url
stringlengths
53
56
repository_url
stringclasses
1 value
labels_url
stringlengths
67
70
comments_url
stringlengths
62
65
events_url
stringlengths
60
63
html_url
stringlengths
41
46
id
int64
450k
1.69B
node_id
stringlengths
18
32
number
int64
1
2.72k
title
stringlengths
1
209
user
dict
labels
list
state
stringclasses
1 value
locked
bool
2 classes
assignee
null
assignees
sequence
milestone
null
comments
sequence
created_at
unknown
updated_at
unknown
closed_at
unknown
author_association
stringclasses
3 values
active_lock_reason
stringclasses
2 values
body
stringlengths
0
104k
reactions
dict
timeline_url
stringlengths
62
65
performed_via_github_app
null
state_reason
stringclasses
2 values
draft
bool
2 classes
pull_request
dict
https://api.github.com/repos/coleifer/peewee/issues/2519
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2519/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2519/comments
https://api.github.com/repos/coleifer/peewee/issues/2519/events
https://github.com/coleifer/peewee/issues/2519
1,113,659,655
I_kwDOAA7yGM5CYRkH
2,519
drop_tables does not delete "schema" tables in MySQL
{ "login": "ascaron37", "id": 8329544, "node_id": "MDQ6VXNlcjgzMjk1NDQ=", "avatar_url": "https://avatars.githubusercontent.com/u/8329544?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ascaron37", "html_url": "https://github.com/ascaron37", "followers_url": "https://api.github.com/users/ascaron37/followers", "following_url": "https://api.github.com/users/ascaron37/following{/other_user}", "gists_url": "https://api.github.com/users/ascaron37/gists{/gist_id}", "starred_url": "https://api.github.com/users/ascaron37/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ascaron37/subscriptions", "organizations_url": "https://api.github.com/users/ascaron37/orgs", "repos_url": "https://api.github.com/users/ascaron37/repos", "events_url": "https://api.github.com/users/ascaron37/events{/privacy}", "received_events_url": "https://api.github.com/users/ascaron37/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "So the reason the drop doesn't work is because Peewee, when \"safe=True\" (the default), attempts to check if the table exists in the current db. To get this working, you can just specify `safe=False`:\r\n\r\n```python\r\n# This will skip the check that the table exists, and then issue the\r\n# appropriate drop table.\r\ndb1.drop_tables([Model2], safe=False)\r\n```", "Perfect, thank you!" ]
"2022-01-25T09:48:21"
"2022-01-25T18:14:51"
"2022-01-25T15:40:33"
NONE
null
I have a problem testing my application with MySQL. I have two separate databases on one MySQL server, say DB1 and DB2. I create a DB Object with `db1 = MySQLDatabase('DB1', ...)`. All my models from DB2 have a schema attribute in the inner Meta class: ``` class DBModel(Model): class Meta: database = db1 class Model2(DBModel): id = AutoField() class Meta: schema = 'DB2' ``` In the application this works well, but for testing purposes I want to create and drop all tables. All models from the first DB get created and deleted properly. My problem now is that creating the tables in the second DB works as expected, but drop_tables does not drop the tables from the second DB. ``` db1.create_tables((Model2,)) # creates the table in the second DB db1.drop_tables((Model2,)) # does not drop tables from second DB ``` Is my usage of the schema attribute a lucky incident from the standpoint that it works in my application? Can I force drop_tables to delete also the tables from the other schemas?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2519/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2519/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2518
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2518/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2518/comments
https://api.github.com/repos/coleifer/peewee/issues/2518/events
https://github.com/coleifer/peewee/issues/2518
1,111,184,135
I_kwDOAA7yGM5CO1MH
2,518
PooledMySQLDatabase MaxConnectionsExceeded
{ "login": "Ahern542968", "id": 32890883, "node_id": "MDQ6VXNlcjMyODkwODgz", "avatar_url": "https://avatars.githubusercontent.com/u/32890883?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Ahern542968", "html_url": "https://github.com/Ahern542968", "followers_url": "https://api.github.com/users/Ahern542968/followers", "following_url": "https://api.github.com/users/Ahern542968/following{/other_user}", "gists_url": "https://api.github.com/users/Ahern542968/gists{/gist_id}", "starred_url": "https://api.github.com/users/Ahern542968/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Ahern542968/subscriptions", "organizations_url": "https://api.github.com/users/Ahern542968/orgs", "repos_url": "https://api.github.com/users/Ahern542968/repos", "events_url": "https://api.github.com/users/Ahern542968/events{/privacy}", "received_events_url": "https://api.github.com/users/Ahern542968/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "http://docs.peewee-orm.com/en/latest/peewee/playhouse.html#pool\r\n\r\nYou are not managing your connections.\r\n\r\nCall connect at the beginning of a request/thread, and close() at the end. This ensures that your conns are returned to the pool." ]
"2022-01-22T02:08:23"
"2022-01-22T13:10:42"
"2022-01-22T13:10:42"
NONE
null
I use PooledMySQLDatabase for a pool , when I with atomic for create , but case MaxConnectionsExceeded ``` def logic_create_event_item(operate_dict): user_id = operate_dict["user_id"] content_type = "event" event_id = operate_dict["event_id"] operate_type = "create" user_operate_log = UserOperateLog.get_or_none( UserOperateLog.user_id == user_id, UserOperateLog.content_type == content_type, UserOperateLog.event_id == event_id, UserOperateLog.operate_type == operate_type) if user_operate_log: return with settings.DB.atomic() as txn: try: user_operate_log = UserOperateLog() user_operate_log.user_id = user_id user_operate_log.content_type = content_type user_operate_log.content_id = event_id user_operate_log.operate_type = operate_type user_operate_log.content = "创建埋点" user_operate_log.save(force_insert=True) except Exception as e: txn.rollback() raise e ``` ![image](https://user-images.githubusercontent.com/32890883/150620761-475a7d32-ecc7-4fdd-8843-a3f1718617f7.png)
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2518/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2518/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2517
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2517/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2517/comments
https://api.github.com/repos/coleifer/peewee/issues/2517/events
https://github.com/coleifer/peewee/issues/2517
1,110,727,085
I_kwDOAA7yGM5CNFmt
2,517
Select with complex where statement don't transform correctly
{ "login": "spacialhufman", "id": 33603355, "node_id": "MDQ6VXNlcjMzNjAzMzU1", "avatar_url": "https://avatars.githubusercontent.com/u/33603355?v=4", "gravatar_id": "", "url": "https://api.github.com/users/spacialhufman", "html_url": "https://github.com/spacialhufman", "followers_url": "https://api.github.com/users/spacialhufman/followers", "following_url": "https://api.github.com/users/spacialhufman/following{/other_user}", "gists_url": "https://api.github.com/users/spacialhufman/gists{/gist_id}", "starred_url": "https://api.github.com/users/spacialhufman/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/spacialhufman/subscriptions", "organizations_url": "https://api.github.com/users/spacialhufman/orgs", "repos_url": "https://api.github.com/users/spacialhufman/repos", "events_url": "https://api.github.com/users/spacialhufman/events{/privacy}", "received_events_url": "https://api.github.com/users/spacialhufman/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Peewee will generate that `0 = 1` when you write an expression that evalutes to `WHERE x IN ()` (which is always false):\r\n\r\n```python\r\n.where(Model.field.in_([]))\r\n```\r\n\r\nSo in your example, at least one of your lists is empty, or alternatively your `tickets` iterator is a generator or something and returns a list, but when iterated the 2nd time it is consumed and returns an empty list." ]
"2022-01-21T17:36:05"
"2022-01-21T17:44:28"
"2022-01-21T17:44:28"
NONE
null
I was building a complex query in my app which is describe below: `query = (self.repository.getSelect(( fieldId, fieldReceivedAt, fieldMac, fieldCode, fieldDomain, fieldStatus )) .where( fieldId.between(initialId, endId + 10000) & SystemEventModel.Message.contains('done query') & self.__getCodeFuncField().in_([t.code for t in tickets]) & SystemEventModel.FromHost.in_([t.mac for t in tickets]) ) .order_by(fieldCode))` The sintax seems correct by when I run the application, the ORM makes my query look like below: `SELECT t1.ID AS id, t1.ReceivedAt AS received_at, t1.FromHost AS mac, TRIM(SUBSTR(SUBSTR(t1.Message, INSTR(t1.Message, '#'), 11), 1, INSTR(SUBSTR(t1.Message, INSTR(t1.Message, '#'), 11), ' '))) AS code, TRIM(IF((INSTR(t1.Message, 'dns name exists') > 0), 'dns name exists, but no appropriate record', SUBSTR(TRIM(SUBSTR(SUBSTR(t1.Message, INSTR(t1.Message, '#'), 100), INSTR(SUBSTR(t1.Message, INSTR(t1.Message, '#'), 100), ' '), 100)), 1, INSTR(TRIM(SUBSTR(SUBSTR(t1.Message, INSTR(t1.Message, '#'), 100), INSTR(SUBSTR(t1.Message, INSTR(t1.Message, '#'), 100), ' '), 100)), ' ')))) AS domain, TRIM(IF((INSTR(t1.Message, 'dns name exists') > 0), 1, IF((INSTR(t1.Message, 'dns does not exist') > 0), 0, 2))) AS status FROM SystemEvents AS t1 WHERE ((((id BETWEEN 23706618 AND 23726554) AND (t1.Message LIKE '%done query%')) AND (TRIM(SUBSTR(SUBSTR(t1.Message, INSTR(t1.Message, '#'), 11), 1, INSTR(SUBSTR(t1.Message, INSTR(t1.Message, '#'), 11), ' '))) IN ('#118729', '#118730', '#12485651', '#12485652', '#12485653', '#129837753', '#129837754', '#1301639'))) AND (0 = 1)) ORDER BY code` Why the last where was transformed into `(0 = 1)` if it is a **where in** statement? Any help would be appreciated. Thanks
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2517/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2517/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2516
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2516/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2516/comments
https://api.github.com/repos/coleifer/peewee/issues/2516/events
https://github.com/coleifer/peewee/issues/2516
1,107,573,649
I_kwDOAA7yGM5CBDuR
2,516
generate_models from DatabaseProxy ValueError
{ "login": "LLjiahai", "id": 30220658, "node_id": "MDQ6VXNlcjMwMjIwNjU4", "avatar_url": "https://avatars.githubusercontent.com/u/30220658?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LLjiahai", "html_url": "https://github.com/LLjiahai", "followers_url": "https://api.github.com/users/LLjiahai/followers", "following_url": "https://api.github.com/users/LLjiahai/following{/other_user}", "gists_url": "https://api.github.com/users/LLjiahai/gists{/gist_id}", "starred_url": "https://api.github.com/users/LLjiahai/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LLjiahai/subscriptions", "organizations_url": "https://api.github.com/users/LLjiahai/orgs", "repos_url": "https://api.github.com/users/LLjiahai/repos", "events_url": "https://api.github.com/users/LLjiahai/events{/privacy}", "received_events_url": "https://api.github.com/users/LLjiahai/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "In this case, just instantiate the introspector directly - or pass it the underlying db instance:\r\n\r\n```python\r\n# Option 1:\r\nintrospector = Introspector(MySQLMetadata(database))\r\nintrospector.generate_models()\r\n\r\n# Option 2:\r\nmodels = generate_models(db.obj)\r\n```", "I gave this another look and made a better fix. Going forward introspection will work with a proxy just fine." ]
"2022-01-19T02:42:34"
"2022-01-19T23:55:00"
"2022-01-19T16:22:41"
NONE
null
db = DatabaseProxy() db.initialize(RetryMySQLDatabase.get_db_instance(**kw_config)) models = generate_models(db) ValueError: Introspection not supported for <peewee.DatabaseProxy object at 0x000002859D87FE40>
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2516/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2516/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2515
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2515/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2515/comments
https://api.github.com/repos/coleifer/peewee/issues/2515/events
https://github.com/coleifer/peewee/issues/2515
1,106,872,487
I_kwDOAA7yGM5B-Yin
2,515
How to get all data from my Model?
{ "login": "peepo5", "id": 72892531, "node_id": "MDQ6VXNlcjcyODkyNTMx", "avatar_url": "https://avatars.githubusercontent.com/u/72892531?v=4", "gravatar_id": "", "url": "https://api.github.com/users/peepo5", "html_url": "https://github.com/peepo5", "followers_url": "https://api.github.com/users/peepo5/followers", "following_url": "https://api.github.com/users/peepo5/following{/other_user}", "gists_url": "https://api.github.com/users/peepo5/gists{/gist_id}", "starred_url": "https://api.github.com/users/peepo5/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/peepo5/subscriptions", "organizations_url": "https://api.github.com/users/peepo5/orgs", "repos_url": "https://api.github.com/users/peepo5/repos", "events_url": "https://api.github.com/users/peepo5/events{/privacy}", "received_events_url": "https://api.github.com/users/peepo5/received_events", "type": "User", "site_admin": false }
[]
closed
true
null
[]
null
[ "You need to read the documentation.\r\n\r\nTake 10 or 15 minutes and go through the quick start.\r\n\r\nhttp://docs.peewee-orm.com/en/latest/peewee/quickstart.html#retrieving-data", "So I have to iterate over the select to get any results? Your answer to my question did not help.", "http://docs.peewee-orm.com/en/latest/peewee/querying.html#selecting-a-single-record\r\n\r\nhttp://docs.peewee-orm.com/en/latest/peewee/querying.html#selecting-multiple-records", "To answer my own question, yes you must iterate. @coleifer are there any ways of not iterating to save time? like recieving an array of classes instead of just something to iterate upon. Thanks. And please give some explanation rather than just sending a link, thank you.", "https://docs.python.org/3/library/functions.html#func-list" ]
"2022-01-18T12:50:29"
"2022-01-18T20:08:55"
"2022-01-18T12:57:12"
NONE
off-topic
My class looks like this: ```py class WaitPayment(BaseModel): txn_id = TextField(unique=True, primary_key=True) discord_id = TextField() ms_expires = DecimalField() ``` and I want to recieve all of its records as a dict. However, when I use `WaitPayment.select()` or `WaitPayment.select().dicts()`, it returns: ```sql SELECT "t1"."txn_id", "t1"."discord_id", "t1"."ms_expires" FROM "waitpayment" AS "t1" ``` in the terminal, which is not what I want. I do not want the sql query, I want the result. Any help? Thanks.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2515/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2515/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2514
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2514/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2514/comments
https://api.github.com/repos/coleifer/peewee/issues/2514/events
https://github.com/coleifer/peewee/issues/2514
1,105,342,191
I_kwDOAA7yGM5B4i7v
2,514
Field argument 'help_text' seem have not realized in MySQL...
{ "login": "xfl12345", "id": 17960863, "node_id": "MDQ6VXNlcjE3OTYwODYz", "avatar_url": "https://avatars.githubusercontent.com/u/17960863?v=4", "gravatar_id": "", "url": "https://api.github.com/users/xfl12345", "html_url": "https://github.com/xfl12345", "followers_url": "https://api.github.com/users/xfl12345/followers", "following_url": "https://api.github.com/users/xfl12345/following{/other_user}", "gists_url": "https://api.github.com/users/xfl12345/gists{/gist_id}", "starred_url": "https://api.github.com/users/xfl12345/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/xfl12345/subscriptions", "organizations_url": "https://api.github.com/users/xfl12345/orgs", "repos_url": "https://api.github.com/users/xfl12345/repos", "events_url": "https://api.github.com/users/xfl12345/events{/privacy}", "received_events_url": "https://api.github.com/users/xfl12345/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "The `help_text` and `verbose_name` are used for integrations with flask-peewee (or other user-facing applications) and do not have any impact on the schema." ]
"2022-01-17T04:03:44"
"2022-01-17T14:23:27"
"2022-01-17T14:23:27"
NONE
null
I think the 'help_text' parameter may become the 'comment' value in MySQL. It did not work as i expect. Neigther of 'verbose_name' parameter. I would appreciate if the little feature has been realized. Thanks a lot. (BTW: peewee is so much easier to use than SQLAlchemy!!!Can't help coding with peewee.)
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2514/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2514/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2513
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2513/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2513/comments
https://api.github.com/repos/coleifer/peewee/issues/2513/events
https://github.com/coleifer/peewee/issues/2513
1,101,868,467
I_kwDOAA7yGM5BrS2z
2,513
Unreliable hybrid_fields order in model_to_dict(model, extra_attrs=hybrid_fields) bug?
{ "login": "neutralvibes", "id": 26578830, "node_id": "MDQ6VXNlcjI2NTc4ODMw", "avatar_url": "https://avatars.githubusercontent.com/u/26578830?v=4", "gravatar_id": "", "url": "https://api.github.com/users/neutralvibes", "html_url": "https://github.com/neutralvibes", "followers_url": "https://api.github.com/users/neutralvibes/followers", "following_url": "https://api.github.com/users/neutralvibes/following{/other_user}", "gists_url": "https://api.github.com/users/neutralvibes/gists{/gist_id}", "starred_url": "https://api.github.com/users/neutralvibes/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/neutralvibes/subscriptions", "organizations_url": "https://api.github.com/users/neutralvibes/orgs", "repos_url": "https://api.github.com/users/neutralvibes/repos", "events_url": "https://api.github.com/users/neutralvibes/events{/privacy}", "received_events_url": "https://api.github.com/users/neutralvibes/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "~~Despite spending a few hours on it, a slap from myself helped me realise that perhaps OrderedDict isn't being used, despite database fields seemingly reproduced in reliable in order.~~ Nope, dict keys became reliably ordered from python 3.7.", "Dict keys are not sorted -- exactly. This is not an issue in my opinion, this is just the nature of dictionaries. A dictionary is an inherently unordered data-structure, and while later Python preserves the insert order, it's not something to rely upon.\r\n\r\nYou already have your list of attrs - so you probably want to do something like: `table.append([getattr(model, attr) for attr in list_of_attrs])`. Or just pull the keys out of the resulting dict in the order you need them.", "First, many thanks for reviewing.\r\n\r\nDicts have guaranteed order as of python 3.7 so can be relied upon. \r\n I have looked at the underlyng code and the reason is you convert the input to a `Set`. The reason for the lack of stability is because of a shortcut you use to convert the input.\r\n\r\n~~~python\r\n_clone_set = lambda s: set(s) if s else set()\r\n~~~\r\n\r\nSets still do not have this assurance, and I wonder if using a List would be better saving on converting many rows of data for the sake of it. I bow to your superior knowledge so shall leave it there.\r\n\r\nThanks again.", "We don't bow to anyone in Peewee!\r\n\r\nI think my recommendation to you is the same:\r\n\r\n* either do not use `model_to_dict()` and just grab the attrs you want explicitly\r\n* or, once you have the dict back, pull the fields out of it explicitly using a list comprehension or the like" ]
"2022-01-13T14:38:27"
"2022-01-13T16:26:25"
"2022-01-13T16:11:07"
NONE
null
# Hybrid fields output order from model_to_dict - bug? ## Problem - Python Version: 3.7.6 I have a case where a gui table requires row data in tuple format. This must be ordered correctly in relation to the provided headers. The data is supplied by a model with a couple of hybrid fields. I am finding that irrespective of the order provided to the `extra_attrs` parameter of `model_to_dict`, while the order of the data directly from the database is predictable, the order of the resultant hybrid fields attached is not in my case. ## Example Here is an example missing definitions but, I hope gives enough to demonstate with a fictious Person table where `status` and `age` are hybrids getting their data from other Person fields. ~~~python # Fields to list, * denotes hybrid field display_list = [ 'id', 'name', '*status', '*age' ] # Get hybrids list hybrid_fields = [ f for f in display_list if f.startswith('*') ] # Printing shows supplied order consistently print(f'hybrids: {hybrid_fields}') # Output: hybrids: ['status', 'age'] # Grab list of non hybrid fields to extract only_prep = [ f for f in display_list if not f.startswith('*') ] # Get attributes for `only_list` from table class only_list = ( attrgetter(*only_prep)(Person) ) if len(only_prep) else () print(only_list) # Output: (<AutoField: Person.id>, <CharField: Person.name>) table_data = [] # to hold return from encapsulating function for rec in Person.select(): row = model_to_dict( rec, only=only_list, extra_attrs=hybrid_fields ) print( row.keys() ) # Expected: dict_keys(['id', 'name', 'status','age']) ....and so on # Sometimes on python restart: dict_keys(['id', 'name', 'age', 'status']) .... notice the change in the last two. ​ table_data.append( tuple(row.values()) ) ~~~ ## Tales of the unexpected What is bizzare is that while python is running the order returned seems fixed. Run the program from scratch a few times after a period then it can alternate. ## Advice appreciated Not sure if I'm doing something wrong. Tried making `hybrid_fields ` a tuple with no joy. Any suggestions would be appreciated as I am not sure if this is a bug or an interface with the keyboard problem.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2513/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2513/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2512
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2512/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2512/comments
https://api.github.com/repos/coleifer/peewee/issues/2512/events
https://github.com/coleifer/peewee/issues/2512
1,097,482,999
I_kwDOAA7yGM5BakL3
2,512
potential raw_unicode_escape error when debugging customized BINARY field
{ "login": "est", "id": 23570, "node_id": "MDQ6VXNlcjIzNTcw", "avatar_url": "https://avatars.githubusercontent.com/u/23570?v=4", "gravatar_id": "", "url": "https://api.github.com/users/est", "html_url": "https://github.com/est", "followers_url": "https://api.github.com/users/est/followers", "following_url": "https://api.github.com/users/est/following{/other_user}", "gists_url": "https://api.github.com/users/est/gists{/gist_id}", "starred_url": "https://api.github.com/users/est/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/est/subscriptions", "organizations_url": "https://api.github.com/users/est/orgs", "repos_url": "https://api.github.com/users/est/repos", "events_url": "https://api.github.com/users/est/events{/privacy}", "received_events_url": "https://api.github.com/users/est/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Did you not read the [code comment](https://github.com/coleifer/peewee/blob/master/peewee.py#L658-L661)? This is not part of the public API.", "> Did you not read the code comment? This is not part of the public API.\r\n\r\nHi coleifer, I am using `logger.debug(\"sql is %s\", query)` with one of the parameter as a customized mysql BINARY field. I think it fits the code comment of \"This function is intended for debugging or logging purposes ONLY\".\r\n\r\nUnless there's another way to debug the SQL I am unaware of?\r\n", "The correct way to do this is:\r\n\r\n```python\r\n\r\nquery = MyModel.select().where(MyModel.foo == 'bar')\r\nsql, params = query.sql()\r\n```\r\n\r\nYou can inspect the sql string, which contains proper placeholders for the parameters, and the parameters themselves." ]
"2022-01-10T05:49:22"
"2022-01-10T16:27:42"
"2022-01-10T14:01:50"
NONE
null
Hi, I created a BINARY field following the doc https://docs.peewee-orm.com/en/latest/peewee/models.html#creating-a-custom-field When debugging the SQL, peewee will try `utf8` or `raw_unicode_escape` the `bytes()` parameter, at one time my bytes happens to be `b"\x0ee\xd4\xee\xff\xc0\x02r\xc2\\U0'\x00"`, which will error in [_query_val_transform](https://github.com/coleifer/peewee/blob/master/peewee.py#L683) ``` UnicodeDecodeError 'rawunicodeescape' codec can't decode bytes in position 9-11: truncated \UXXXXXXXX escape ``` I think it's because we assume the in SQL parameter, `bytes()` are displayable strings but unfortunately it will fail when there's an illegal "\U" literal inside the string.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2512/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2512/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2511
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2511/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2511/comments
https://api.github.com/repos/coleifer/peewee/issues/2511/events
https://github.com/coleifer/peewee/issues/2511
1,095,551,899
I_kwDOAA7yGM5BTMub
2,511
Joined and grouped foreign key values are not cached
{ "login": "conqp", "id": 3766192, "node_id": "MDQ6VXNlcjM3NjYxOTI=", "avatar_url": "https://avatars.githubusercontent.com/u/3766192?v=4", "gravatar_id": "", "url": "https://api.github.com/users/conqp", "html_url": "https://github.com/conqp", "followers_url": "https://api.github.com/users/conqp/followers", "following_url": "https://api.github.com/users/conqp/following{/other_user}", "gists_url": "https://api.github.com/users/conqp/gists{/gist_id}", "starred_url": "https://api.github.com/users/conqp/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/conqp/subscriptions", "organizations_url": "https://api.github.com/users/conqp/orgs", "repos_url": "https://api.github.com/users/conqp/repos", "events_url": "https://api.github.com/users/conqp/events{/privacy}", "received_events_url": "https://api.github.com/users/conqp/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Please consult the docs here: http://docs.peewee-orm.com/en/latest/peewee/relationships.html#avoiding-the-n-1-problem\r\n\r\nFurthermore, you cannot go from \"many\" -> \"many\" in a single go without using prefetch. You can join \"upwards\" no problem, e.g., select all Systems and also grab their Deployment:\r\n\r\n```python\r\n# This executes in a single query.\r\nquery = System.select(System, Deployment).join(Deployment)\r\nfor system in query:\r\n print(system.id, system.deployment.annotation)\r\n```\r\n\r\nWe cannot do the reverse automatically, e.g., get deployments and all their systems, because of SQL's tabular nature.\r\n\r\nSo you can either:\r\n\r\n1. Use `prefetch()` (although I strongly advise profiling this beforehand to see if it actually provides any measurable improvements)\r\n2. Use N+1, e.g., for each deployment also query its systems.\r\n\r\nAlso note that in your first query you are not actually aggregating anything, and the query probably doesn't do what you think it does.\r\n\r\nThe docs provide many examples: http://docs.peewee-orm.com/en/latest/peewee/relationships.html#avoiding-the-n-1-problem\r\n\r\nLastly, queries generated by accessing a backref are not cached -- accessing the backref creates a new query. If you want to iterate over a backref multiple times you can just assign it to a local variable:\r\n\r\n```python\r\n\r\nuser = User.get(User.username == 'charlie')\r\ntweets = user.tweets\r\n# Query is only evaluated once:\r\nfor t in tweets: pass\r\nfor t in tweets: pass\r\n```" ]
"2022-01-06T17:54:30"
"2022-01-06T19:10:06"
"2022-01-06T19:06:58"
CONTRIBUTOR
null
I have a case where I want to select a one-to-many relation. According to the [docs](http://docs.peewee-orm.com/en/latest/peewee/querying.html#aggregating-records), this can be done with appropriate selects, joins and group-bys. However, peewee does not cache query, but executes it again, if I access the backref: ```python from logging import DEBUG, basicConfig basicConfig(level=DEBUG) from hwdb import System, Deployment select = Deployment.select(System, Deployment).join(System, on=System.deployment == Deployment.id).where(Deployment.id == 354).group_by(Deployment) dep = select.get() print(dep.systems) print(list(dep.systems)) ``` Result: ```python DEBUG:peewee:('SELECT `t1`.`id`, `t1`.`Group`, `t1`.`deployment`, `t1`.`dataset`, `t1`.`openvpn`, `t1`.`ipv6address`, `t1`.`pubkey`, `t1`.`created`, `t1`.`configured`, `t1`.`fitted`, `t1`.`operating_system`, `t1`.`monitor`, `t1`.`serial_number`, `t1`.`model`, `t1`.`last_sync`, `t2`.`id`, `t2`.`customer`, `t2`.`type`, `t2`.`connection`, `t2`.`address`, `t2`.`lpt_address`, `t2`.`scheduled`, `t2`.`annotation`, `t2`.`testing`, `t2`.`timestamp` FROM `hwdb`.`deployment` AS `t2` INNER JOIN `hwdb`.`system` AS `t1` ON (`t1`.`deployment` = `t2`.`id`) WHERE (`t2`.`id` = %s) GROUP BY `t2`.`id`, `t2`.`customer`, `t2`.`type`, `t2`.`connection`, `t2`.`address`, `t2`.`lpt_address`, `t2`.`scheduled`, `t2`.`annotation`, `t2`.`testing`, `t2`.`timestamp` LIMIT %s OFFSET %s', [354, 1, 0]) SELECT `t1`.`id`, `t1`.`Group`, `t1`.`deployment`, `t1`.`dataset`, `t1`.`openvpn`, `t1`.`ipv6address`, `t1`.`pubkey`, `t1`.`created`, `t1`.`configured`, `t1`.`fitted`, `t1`.`operating_system`, `t1`.`monitor`, `t1`.`serial_number`, `t1`.`model`, `t1`.`last_sync` FROM `hwdb`.`system` AS `t1` WHERE (`t1`.`deployment` = 354) DEBUG:peewee:('SELECT `t1`.`id`, `t1`.`Group`, `t1`.`deployment`, `t1`.`dataset`, `t1`.`openvpn`, `t1`.`ipv6address`, `t1`.`pubkey`, `t1`.`created`, `t1`.`configured`, `t1`.`fitted`, `t1`.`operating_system`, `t1`.`monitor`, `t1`.`serial_number`, `t1`.`model`, `t1`.`last_sync` FROM `hwdb`.`system` AS `t1` WHERE (`t1`.`deployment` = %s)', [354]) [<System: 36>, <System: 359>] ``` As you can see from the last debug output, peewee executes another query to retrieve the related records. Shouldn't it have cached them from the initial join? Vanilla peewee test script: ```python from configparser import ConfigParser from datetime import datetime from logging import DEBUG, basicConfig from peewee import Model, MySQLDatabase from peewee import BooleanField, CharField, DateField, DateTimeField, ForeignKeyField, FixedCharField basicConfig(level=DEBUG) CONFIG = ConfigParser() CONFIG.read('/usr/local/etc/hwdb.conf') DATABASE = MySQLDatabase( 'hwdb', host='localhost', user='hwdb', passwd=CONFIG.get('db', 'passwd') ) class Deployment(Model): """A customer-specific deployment of a terminal.""" class Meta: database = DATABASE schema = database.database scheduled = DateField(null=True) annotation = CharField(255, null=True) testing = BooleanField(default=False) timestamp = DateTimeField(null=True) class System(Model): """A physical computer system out in the field.""" class Meta: database = DATABASE schema = database.database deployment = ForeignKeyField( Deployment, null=True, column_name='deployment', backref='systems', on_delete='SET NULL', on_update='CASCADE', lazy_load=False) pubkey = FixedCharField(44, null=True, unique=True) created = DateTimeField(default=datetime.now) configured = DateTimeField(null=True) fitted = BooleanField(default=False) monitor = BooleanField(null=True) serial_number = CharField(255, null=True) model = CharField(255, null=True) # Hardware model. last_sync = DateTimeField(null=True) select = Deployment.select(System, Deployment).join(System, on=System.deployment == Deployment.id).where(Deployment.id == 354).group_by(Deployment) dep = select.get() print(dep.systems) print(list(dep.systems)) ``` Result: ```python DEBUG:peewee:('SELECT `t1`.`id`, `t1`.`deployment`, `t1`.`pubkey`, `t1`.`created`, `t1`.`configured`, `t1`.`fitted`, `t1`.`monitor`, `t1`.`serial_number`, `t1`.`model`, `t1`.`last_sync`, `t2`.`id`, `t2`.`scheduled`, `t2`.`annotation`, `t2`.`testing`, `t2`.`timestamp` FROM `hwdb`.`deployment` AS `t2` INNER JOIN `hwdb`.`system` AS `t1` ON (`t1`.`deployment` = `t2`.`id`) WHERE (`t2`.`id` = %s) GROUP BY `t2`.`id`, `t2`.`scheduled`, `t2`.`annotation`, `t2`.`testing`, `t2`.`timestamp` LIMIT %s OFFSET %s', [354, 1, 0]) SELECT `t1`.`id`, `t1`.`deployment`, `t1`.`pubkey`, `t1`.`created`, `t1`.`configured`, `t1`.`fitted`, `t1`.`monitor`, `t1`.`serial_number`, `t1`.`model`, `t1`.`last_sync` FROM `hwdb`.`system` AS `t1` WHERE (`t1`.`deployment` = 354) DEBUG:peewee:('SELECT `t1`.`id`, `t1`.`deployment`, `t1`.`pubkey`, `t1`.`created`, `t1`.`configured`, `t1`.`fitted`, `t1`.`monitor`, `t1`.`serial_number`, `t1`.`model`, `t1`.`last_sync` FROM `hwdb`.`system` AS `t1` WHERE (`t1`.`deployment` = %s)', [354]) [<System: 36>, <System: 359>] ``` If I remove the select and join, it still works: ```python select = Deployment.select().where(Deployment.id == 354).group_by(Deployment) ``` It seems like `lazy_load=False` is being entirely ignored.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2511/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2511/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2510
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2510/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2510/comments
https://api.github.com/repos/coleifer/peewee/issues/2510/events
https://github.com/coleifer/peewee/issues/2510
1,094,163,850
I_kwDOAA7yGM5BN52K
2,510
table_exists(MyModel) should raise error
{ "login": "jonashaag", "id": 175722, "node_id": "MDQ6VXNlcjE3NTcyMg==", "avatar_url": "https://avatars.githubusercontent.com/u/175722?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jonashaag", "html_url": "https://github.com/jonashaag", "followers_url": "https://api.github.com/users/jonashaag/followers", "following_url": "https://api.github.com/users/jonashaag/following{/other_user}", "gists_url": "https://api.github.com/users/jonashaag/gists{/gist_id}", "starred_url": "https://api.github.com/users/jonashaag/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/jonashaag/subscriptions", "organizations_url": "https://api.github.com/users/jonashaag/orgs", "repos_url": "https://api.github.com/users/jonashaag/repos", "events_url": "https://api.github.com/users/jonashaag/events{/privacy}", "received_events_url": "https://api.github.com/users/jonashaag/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Thanks, this method now accepts a model class." ]
"2022-01-05T09:45:29"
"2022-01-05T13:41:17"
"2022-01-05T13:41:06"
CONTRIBUTOR
null
Misuse of `table_exists` API always returns `False`: ```py db.table_exists(MyModel) ``` IMO this should either raise an error, or be identical to ```py db.table_exists(MyModel.table_name) ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2510/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2510/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2509
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2509/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2509/comments
https://api.github.com/repos/coleifer/peewee/issues/2509/events
https://github.com/coleifer/peewee/issues/2509
1,092,086,129
I_kwDOAA7yGM5BF-lx
2,509
Add ModelSelect to __all__ for linting
{ "login": "1kastner", "id": 5236165, "node_id": "MDQ6VXNlcjUyMzYxNjU=", "avatar_url": "https://avatars.githubusercontent.com/u/5236165?v=4", "gravatar_id": "", "url": "https://api.github.com/users/1kastner", "html_url": "https://github.com/1kastner", "followers_url": "https://api.github.com/users/1kastner/followers", "following_url": "https://api.github.com/users/1kastner/following{/other_user}", "gists_url": "https://api.github.com/users/1kastner/gists{/gist_id}", "starred_url": "https://api.github.com/users/1kastner/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/1kastner/subscriptions", "organizations_url": "https://api.github.com/users/1kastner/orgs", "repos_url": "https://api.github.com/users/1kastner/repos", "events_url": "https://api.github.com/users/1kastner/events{/privacy}", "received_events_url": "https://api.github.com/users/1kastner/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I'm going to pass - I don't really see a need to export this." ]
"2022-01-02T19:35:36"
"2022-01-04T14:28:11"
"2022-01-04T14:28:11"
NONE
null
First of all thank you very much for creating this project and sharing it under the MIT license! Recently I have started to rely more on type-hinting in my projects to have better tool support and in some cases making my intentions more obvious for the readers of the code. While doing that, I got the warning that ModelSelect is not part of the official interface as it is not listed in `__all__`. However, I would like to pass a query which is not yet executed to another method or function and only there execute it. Thus my request to add `ModelSelect` to `__all__` and treat it as an official part of the API. Of course there could be design choices or other ideas that contradict this. In that case I would love to learn more about the future intentions of how this project will continue.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2509/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2509/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2508
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2508/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2508/comments
https://api.github.com/repos/coleifer/peewee/issues/2508/events
https://github.com/coleifer/peewee/issues/2508
1,091,487,002
I_kwDOAA7yGM5BDsUa
2,508
how to search by pair of columns
{ "login": "sherrrrr", "id": 31510228, "node_id": "MDQ6VXNlcjMxNTEwMjI4", "avatar_url": "https://avatars.githubusercontent.com/u/31510228?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sherrrrr", "html_url": "https://github.com/sherrrrr", "followers_url": "https://api.github.com/users/sherrrrr/followers", "following_url": "https://api.github.com/users/sherrrrr/following{/other_user}", "gists_url": "https://api.github.com/users/sherrrrr/gists{/gist_id}", "starred_url": "https://api.github.com/users/sherrrrr/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sherrrrr/subscriptions", "organizations_url": "https://api.github.com/users/sherrrrr/orgs", "repos_url": "https://api.github.com/users/sherrrrr/repos", "events_url": "https://api.github.com/users/sherrrrr/events{/privacy}", "received_events_url": "https://api.github.com/users/sherrrrr/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Here is an example (tested with mariadb 10.6):\r\n\r\n```python\r\nclass Reg(Model):\r\n key = CharField()\r\n value = TextField()\r\n\r\nReg.insert_many([('k%02d' % i, 'v%02d' % i) for i in range(100)]).execute()\r\ncols = Tuple(Reg.key, Reg.value)\r\nvalues = [('k02', 'v02'), ('k04', 'v04'), ('k06', 'v06')]\r\nquery = (Reg.select()\r\n .where(cols.in_(values))\r\n .order_by(Reg.key))\r\nfor reg in query:\r\n print(reg.key, reg.value)\r\n\r\n# Prints:\r\n# k02 v02\r\n# k04 v04\r\n# k06 v06\r\n```" ]
"2021-12-31T09:08:14"
"2021-12-31T14:29:40"
"2021-12-31T14:28:25"
NONE
null
How can I write something like this using peewee in mysql? ``` SELECT whatever FROM t WHERE (col1, col2) IN ((val1a, val2a), (val1b, val2b), ...) ; ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2508/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2508/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2507
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2507/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2507/comments
https://api.github.com/repos/coleifer/peewee/issues/2507/events
https://github.com/coleifer/peewee/issues/2507
1,090,305,263
I_kwDOAA7yGM5A_Lzv
2,507
Does not accept pandas dataframe int/float columns with NA
{ "login": "ashishkumarazuga", "id": 78608338, "node_id": "MDQ6VXNlcjc4NjA4MzM4", "avatar_url": "https://avatars.githubusercontent.com/u/78608338?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ashishkumarazuga", "html_url": "https://github.com/ashishkumarazuga", "followers_url": "https://api.github.com/users/ashishkumarazuga/followers", "following_url": "https://api.github.com/users/ashishkumarazuga/following{/other_user}", "gists_url": "https://api.github.com/users/ashishkumarazuga/gists{/gist_id}", "starred_url": "https://api.github.com/users/ashishkumarazuga/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ashishkumarazuga/subscriptions", "organizations_url": "https://api.github.com/users/ashishkumarazuga/orgs", "repos_url": "https://api.github.com/users/ashishkumarazuga/repos", "events_url": "https://api.github.com/users/ashishkumarazuga/events{/privacy}", "received_events_url": "https://api.github.com/users/ashishkumarazuga/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "This error is being thrown by your database, via your database driver, and is not raised by Peewee." ]
"2021-12-29T06:42:55"
"2021-12-29T13:49:30"
"2021-12-29T13:49:30"
NONE
null
The pandas dataframe containing columns which are int64 or float64 in nature, containing NA or NaN are not accepted by peewee. And throws a `DataError('integer out of range\n')`, during an insert.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2507/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2507/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2506
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2506/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2506/comments
https://api.github.com/repos/coleifer/peewee/issues/2506/events
https://github.com/coleifer/peewee/issues/2506
1,090,269,083
I_kwDOAA7yGM5A_C-b
2,506
distutils has been deprecated in Python 3.10
{ "login": "tirkarthi", "id": 3972343, "node_id": "MDQ6VXNlcjM5NzIzNDM=", "avatar_url": "https://avatars.githubusercontent.com/u/3972343?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tirkarthi", "html_url": "https://github.com/tirkarthi", "followers_url": "https://api.github.com/users/tirkarthi/followers", "following_url": "https://api.github.com/users/tirkarthi/following{/other_user}", "gists_url": "https://api.github.com/users/tirkarthi/gists{/gist_id}", "starred_url": "https://api.github.com/users/tirkarthi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tirkarthi/subscriptions", "organizations_url": "https://api.github.com/users/tirkarthi/orgs", "repos_url": "https://api.github.com/users/tirkarthi/repos", "events_url": "https://api.github.com/users/tirkarthi/events{/privacy}", "received_events_url": "https://api.github.com/users/tirkarthi/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Steve Dower @zooba, a Microsoft employee, is railroading through this idiotic pep. From what I can see there are quite a few people who disagree with his decision:\r\n\r\nhttps://discuss.python.org/t/pep-632-deprecate-distutils-module/5134/29\r\n\r\nHe comes off as a smug, smarmy nitwit who cannot appreciate that real world users will be affected by this decision. Once again, Microsoft employees shitting up the party? Or just another petty tyrant who found a source of narcissistic supply in open source? You decide.", "As much as I sympathize with the reaction, the code should be updated in the future. It seems that the pep isn't going away.", "Fallback to the vendored distutils inside setuptools for now: 7ab5e401d85c1eec7aafbbe2fdf3c0344f82422b", "Anyone who stumbles upon this and can find examples of prior art for projects that build C extensions and are going through a similar migration, I'd appreciate links. I glanced around at a few projects I'm familiar with, and all of them are still on `distutils`." ]
"2021-12-29T04:56:21"
"2021-12-30T01:02:47"
"2021-12-29T20:47:49"
NONE
null
https://www.python.org/dev/peps/pep-0632/#migration-advice https://github.com/coleifer/peewee/blob/5a55933ed28b261fbe100568125fffe0ba854b99/setup.py#L5-L8
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2506/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2506/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2505
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2505/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2505/comments
https://api.github.com/repos/coleifer/peewee/issues/2505/events
https://github.com/coleifer/peewee/issues/2505
1,083,846,770
I_kwDOAA7yGM5AmjBy
2,505
formatargspec was removed in Python 3.11
{ "login": "tirkarthi", "id": 3972343, "node_id": "MDQ6VXNlcjM5NzIzNDM=", "avatar_url": "https://avatars.githubusercontent.com/u/3972343?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tirkarthi", "html_url": "https://github.com/tirkarthi", "followers_url": "https://api.github.com/users/tirkarthi/followers", "following_url": "https://api.github.com/users/tirkarthi/following{/other_user}", "gists_url": "https://api.github.com/users/tirkarthi/gists{/gist_id}", "starred_url": "https://api.github.com/users/tirkarthi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tirkarthi/subscriptions", "organizations_url": "https://api.github.com/users/tirkarthi/orgs", "repos_url": "https://api.github.com/users/tirkarthi/repos", "events_url": "https://api.github.com/users/tirkarthi/events{/privacy}", "received_events_url": "https://api.github.com/users/tirkarthi/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "That doesn't matter. The stdlib-provided mock is used for any python 3 or newer, and that old version is vendored for Python 2.x users.", "Got it, sorry for the noise." ]
"2021-12-18T14:43:19"
"2021-12-19T01:41:07"
"2021-12-19T01:01:30"
NONE
null
https://github.com/coleifer/peewee/blob/5a55933ed28b261fbe100568125fffe0ba854b99/tests/libs/mock.py#L190 https://docs.python.org/3.11/whatsnew/3.11.html#removed > the formatargspec function, deprecated since Python 3.5; use the inspect.signature() function and Signature object directly.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2505/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2505/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2504
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2504/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2504/comments
https://api.github.com/repos/coleifer/peewee/issues/2504/events
https://github.com/coleifer/peewee/issues/2504
1,083,132,854
I_kwDOAA7yGM5Aj0u2
2,504
PostgresqlExtDatabase. Model with JSONField raise exception on any action with search (get, get_or_create, etc.)
{ "login": "olvinroght", "id": 49984942, "node_id": "MDQ6VXNlcjQ5OTg0OTQy", "avatar_url": "https://avatars.githubusercontent.com/u/49984942?v=4", "gravatar_id": "", "url": "https://api.github.com/users/olvinroght", "html_url": "https://github.com/olvinroght", "followers_url": "https://api.github.com/users/olvinroght/followers", "following_url": "https://api.github.com/users/olvinroght/following{/other_user}", "gists_url": "https://api.github.com/users/olvinroght/gists{/gist_id}", "starred_url": "https://api.github.com/users/olvinroght/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/olvinroght/subscriptions", "organizations_url": "https://api.github.com/users/olvinroght/orgs", "repos_url": "https://api.github.com/users/olvinroght/repos", "events_url": "https://api.github.com/users/olvinroght/events{/privacy}", "received_events_url": "https://api.github.com/users/olvinroght/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Refer to the docs: https://www.postgresql.org/docs/current/datatype-json.html\r\n\r\nBut, simply, you cannot compare `json` values. I am not aware of any reason why you would want to use `json` over `jsonb`, so I would suggest that you just use `BinaryJSONField` (which additionally supports indexing).\r\n\r\nIf you really want to use `JSONField` then you should not use `get_or_create()` but instead use `insert(...).on_conflict()` and rely on a unique constraint." ]
"2021-12-17T10:50:01"
"2021-12-17T14:04:18"
"2021-12-17T14:02:59"
NONE
null
```python from playhouse.postgres_ext import PostgresqlExtDatabase, JSONField from peewee import Model, CharField database = PostgresqlExtDatabase("Test", user="user", password="password", host="localhost", port=5432, autorollback=True) class TestJSONModel(Model): name = CharField(unique=True) json_value = JSONField() class Meta: database = database TestJSONModel.create_table() model, created = TestJSONModel.get_or_create( name="test_json_model_1", json_value={"key1": "value1"} ) ``` It raises an exception: ```none Traceback (most recent call last): File "\venv\lib\site-packages\peewee.py", line 3160, in execute_sql cursor.execute(sql, params or ()) psycopg2.errors.UndefinedFunction: ERROR: operator does not exist: json = json LINE 1: ..."name" = 'test_json_model_1'), ("t1"."json_value" = CAST('{"... ^ HINT: No operator matches the given name and argument type(s). You might need to add explicit type casts. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "\test.py", line 24, in <module> model, created = TestJSONModel.select( File "\venv\lib\site-packages\peewee.py", line 1911, in inner return method(self, database, *args, **kwargs) File "\venv\lib\site-packages\peewee.py", line 1982, in execute return self._execute(database) File "\venv\lib\site-packages\peewee.py", line 2155, in _execute cursor = database.execute(self) File "\venv\lib\site-packages\playhouse\postgres_ext.py", line 490, in execute cursor = self.execute_sql(sql, params, commit=commit) File "\venv\lib\site-packages\peewee.py", line 3167, in execute_sql self.commit() File "\venv\lib\site-packages\peewee.py", line 2933, in __exit__ reraise(new_type, new_type(exc_value, *exc_args), traceback) File "\venv\lib\site-packages\peewee.py", line 191, in reraise raise value.with_traceback(tb) File "\venv\lib\site-packages\peewee.py", line 3160, in execute_sql cursor.execute(sql, params or ()) peewee.ProgrammingError: ERROR: operator does not exist: json = json LINE 1: ..."name" = 'test_json_model_1'), ("t1"."json_value" = CAST('{"... ^ HINT: No operator matches the given name and argument type(s). You might need to add explicit type casts. ``` Same happens on any other query where `json_value` comparison used. Code above produces next query: ```sql SELECT ("t1"."name" = %s), ("t1"."json_value" = CAST(%s AS json)) FROM "testjsonmodel" AS "t1" ``` As far as I understood, it's not possible to compare JSON fields, only their string values, so it will be good if peewee handles it internally instead of raising an exception. This happens only with `JSONField`, `BinaryJSONField` works well. There's [a question](https://stackoverflow.com/q/32843213/10824407) on Stack Overflow with same issue.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2504/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2504/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2503
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2503/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2503/comments
https://api.github.com/repos/coleifer/peewee/issues/2503/events
https://github.com/coleifer/peewee/issues/2503
1,079,774,294
I_kwDOAA7yGM5AXAxW
2,503
how to convert peewee column field_type to db column_type
{ "login": "LLjiahai", "id": 30220658, "node_id": "MDQ6VXNlcjMwMjIwNjU4", "avatar_url": "https://avatars.githubusercontent.com/u/30220658?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LLjiahai", "html_url": "https://github.com/LLjiahai", "followers_url": "https://api.github.com/users/LLjiahai/followers", "following_url": "https://api.github.com/users/LLjiahai/following{/other_user}", "gists_url": "https://api.github.com/users/LLjiahai/gists{/gist_id}", "starred_url": "https://api.github.com/users/LLjiahai/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LLjiahai/subscriptions", "organizations_url": "https://api.github.com/users/LLjiahai/orgs", "repos_url": "https://api.github.com/users/LLjiahai/repos", "events_url": "https://api.github.com/users/LLjiahai/events{/privacy}", "received_events_url": "https://api.github.com/users/LLjiahai/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[]
"2021-12-14T13:41:15"
"2021-12-14T14:34:46"
"2021-12-14T14:34:46"
NONE
null
ps: field_a = peewee .CharField(max_length=255) to mysql column_type varchar(255)
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2503/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2503/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2502
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2502/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2502/comments
https://api.github.com/repos/coleifer/peewee/issues/2502/events
https://github.com/coleifer/peewee/issues/2502
1,077,359,121
I_kwDOAA7yGM5ANzIR
2,502
AttributeError: 'Expression' object has no attribute 'execute'
{ "login": "thisnugroho", "id": 49790011, "node_id": "MDQ6VXNlcjQ5NzkwMDEx", "avatar_url": "https://avatars.githubusercontent.com/u/49790011?v=4", "gravatar_id": "", "url": "https://api.github.com/users/thisnugroho", "html_url": "https://github.com/thisnugroho", "followers_url": "https://api.github.com/users/thisnugroho/followers", "following_url": "https://api.github.com/users/thisnugroho/following{/other_user}", "gists_url": "https://api.github.com/users/thisnugroho/gists{/gist_id}", "starred_url": "https://api.github.com/users/thisnugroho/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/thisnugroho/subscriptions", "organizations_url": "https://api.github.com/users/thisnugroho/orgs", "repos_url": "https://api.github.com/users/thisnugroho/repos", "events_url": "https://api.github.com/users/thisnugroho/events{/privacy}", "received_events_url": "https://api.github.com/users/thisnugroho/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "The `get_or_none()`, when used on a select query, does not accept any parameters -- though the error message is unhelpful here.\r\n\r\nIn your code you are:\r\n\r\n1. `SomeModel.select()` which creates a `ModelSelect` query\r\n2. Call the `get_or_none()` method on the `ModelSelect` query\r\n\r\nI think what you intended to do is this:\r\n\r\n```python\r\n\r\nSomeModel.select().where(SomeModel.field == value).get_or_none()\r\n```\r\n\r\nAlternatively, you can use the *shortcut* method on model, which *does* accept an expression:\r\n\r\n```python\r\n\r\nSomeModel.get_or_none(SomeModel.field == value)\r\n```\r\n\r\n* http://docs.peewee-orm.com/en/latest/peewee/api.html#Model.get_or_none" ]
"2021-12-11T02:11:25"
"2021-12-11T12:25:38"
"2021-12-11T12:25:38"
NONE
null
It's appear after upgrading from version 3.14.4 to peewee latest version 3.14.8, i decide to upgrade because i want to using `Model.get_or_none()` function, it's not available in previous version. and the error appear i'm trying to running this: ```python SomeModel().select().get_or_none(SomeModel.field== value) ``` my configuration looklike this: ```python db = MySQLDatabase( env("DB_DATABASE"), user=env("DB_USERNAME"), password=env("DB_PASSWORD"), host=env("DB_HOST"), port=int(env("DB_PORT")), ) ``` is there something that i need to adjust ? or i'm just missing something. screenshoot: ![image](https://user-images.githubusercontent.com/49790011/145660391-f7bf7cf7-b00a-4073-998d-9122faabe720.png) details OS:Ubuntu 20.04 MYSQL Version = 8.0.27
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2502/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2502/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2501
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2501/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2501/comments
https://api.github.com/repos/coleifer/peewee/issues/2501/events
https://github.com/coleifer/peewee/pull/2501
1,076,661,651
PR_kwDOAA7yGM4vq1ly
2,501
Allow to use peewee.SQL from `from_`
{ "login": "d0d0", "id": 553292, "node_id": "MDQ6VXNlcjU1MzI5Mg==", "avatar_url": "https://avatars.githubusercontent.com/u/553292?v=4", "gravatar_id": "", "url": "https://api.github.com/users/d0d0", "html_url": "https://github.com/d0d0", "followers_url": "https://api.github.com/users/d0d0/followers", "following_url": "https://api.github.com/users/d0d0/following{/other_user}", "gists_url": "https://api.github.com/users/d0d0/gists{/gist_id}", "starred_url": "https://api.github.com/users/d0d0/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/d0d0/subscriptions", "organizations_url": "https://api.github.com/users/d0d0/orgs", "repos_url": "https://api.github.com/users/d0d0/repos", "events_url": "https://api.github.com/users/d0d0/events{/privacy}", "received_events_url": "https://api.github.com/users/d0d0/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I don't really like this solution. The join will not correctly set-up the model graph, so the best solution in my opinion is to write:\r\n\r\n```python\r\nq = User.select(User, Tweet).from_(SQL(f'`users` as `t1` USE INDEX(`user_id`) ')).join(Tweet)\r\nfor x in q.objects():\r\n # assume username is on User table, and content is on Tweet table.\r\n print(x.username, x.content)\r\n```" ]
"2021-12-10T09:58:34"
"2021-12-10T17:46:55"
"2021-12-10T17:46:54"
NONE
null
**Problem:** Tried to apply custom SQL to `from_` function from this example https://github.com/coleifer/peewee/issues/1289#issuecomment-307128658 However in my query, I use some joins. Here is example ```python q = User.select(User, Tweet).from_(SQL(f'`users` as `t1` USE INDEX(`user_id`) ')).join(Tweet) for x in q: print(x) ``` But without fix, exception is thrown ``` Traceback (most recent call last): File "JetBrains\PyCharm2021.2\scratches\scratch_71.py", line 27, in <module> for file in q: File "venv\lib\site-packages\peewee.py", line 4408, in next self.cursor_wrapper.iterate() File "venv\lib\site-packages\peewee.py", line 4325, in iterate self.initialize() # Lazy initialization. File "venv\lib\site-packages\peewee.py", line 7587, in initialize if curr not in self.joins: TypeError: unhashable type: 'SQL' ``` **Solution** When initializing ModelCursorWrapper skip SQL instances
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2501/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2501/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2501", "html_url": "https://github.com/coleifer/peewee/pull/2501", "diff_url": "https://github.com/coleifer/peewee/pull/2501.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2501.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2500
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2500/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2500/comments
https://api.github.com/repos/coleifer/peewee/issues/2500/events
https://github.com/coleifer/peewee/issues/2500
1,074,152,643
I_kwDOAA7yGM5ABkTD
2,500
Query "where not in empty list" does not work as expected
{ "login": "skontar", "id": 10827040, "node_id": "MDQ6VXNlcjEwODI3MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/10827040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/skontar", "html_url": "https://github.com/skontar", "followers_url": "https://api.github.com/users/skontar/followers", "following_url": "https://api.github.com/users/skontar/following{/other_user}", "gists_url": "https://api.github.com/users/skontar/gists{/gist_id}", "starred_url": "https://api.github.com/users/skontar/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/skontar/subscriptions", "organizations_url": "https://api.github.com/users/skontar/orgs", "repos_url": "https://api.github.com/users/skontar/repos", "events_url": "https://api.github.com/users/skontar/events{/privacy}", "received_events_url": "https://api.github.com/users/skontar/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "The problem is python operator precedence. The following work as expected:\r\n\r\n```python\r\n\r\nPerson.select().where(Person.name.not_in([]))\r\n\r\nPerson.select().where(~(Person.name << []))\r\n\r\nPerson.select().where(~(Person.name.in_([])))\r\n```", "Thanks for reply. I see now, I got confused by SQL operator precedence. \r\nWould it not be better to translate to `WHERE (NOT \"t1\".\"name\" IN ())` for empty list, similarly as it does for non-empty list?\r\n\r\nI understand the reasoning and the problem, but I think this can potentially mask the issue until the list becomes empty one day and return unexpected results." ]
"2021-12-08T08:43:36"
"2021-12-08T15:59:19"
"2021-12-08T15:33:57"
NONE
null
I was using equivalent of `.where(~Person.name << some_list)` and it behaves unexpectedly when the list is empty. See below example. ```python import peewee from peewee import * db = SqliteDatabase(':memory:') class Person(Model): name = CharField() class Meta: database = db db.create_tables([Person]) Person(name="Alice").save() Person(name="Bob").save() Person(name="Cliff").save() q = Person.select().where(Person.name << ['Bob']) print(q) # SELECT "t1"."id", "t1"."name" FROM "person" AS "t1" WHERE ("t1"."name" IN ('Bob')) print(list(q.tuples())) # [(2, 'Bob')] q = Person.select().where(~Person.name << ['Bob']) print(q) # SELECT "t1"."id", "t1"."name" FROM "person" AS "t1" WHERE (NOT "t1"."name" IN ('Bob')) print(list(q.tuples())) # [(1, 'Alice'), (3, 'Cliff')] q = Person.select().where(Person.name << []) print(q) # SELECT "t1"."id", "t1"."name" FROM "person" AS "t1" WHERE (0 = 1) print(list(q.tuples())) # [] q = Person.select().where(~Person.name << []) print(q) # SELECT "t1"."id", "t1"."name" FROM "person" AS "t1" WHERE (0 = 1) # Expected: SELECT "t1"."id", "t1"."name" FROM "person" AS "t1" WHERE (1 = 1) ??? Maybe? print(list(q.tuples())) # [] # Expected: [(1, 'Alice'), (2, 'Bob'), (3, 'Cliff')] print(peewee.__version__) # 3.14.8 print(peewee.sqlite3.sqlite_version) # 3.34.1 ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2500/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2500/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2499
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2499/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2499/comments
https://api.github.com/repos/coleifer/peewee/issues/2499/events
https://github.com/coleifer/peewee/pull/2499
1,073,938,441
PR_kwDOAA7yGM4vh8NM
2,499
Add BlackSheep integration
{ "login": "q0w", "id": 43147888, "node_id": "MDQ6VXNlcjQzMTQ3ODg4", "avatar_url": "https://avatars.githubusercontent.com/u/43147888?v=4", "gravatar_id": "", "url": "https://api.github.com/users/q0w", "html_url": "https://github.com/q0w", "followers_url": "https://api.github.com/users/q0w/followers", "following_url": "https://api.github.com/users/q0w/following{/other_user}", "gists_url": "https://api.github.com/users/q0w/gists{/gist_id}", "starred_url": "https://api.github.com/users/q0w/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/q0w/subscriptions", "organizations_url": "https://api.github.com/users/q0w/orgs", "repos_url": "https://api.github.com/users/q0w/repos", "events_url": "https://api.github.com/users/q0w/events{/privacy}", "received_events_url": "https://api.github.com/users/q0w/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I really can't keep up with these new async-framework-of-the-day things. Additionally, connection handling doesn't work the way you might expect with async, so this is probably incorrect anyways." ]
"2021-12-08T02:20:08"
"2021-12-09T08:27:38"
"2021-12-08T23:43:29"
NONE
null
null
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2499/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2499/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2499", "html_url": "https://github.com/coleifer/peewee/pull/2499", "diff_url": "https://github.com/coleifer/peewee/pull/2499.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2499.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2498
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2498/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2498/comments
https://api.github.com/repos/coleifer/peewee/issues/2498/events
https://github.com/coleifer/peewee/issues/2498
1,071,183,911
I_kwDOAA7yGM4_2Pgn
2,498
Peewee 2 does not work with python 3.10
{ "login": "ribx", "id": 4598953, "node_id": "MDQ6VXNlcjQ1OTg5NTM=", "avatar_url": "https://avatars.githubusercontent.com/u/4598953?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ribx", "html_url": "https://github.com/ribx", "followers_url": "https://api.github.com/users/ribx/followers", "following_url": "https://api.github.com/users/ribx/following{/other_user}", "gists_url": "https://api.github.com/users/ribx/gists{/gist_id}", "starred_url": "https://api.github.com/users/ribx/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ribx/subscriptions", "organizations_url": "https://api.github.com/users/ribx/orgs", "repos_url": "https://api.github.com/users/ribx/repos", "events_url": "https://api.github.com/users/ribx/events{/privacy}", "received_events_url": "https://api.github.com/users/ribx/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "That version is not receiving patches, unfortunately. Your best bet is to fork it and patch it. You won't lose sync with upstream if you stay on 2.x." ]
"2021-12-04T11:06:22"
"2021-12-04T13:11:43"
"2021-12-04T13:11:42"
NONE
null
The problem is, that the exports of the `collections` modules were moved to `collections.abc`: ``` docker run --rm -ti python:3.10 sh -c 'pip install cython &>/dev/null && pip install peewee==2.10.2' .... Collecting peewee==2.10.2 Downloading peewee-2.10.2.tar.gz (516 kB) |████████████████████████████████| 516 kB 3.7 MB/s ERROR: Command errored out with exit status 1: command: /usr/local/bin/python -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-i073ept4/peewee_0592019e5eb643a48b6ef02e66fcd5d8/setup.py'"'"'; __file__='"'"'/tmp/pip-install-i073ept4/peewee_0592019e5eb643a48b6ef02e66fcd5d8/setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(__file__) if os.path.exists(__file__) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' egg_info --egg-base /tmp/pip-pip-egg-info-efaxgva5 cwd: /tmp/pip-install-i073ept4/peewee_0592019e5eb643a48b6ef02e66fcd5d8/ Complete output (7 lines): Traceback (most recent call last): File "<string>", line 1, in <module> File "/tmp/pip-install-i073ept4/peewee_0592019e5eb643a48b6ef02e66fcd5d8/setup.py", line 64, in <module> version=__import__('peewee').__version__, File "/tmp/pip-install-i073ept4/peewee_0592019e5eb643a48b6ef02e66fcd5d8/peewee.py", line 124, in <module> from collections import Callable ImportError: cannot import name 'Callable' from 'collections' (/usr/local/lib/python3.10/collections/__init__.py) ---------------------------------------- ``` ``` $ docker run --rm -ti python:3.9 >>> from collections import Callable <stdin>:1: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3, and in 3.10 it will stop working ``` The fix should be quite simple, by just replacing `collections` with `collections.abc`. I am stuck to peewee 2 because I am using this extension: https://github.com/cour4g3/peewee-mssql I can also create a PR.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2498/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2498/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2497
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2497/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2497/comments
https://api.github.com/repos/coleifer/peewee/issues/2497/events
https://github.com/coleifer/peewee/issues/2497
1,070,778,439
I_kwDOAA7yGM4_0shH
2,497
Support for "<@" in PostgreSQL arrays
{ "login": "joaodlf", "id": 3275379, "node_id": "MDQ6VXNlcjMyNzUzNzk=", "avatar_url": "https://avatars.githubusercontent.com/u/3275379?v=4", "gravatar_id": "", "url": "https://api.github.com/users/joaodlf", "html_url": "https://github.com/joaodlf", "followers_url": "https://api.github.com/users/joaodlf/followers", "following_url": "https://api.github.com/users/joaodlf/following{/other_user}", "gists_url": "https://api.github.com/users/joaodlf/gists{/gist_id}", "starred_url": "https://api.github.com/users/joaodlf/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/joaodlf/subscriptions", "organizations_url": "https://api.github.com/users/joaodlf/orgs", "repos_url": "https://api.github.com/users/joaodlf/repos", "events_url": "https://api.github.com/users/joaodlf/events{/privacy}", "received_events_url": "https://api.github.com/users/joaodlf/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Thanks for the suggestion - this is implemented as `contained_by`.", "Thanks for the **incredible** feedback in such a short notice, do you plan on releasing a minor version with this feature any time in the near future?", "Besides this one change, there aren't really any other notable changes yet to justify a new release, in my opinion -- everything else is just docs or example updates. I'll probably wait until after the holidays.\r\n\r\nIf you want to use this now, but don't want to pin to `master` branch, I'd suggest just implementing your own subclass of ArrayField that implements `contained_by()` and using that in the meantime." ]
"2021-12-03T16:51:46"
"2021-12-06T13:57:46"
"2021-12-03T17:10:59"
NONE
null
Peewee currently supports `@>` (`.contains()`) and `&&` (`contains_any()`) - [docs](https://docs.peewee-orm.com/en/latest/peewee/playhouse.html?highlight=array#ArrayField.contains) It would be nice for Peewee to also support `<@`, as expressed in the [PostgreSQL docs](https://www.postgresql.org/docs/current/functions-array.html): > anyarray <@ anyarray → boolean > > Is the first array contained by the second? > > ARRAY[2,2,7] <@ ARRAY[1,7,4,2,6] → t Note that this is a completely different behavior to `@>`: > anyarray @> anyarray → boolean > > Does the first array contain the second, that is, does each element appearing in the second array equal some element of the first array? (Duplicates are not treated specially, thus ARRAY[1] and ARRAY[1,1] are each considered to contain the other.) > > ARRAY[1,4,3] @> ARRAY[3,1,3] → t I would imagine this implemented as something like `.contained_by()`?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2497/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2497/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2496
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2496/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2496/comments
https://api.github.com/repos/coleifer/peewee/issues/2496/events
https://github.com/coleifer/peewee/pull/2496
1,070,550,586
PR_kwDOAA7yGM4vW_gK
2,496
Fix link to SQLAlchemy Hybrid Attributes docs
{ "login": "jonathanmach", "id": 12788052, "node_id": "MDQ6VXNlcjEyNzg4MDUy", "avatar_url": "https://avatars.githubusercontent.com/u/12788052?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jonathanmach", "html_url": "https://github.com/jonathanmach", "followers_url": "https://api.github.com/users/jonathanmach/followers", "following_url": "https://api.github.com/users/jonathanmach/following{/other_user}", "gists_url": "https://api.github.com/users/jonathanmach/gists{/gist_id}", "starred_url": "https://api.github.com/users/jonathanmach/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/jonathanmach/subscriptions", "organizations_url": "https://api.github.com/users/jonathanmach/orgs", "repos_url": "https://api.github.com/users/jonathanmach/repos", "events_url": "https://api.github.com/users/jonathanmach/events{/privacy}", "received_events_url": "https://api.github.com/users/jonathanmach/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[]
"2021-12-03T12:39:40"
"2021-12-03T17:35:56"
"2021-12-03T14:56:05"
CONTRIBUTOR
null
Broken link: http://docs.sqlalchemy.org/en/improve_toc/orm/extensions/hybrid.html New link: https://docs.sqlalchemy.org/en/14/orm/extensions/hybrid.html
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2496/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2496/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2496", "html_url": "https://github.com/coleifer/peewee/pull/2496", "diff_url": "https://github.com/coleifer/peewee/pull/2496.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2496.patch", "merged_at": "2021-12-03T14:56:05" }
https://api.github.com/repos/coleifer/peewee/issues/2495
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2495/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2495/comments
https://api.github.com/repos/coleifer/peewee/issues/2495/events
https://github.com/coleifer/peewee/issues/2495
1,069,862,567
I_kwDOAA7yGM4_xM6n
2,495
TypeError: int() argument must be a string, a bytes-like object or a number when importing models
{ "login": "koadjunky", "id": 8714421, "node_id": "MDQ6VXNlcjg3MTQ0MjE=", "avatar_url": "https://avatars.githubusercontent.com/u/8714421?v=4", "gravatar_id": "", "url": "https://api.github.com/users/koadjunky", "html_url": "https://github.com/koadjunky", "followers_url": "https://api.github.com/users/koadjunky/followers", "following_url": "https://api.github.com/users/koadjunky/following{/other_user}", "gists_url": "https://api.github.com/users/koadjunky/gists{/gist_id}", "starred_url": "https://api.github.com/users/koadjunky/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/koadjunky/subscriptions", "organizations_url": "https://api.github.com/users/koadjunky/orgs", "repos_url": "https://api.github.com/users/koadjunky/repos", "events_url": "https://api.github.com/users/koadjunky/events{/privacy}", "received_events_url": "https://api.github.com/users/koadjunky/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I'm not going to download or run that code, but you are welcome to paste a short snippet demonstrating the issue - or even just the actual traceback...", "```\r\n(PeeWee-error) maciej@pippin:~/PeeWee-error$ python dsp_be/logic/planet.py \r\nTraceback (most recent call last):\r\n File \"dsp_be/logic/planet.py\", line 14, in <module>\r\n factory = Factory.create(planet=earth)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 6393, in create\r\n inst.save(force_insert=True)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 6603, in save\r\n pk = self.insert(**field_dict).execute()\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 1911, in inner\r\n return method(self, database, *args, **kwargs)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 1982, in execute\r\n return self._execute(database)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 2761, in _execute\r\n return super(Insert, self)._execute(database)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 2479, in _execute\r\n cursor = database.execute(self)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 3172, in execute\r\n sql, params = ctx.sql(query).query()\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 614, in sql\r\n return obj.__sql__(self)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 2738, in __sql__\r\n self._simple_insert(ctx)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 2588, in _simple_insert\r\n return self._generate_insert((self._insert,), ctx)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 2710, in _generate_insert\r\n return ctx.sql(CommaNodeList(all_values))\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 614, in sql\r\n return obj.__sql__(self)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 1777, in __sql__\r\n ctx.sql(self.nodes[n_nodes - 1])\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 614, in sql\r\n return obj.__sql__(self)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 1777, in __sql__\r\n ctx.sql(self.nodes[n_nodes - 1])\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 614, in sql\r\n return obj.__sql__(self)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 1383, in __sql__\r\n return ctx.value(self.value, self.converter)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 626, in value\r\n value = converter(value)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 5242, in db_value\r\n return self.rel_field.db_value(value)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 4569, in db_value\r\n return value if value is None else self.adapt(value)\r\n File \"/home/maciej/.pyenv/versions/PeeWee-error/lib/python3.8/site-packages/peewee.py\", line 4629, in adapt\r\n return int(value)\r\nTypeError: int() argument must be a string, a bytes-like object or a number, not 'Planet'\r\n```", "Looks like you have code here on line 14:\r\n\r\n```python\r\nfactory = Factory.create(planet=earth)\r\n```\r\n\r\nThe definition of `Factory.planet` (I'm guessing) is not a `ForeignKeyField` perhaps, when it should be. Possibly you put it as an IntegerField or something?", "It only happens if models are in different files. Merge it into one - and it works.\r\n\r\n```\r\n(PeeWee-error) maciej@pippin:~/PeeWee-error$ cat dsp_be/logic/__init__.py \r\nfrom typing import List, Type\r\n\r\nfrom peewee import DatabaseProxy, SqliteDatabase, Model\r\n\r\ndb_proxy = DatabaseProxy()\r\n\r\n\r\nclass BaseModel(Model):\r\n class Meta:\r\n database = db_proxy\r\n\r\n\r\ndef test_database(tables: List[Type[Model]]):\r\n db = SqliteDatabase(':memory:')\r\n db_proxy.initialize(db)\r\n db_proxy.connect()\r\n db_proxy.create_tables(tables)\r\n```\r\n\r\n```\r\n(PeeWee-error) maciej@pippin:~/PeeWee-error$ cat dsp_be/logic/planet.py \r\nfrom peewee import Field, CharField\r\n\r\nfrom dsp_be.logic import test_database, BaseModel\r\n\r\n\r\nclass Planet(BaseModel):\r\n name: Field = CharField()\r\n\r\n\r\nif __name__ == '__main__':\r\n from dsp_be.logic.factory import Factory\r\n test_database([Planet, Factory])\r\n earth = Planet.create(name='Earth')\r\n factory = Factory.create(planet=earth)\r\n```\r\n\r\n```\r\n(PeeWee-error) maciej@pippin:~/PeeWee-error$ cat dsp_be/logic/factory.py \r\nfrom peewee import Field, ForeignKeyField\r\n\r\nfrom dsp_be.logic import test_database, BaseModel\r\nfrom dsp_be.logic.planet import Planet\r\n\r\n\r\nclass Factory(BaseModel):\r\n planet: Field = ForeignKeyField(Planet, backref=\"planets\")\r\n\r\n\r\nif __name__ == '__main__':\r\n test_database([Planet, Factory])\r\n earth = Planet.create(name='Earth')\r\n factory = Factory.create(planet=earth)\r\n```", "You've got some bug in your code. This works fine for me:\r\n\r\n```python\r\n# testx/__init__.py\r\nfrom peewee import *\r\n\r\ndb = DatabaseProxy()\r\n\r\nclass BaseModel(Model):\r\n class Meta:\r\n database = db\r\n\r\n# testx/factory.py\r\nfrom peewee import ForeignKeyField\r\nfrom testx import BaseModel\r\nfrom testx.planet import Planet\r\n\r\nclass Factory(BaseModel):\r\n planet = ForeignKeyField(Planet)\r\n\r\n# testx/planet.py\r\nfrom peewee import TextField\r\nfrom testx import BaseModel\r\n\r\nclass Planet(BaseModel):\r\n name = TextField()\r\n\r\n# testx/main.py\r\nfrom peewee import *\r\nfrom testx import db\r\nfrom testx.planet import Planet\r\nfrom testx.factory import Factory\r\n\r\n\r\n_db = SqliteDatabase(':memory:')\r\ndb.initialize(_db)\r\ndb.create_tables([Factory, Planet])\r\n\r\np = Planet.create(name='earth')\r\nFactory.create(planet=p)\r\nprint(Factory.get().planet.name) # \"earth\"\r\n```", "I copied part of planet.py into main.py similar to your snippet and now it works. Weird.\r\n```\r\n# dsp_be/logic/main.py\r\nfrom peewee import SqliteDatabase\r\n\r\nfrom dsp_be.logic import test_database\r\n\r\n\r\nif __name__ == '__main__':\r\n from dsp_be.logic.factory import Factory, Planet\r\n test_database([Factory, Planet])\r\n earth = Planet.create(name='Earth')\r\n factory = Factory.create(planet=earth)\r\n```\r\nCan be used as an workaround through. Thanks for hint!" ]
"2021-12-02T18:33:08"
"2021-12-02T20:47:19"
"2021-12-02T20:01:14"
NONE
null
Python: 3.8.11 peewee: 3.14.8 I have two files with two models connected with foreign key: factory.py and planet.py. Each file has identical test code (except imports). Calling 'python dsp_be/logic/factory.py' works. Calling 'python dsp_be/logic/planet.py' throws an exception: TypeError: int() argument must be a string, a bytes-like object or a number, not 'Planet' Minimal code demonstrating the issue attached. Add "." to PYTHONPATH before executing. [PeeWee-error.zip](https://github.com/coleifer/peewee/files/7644256/PeeWee-error.zip) The Internet is silent about the issue, I only found this https://stackoverflow.com/questions/26181642/peewee-foreignkeyfield-doesnt-accept-appropriate-foreign-key-field but the solution is not working for me (outdated?). I'll be grateful for any hint.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2495/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2495/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2494
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2494/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2494/comments
https://api.github.com/repos/coleifer/peewee/issues/2494/events
https://github.com/coleifer/peewee/issues/2494
1,067,754,583
I_kwDOAA7yGM4_pKRX
2,494
pwiz uses wildcard imports of peewee
{ "login": "tdennisliu", "id": 34557499, "node_id": "MDQ6VXNlcjM0NTU3NDk5", "avatar_url": "https://avatars.githubusercontent.com/u/34557499?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tdennisliu", "html_url": "https://github.com/tdennisliu", "followers_url": "https://api.github.com/users/tdennisliu/followers", "following_url": "https://api.github.com/users/tdennisliu/following{/other_user}", "gists_url": "https://api.github.com/users/tdennisliu/gists{/gist_id}", "starred_url": "https://api.github.com/users/tdennisliu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tdennisliu/subscriptions", "organizations_url": "https://api.github.com/users/tdennisliu/orgs", "repos_url": "https://api.github.com/users/tdennisliu/repos", "events_url": "https://api.github.com/users/tdennisliu/events{/privacy}", "received_events_url": "https://api.github.com/users/tdennisliu/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "This is a convention that is used throughout peewee examples, as well as my own code. Peewee module defines `__all__`, which controls what objects are exported, and I am happy with that. In your own code you are welcome to use whatever style you like, though." ]
"2021-11-30T22:57:07"
"2021-12-01T00:38:23"
"2021-12-01T00:38:23"
NONE
null
Hiya, I've found pwiz very helpful in setting up my scripts, but since it wildcard imports `peewee` into the script. See below from the README: ``` from peewee import * import datetime db = SqliteDatabase('my_database.db') class BaseModel(Model): class Meta: database = db class User(BaseModel): username = CharField(unique=True) class Tweet(BaseModel): user = ForeignKeyField(User, backref='tweets') message = TextField() created_date = DateTimeField(default=datetime.datetime.now) is_published = BooleanField(default=True) ``` I find this very difficult to troubleshoot if something goes wrong, and it takes a long time to simply find and replace the classes. What I would like to see: ``` import peewee as pw import datetime db = pw.SqliteDatabase('my_database.db') class BaseModel(pw.Model): class Meta: database = db class User(BaseModel): username = pw.CharField(unique=True) class Tweet(BaseModel): user = pw.ForeignKeyField(User, backref='tweets') message = pw.TextField() created_date = pw.DateTimeField(default=datetime.datetime.now) is_published = pw.BooleanField(default=True) ``` The [python docs](https://docs.python.org/3/tutorial/modules.html#importing-from-a-package) do state that this practise should be avoided. As far as I can tell it's not too difficult to change this behaviour for pwiz, and I can set up a PR. Thoughts?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2494/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2494/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2493
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2493/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2493/comments
https://api.github.com/repos/coleifer/peewee/issues/2493/events
https://github.com/coleifer/peewee/issues/2493
1,065,846,586
I_kwDOAA7yGM4_h4c6
2,493
fn.Avg() gets coerced into integer
{ "login": "gjask", "id": 3306648, "node_id": "MDQ6VXNlcjMzMDY2NDg=", "avatar_url": "https://avatars.githubusercontent.com/u/3306648?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gjask", "html_url": "https://github.com/gjask", "followers_url": "https://api.github.com/users/gjask/followers", "following_url": "https://api.github.com/users/gjask/following{/other_user}", "gists_url": "https://api.github.com/users/gjask/gists{/gist_id}", "starred_url": "https://api.github.com/users/gjask/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gjask/subscriptions", "organizations_url": "https://api.github.com/users/gjask/orgs", "repos_url": "https://api.github.com/users/gjask/repos", "events_url": "https://api.github.com/users/gjask/events{/privacy}", "received_events_url": "https://api.github.com/users/gjask/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "You want to specify `.coerce(False)`:\r\n\r\n```python\r\n\r\nTestTable.select(\r\n fn.Avg(TestTable.value).coerce(False).alias(\"avg_value\"),\r\n).scalar()\r\n```", "http://docs.peewee-orm.com/en/latest/peewee/api.html#Function.coerce" ]
"2021-11-29T10:09:09"
"2021-11-29T13:51:52"
"2021-11-29T13:50:55"
NONE
null
When using `fn.Avg()` on integer field, result will also gets coerced into integer. To illustrate problem I have created this example. ```python class TestTable(Model): id = AutoField() value = IntegerField() ``` ```sql insert into testtable (value) values (3), (4); select * from testtable; +----+-------+ | id | value | +----+-------+ | 1 | 3 | | 2 | 4 | +----+-------+ ``` ```python val = TestTable.select( fn.Avg(TestTable.value).alias("avg_value"), ).scalar() print(val, type(val) # 3 <class 'int'> ``` The expected result would be `3.5` or `Decimal(3.5)` peewee = 3.14.8 database is MySQL
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2493/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2493/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2492
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2492/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2492/comments
https://api.github.com/repos/coleifer/peewee/issues/2492/events
https://github.com/coleifer/peewee/issues/2492
1,064,369,878
I_kwDOAA7yGM4_cP7W
2,492
Bug related to create_table function in your Model
{ "login": "kezzhang", "id": 42364982, "node_id": "MDQ6VXNlcjQyMzY0OTgy", "avatar_url": "https://avatars.githubusercontent.com/u/42364982?v=4", "gravatar_id": "", "url": "https://api.github.com/users/kezzhang", "html_url": "https://github.com/kezzhang", "followers_url": "https://api.github.com/users/kezzhang/followers", "following_url": "https://api.github.com/users/kezzhang/following{/other_user}", "gists_url": "https://api.github.com/users/kezzhang/gists{/gist_id}", "starred_url": "https://api.github.com/users/kezzhang/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/kezzhang/subscriptions", "organizations_url": "https://api.github.com/users/kezzhang/orgs", "repos_url": "https://api.github.com/users/kezzhang/repos", "events_url": "https://api.github.com/users/kezzhang/events{/privacy}", "received_events_url": "https://api.github.com/users/kezzhang/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I could swear I ran all the examples and they worked so possibly something has changed. 2.x is unmaintained since Feb 2018 so hard to say." ]
"2021-11-26T11:22:24"
"2021-11-26T12:44:29"
"2021-11-26T12:44:29"
NONE
null
It is actually about one of your examples in version 2.9.1 and before, redis_vtable, in line 208, RedisView.create_table(), this will report error. But when I change that line to something like cursor.execute("create virtual table redisview using RedisModule()"), the whole program worked. So I highly doubt that there is something wrong with your create_table function especially under the apsw_ext situation. Besides, when you truly want to do extensions for something, doing it halfway is worse than nothing.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2492/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2492/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2491
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2491/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2491/comments
https://api.github.com/repos/coleifer/peewee/issues/2491/events
https://github.com/coleifer/peewee/issues/2491
1,062,176,736
I_kwDOAA7yGM4_T4fg
2,491
FlaskDb wrapper from extension
{ "login": "nabinbhusal80", "id": 24907850, "node_id": "MDQ6VXNlcjI0OTA3ODUw", "avatar_url": "https://avatars.githubusercontent.com/u/24907850?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nabinbhusal80", "html_url": "https://github.com/nabinbhusal80", "followers_url": "https://api.github.com/users/nabinbhusal80/followers", "following_url": "https://api.github.com/users/nabinbhusal80/following{/other_user}", "gists_url": "https://api.github.com/users/nabinbhusal80/gists{/gist_id}", "starred_url": "https://api.github.com/users/nabinbhusal80/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/nabinbhusal80/subscriptions", "organizations_url": "https://api.github.com/users/nabinbhusal80/orgs", "repos_url": "https://api.github.com/users/nabinbhusal80/repos", "events_url": "https://api.github.com/users/nabinbhusal80/events{/privacy}", "received_events_url": "https://api.github.com/users/nabinbhusal80/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I've implemented this, thanks for the suggestion. The endpoints can be specified using:\r\n\r\n```python\r\nFLASKDB_EXCLUDED_ROUTES = ('view1', 'view2'...)\r\n\r\n# Or,\r\nflask_db = FlaskDB(app, excluded_routes=set('view1', 'view2',...))\r\n```", "Hey @coleifer thank you so much for the solution. Somehow i missed it in docs. I was banging my head on this for a while. So amazing to see the quick response on this. \r\nI really appreciate it." ]
"2021-11-24T09:13:45"
"2021-11-24T18:49:11"
"2021-11-24T14:30:09"
NONE
null
Hi I am using FlaskDb on my flask application which always opens connection to db on each request and closes connection once request ends. This is a great feature but I have some endpoints which don't require anything from db and they kind of live for long time because complex computations. How do I exempt db connections for specific endpoints? Or is there any other way i can use this? Your help will be appreciated.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2491/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2491/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2490
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2490/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2490/comments
https://api.github.com/repos/coleifer/peewee/issues/2490/events
https://github.com/coleifer/peewee/issues/2490
1,060,162,877
I_kwDOAA7yGM4_MM09
2,490
Remote Postgres Use Case
{ "login": "aiqc", "id": 74990642, "node_id": "MDQ6VXNlcjc0OTkwNjQy", "avatar_url": "https://avatars.githubusercontent.com/u/74990642?v=4", "gravatar_id": "", "url": "https://api.github.com/users/aiqc", "html_url": "https://github.com/aiqc", "followers_url": "https://api.github.com/users/aiqc/followers", "following_url": "https://api.github.com/users/aiqc/following{/other_user}", "gists_url": "https://api.github.com/users/aiqc/gists{/gist_id}", "starred_url": "https://api.github.com/users/aiqc/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/aiqc/subscriptions", "organizations_url": "https://api.github.com/users/aiqc/orgs", "repos_url": "https://api.github.com/users/aiqc/repos", "events_url": "https://api.github.com/users/aiqc/events{/privacy}", "received_events_url": "https://api.github.com/users/aiqc/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Yes, peewee supports all the same parameters as psycopg2, so:\r\n\r\n```python\r\n\r\n# for example:\r\ndb = PostgresqlDatabase('db_name', host='192.168.1.30', port=5432, user='postgres', password='xyz')\r\n```\r\n\r\nPsycopg2 connection doc: https://www.psycopg.org/docs/module.html#psycopg2.connect" ]
"2021-11-22T13:32:39"
"2021-11-22T13:42:07"
"2021-11-22T13:42:07"
NONE
null
Hiya. It sounded like you were fishing for use cases with respect to Postgres in the docs. > https://docs.peewee-orm.com/en/latest/peewee/playhouse.html?highlight=postgresql#PostgresqlExtDatabase > > In the future I would like to add support for more of postgresql’s features. If there is a particular feature you would like to see added, please open a Github issue. I'm using and love PeeWee w SQLite for local development and vertical scale (single desktop/ server). It's a great way to provide a 1-line install for contributors and open source users. I was hoping to drop a SQLite file into AWS EFS and mount many jobs/users to it simultaneously, but concurrent writes over NFS are a no-go zone for SQLite. Right now I have to manually manage my own lock file. So my solution would be to serve PSQL via AWS RDS. I searched for "postgresql" in the peewee docs because I don't want to switch ORMs. However, I didn't see anything like database IP address or port - so I was wondering if this implementation would support a remote psql db?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2490/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2490/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2489
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2489/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2489/comments
https://api.github.com/repos/coleifer/peewee/issues/2489/events
https://github.com/coleifer/peewee/issues/2489
1,058,826,022
I_kwDOAA7yGM4_HGcm
2,489
CharField with unique constraint is case insensitive
{ "login": "rmskinsa2", "id": 86251795, "node_id": "MDQ6VXNlcjg2MjUxNzk1", "avatar_url": "https://avatars.githubusercontent.com/u/86251795?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rmskinsa2", "html_url": "https://github.com/rmskinsa2", "followers_url": "https://api.github.com/users/rmskinsa2/followers", "following_url": "https://api.github.com/users/rmskinsa2/following{/other_user}", "gists_url": "https://api.github.com/users/rmskinsa2/gists{/gist_id}", "starred_url": "https://api.github.com/users/rmskinsa2/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/rmskinsa2/subscriptions", "organizations_url": "https://api.github.com/users/rmskinsa2/orgs", "repos_url": "https://api.github.com/users/rmskinsa2/repos", "events_url": "https://api.github.com/users/rmskinsa2/events{/privacy}", "received_events_url": "https://api.github.com/users/rmskinsa2/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "https://stackoverflow.com/questions/463764/are-unique-indices-case-sensitive-in-mysql", "@coleifer Is there a way to make the column case sensitive via PeeWee, or is this something we need to do directly in the db?", "You should be able to specify the collation directly via Peewee:\r\n\r\n```python\r\nkey = CharField(collation='utf8_bin', unique=True)\r\n\r\n# Specify character set as well:\r\nkey = CharField(collation='utf8_bin', constraints=[SQL('character set utf8')], unique=True)\r\n```\r\n\r\nI haven't tested this myself, but this is based somewhat on the above-linked StackOverflow answer.", "Thank you! I'll try that and report back here in case someone else needs the info in future.", "For future folks: changing the `collation` and `constraints` kwargs on the `CharField`, followed by dropping/adding the column with Playhouse didn't change anything. But running some raw sql did:\r\n```\r\n # Make the name column case sensitive\r\n collate_sql = \"\"\"\r\n ALTER TABLE db.tbl\r\n MODIFY COLUMN col VARCHAR(255) COLLATE utf8_bin;\r\n \"\"\"\r\n db.execute_sql(collate_sql)\r\n```\r\nVerified by running `SHOW FULL COLUMNS FROM tbl;`\r\n\r\nNOTE the `MODIFY COLUMN` vs `ALTER COLUMN`. When I used `ALTER COLUMN` I kept getting a syntax error. But for some reason `MODIFY COLUMN` worked. I'm too unfamiliar with db administration and too strapped on time to know why, but figured I'd include that factoid." ]
"2021-11-19T19:01:51"
"2021-11-23T17:26:46"
"2021-11-19T19:46:46"
NONE
null
I have a `CharField` defined like so: ``` class ToucanDomain(db.Model): name = CharField(unique=True, null=False) ``` I would like to create two entries, one with `name='SaTScan'` and another with `name='satscan'`. However, I get the following error, because on `create` the field appears to be case insensitive. ``` >>> ToucanDomain.create(name='SaTScan') <ToucanDomain: 1> >>> ToucanDomain.create(name='satscan') Traceback (most recent call last): File "/usr/local/lib/python3.7/site-packages/peewee.py", line 3099, in execute_sql cursor.execute(sql, params or ()) File "/usr/local/lib/python3.7/site-packages/pymysql/cursors.py", line 170, in execute result = self._query(query) File "/usr/local/lib/python3.7/site-packages/pymysql/cursors.py", line 328, in _query conn.query(q) File "/usr/local/lib/python3.7/site-packages/pymysql/connections.py", line 517, in query self._affected_rows = self._read_query_result(unbuffered=unbuffered) File "/usr/local/lib/python3.7/site-packages/pymysql/connections.py", line 732, in _read_query_result result.read() File "/usr/local/lib/python3.7/site-packages/pymysql/connections.py", line 1075, in read first_packet = self.connection._read_packet() File "/usr/local/lib/python3.7/site-packages/pymysql/connections.py", line 684, in _read_packet packet.check_error() File "/usr/local/lib/python3.7/site-packages/pymysql/protocol.py", line 220, in check_error err.raise_mysql_exception(self._data) File "/usr/local/lib/python3.7/site-packages/pymysql/err.py", line 109, in raise_mysql_exception raise errorclass(errno, errval) pymysql.err.IntegrityError: (1062, "Duplicate entry 'satscan' for key 'toucandomain_name'") ``` Is this expected behavior? If so, is there a way around it?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2489/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2489/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2488
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2488/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2488/comments
https://api.github.com/repos/coleifer/peewee/issues/2488/events
https://github.com/coleifer/peewee/issues/2488
1,058,357,797
I_kwDOAA7yGM4_FUIl
2,488
how to make order_by smarter
{ "login": "viponedream", "id": 16624529, "node_id": "MDQ6VXNlcjE2NjI0NTI5", "avatar_url": "https://avatars.githubusercontent.com/u/16624529?v=4", "gravatar_id": "", "url": "https://api.github.com/users/viponedream", "html_url": "https://github.com/viponedream", "followers_url": "https://api.github.com/users/viponedream/followers", "following_url": "https://api.github.com/users/viponedream/following{/other_user}", "gists_url": "https://api.github.com/users/viponedream/gists{/gist_id}", "starred_url": "https://api.github.com/users/viponedream/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/viponedream/subscriptions", "organizations_url": "https://api.github.com/users/viponedream/orgs", "repos_url": "https://api.github.com/users/viponedream/repos", "events_url": "https://api.github.com/users/viponedream/events{/privacy}", "received_events_url": "https://api.github.com/users/viponedream/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[]
"2021-11-19T10:08:36"
"2021-11-19T14:10:53"
"2021-11-19T14:10:53"
NONE
null
i want to accept a parameter, def get_all_order_by(self, _order_by) -> list: table = self.get_table() res = table.select().order_by(table._order_by.desc()) --> this doesnot work item_list = [d.to_item() for d in res] return item_list
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2488/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2488/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2487
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2487/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2487/comments
https://api.github.com/repos/coleifer/peewee/issues/2487/events
https://github.com/coleifer/peewee/issues/2487
1,053,927,057
I_kwDOAA7yGM4-0aaR
2,487
JSONField - Peewee cannot serialize Pendulum DateTime
{ "login": "johnziebro", "id": 90479072, "node_id": "MDQ6VXNlcjkwNDc5MDcy", "avatar_url": "https://avatars.githubusercontent.com/u/90479072?v=4", "gravatar_id": "", "url": "https://api.github.com/users/johnziebro", "html_url": "https://github.com/johnziebro", "followers_url": "https://api.github.com/users/johnziebro/followers", "following_url": "https://api.github.com/users/johnziebro/following{/other_user}", "gists_url": "https://api.github.com/users/johnziebro/gists{/gist_id}", "starred_url": "https://api.github.com/users/johnziebro/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/johnziebro/subscriptions", "organizations_url": "https://api.github.com/users/johnziebro/orgs", "repos_url": "https://api.github.com/users/johnziebro/repos", "events_url": "https://api.github.com/users/johnziebro/events{/privacy}", "received_events_url": "https://api.github.com/users/johnziebro/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ ">I'd like to better understand the roundtrip process for serializing and deserializing.\r\n\r\nThe sqlite3 json data-type requires the data to be serialized into json. Simple as that. When reading row data off a cursor the JSONField will deserialize it back to an object.\r\n\r\nThe problem is actually rather from python's own `json` module not supporting `datetime`-like out of the box. Which is a pretty glaring oversight in my opinion, but something to take up with the core team.", "@coleifer could you let me know if this would be a good strategy to follow?\r\n\r\nGenerally speaking, the [direction for encoding per this stackoverflow](https://stackoverflow.com/a/27058505/1663382) is to subclass json.JSONEncoder/JSONDecoder with a custom class, then provide it to json.dumps/loads as the 'cls' parameter using functools.partial. I believe what I was hung up on from the Peewee doc's was how to use the custom encoder and that json_loads was missing from the example.\r\n\r\n```\r\nclass DateTimeEncoder(JSONEncoder)\r\n # custom encode implementation here\r\n ...\r\n\r\nclass DateTimeDecoder(JSONDecoder)\r\n # custom decode implementation here\r\n ...\r\n\r\n# assign dumps and loads functions\r\nmy_json_dumps = functools.partial(json.dumps, cls=DateTimeEncoder)\r\nmy_json_loads = functools.partial(json.loads, cls=DateTimeDecoder)\r\n\r\nclass SomeModel(Model):\r\n # Specify our custom serialization function.\r\n json_data = JSONField(json_dumps=my_json_dumps, json_loads=my_json_loads)\r\n```", "There are a number of options: https://stackoverflow.com/questions/11875770/how-to-overcome-datetime-datetime-not-json-serializable", "For posterity or if anyone else needs an example as I did not see this in any other closed issues, @coleifer thanks again! \r\n\r\n### Solution\r\n```\r\nimport json\r\nimport pendulum as pdlm\r\nfrom functools import partial\r\n\r\n\r\nclass PendulumDateTimeEncoder(json.JSONEncoder):\r\n \"\"\"\r\n Provides json encoding as string for pendulum.DateTime objects\r\n including timezone or UTC.\r\n \"\"\"\r\n\r\n def default(self, object):\r\n\r\n # handle pendulum.DateTime objects\r\n if isinstance(object, pdlm.DateTime):\r\n\r\n # generate ISO 8601 string\r\n return object.to_iso8601_string()\r\n\r\n return super().default(object)\r\n\r\n\r\nclass PendulumDateTimeDecoder(json.JSONDecoder):\r\n \"\"\"\r\n Provides json datetime stored as string decoding for\r\n pendulum.DateTime objects.\r\n \"\"\"\r\n\r\n def decode(self, object):\r\n\r\n # handle pendulum.DateTime objects\r\n if isinstance(object, str):\r\n\r\n # load as json string\r\n object = json.loads(object)\r\n\r\n # return ISO 8601 string as pendulum DateTime\r\n return pdlm.parse(object)\r\n\r\n return super().decode(object)\r\n\r\n\r\n# assign dumps and loads functions\r\ncustom_dumps = partial(json.dumps, cls=PendulumDateTimeEncoder)\r\ncustom_loads = partial(json.loads, cls=PendulumDateTimeDecoder)\r\n```\r\n\r\n### Test\r\n```\r\nnow = pdlm.now('UTC')\r\ndumped = custom_dumps(now)\r\nloaded = custom_loads(dumped)\r\nassert loaded == now, \"Round-trip serialization failed.\"\r\nassert loaded.tz.name == \"UTC\", \"Timezone round-trip failure.\"\r\n```\r\n\r\n### ORM Class\r\n```\r\nclass TimelineModel(BaseModel):\r\n \"\"\" ORM model for Timeline. \"\"\"\r\n intervals = JSONField(json_dumps=custom_dumps, json_loads=custom_loads)\r\n```" ]
"2021-11-15T17:38:08"
"2021-11-15T20:21:22"
"2021-11-15T17:55:31"
NONE
null
### Issue The Pendulum module's purpose is to be a drop-in replacement for the standard datetime class. In most cases Pendulum's DateTime class can be treated the same as datetime.datetime. However, when attempting to save with Peewee into a JSONField, DateTime throws a TypeError as not serializable. Per [Pendulum's FAQ](https://pendulum.eustace.io/faq/) this issue is to be expected: > Unlike other datetime libraries for Python, Pendulum is a drop-in replacement for the standard datetime class (it inherits from it), so, basically, you can replace all your datetime instances by DateTime instances in you code (**_exceptions exist for libraries that check the type of the objects by using the type function like sqlite3 or PyMySQL for instance_**). ### Peewee Doc's Solution Per Peewee docs for [JSONField](http://docs.peewee-orm.com/en/latest/peewee/sqlite_ext.html?highlight=JSONField#JSONField), the solution is to customize the serialization: > To customize the JSON serialization or de-serialization, you can specify a custom json_dumps and json_loads callables. These functions should accept a single paramter: the object to serialize, and the JSON string, respectively. To modify the parameters of the stdlib JSON functions, you can use functools.partial ``` # Do not escape unicode code-points. my_json_dumps = functools.partial(json.dumps, ensure_ascii=False) class SomeModel(Model): # Specify our custom serialization function. json_data = JSONField(json_dumps=my_json_dumps) ``` ### Questions I'd like to better understand the roundtrip process for serializing and deserializing. In the future, it would be helpful for a more extensive example in the docs. The use case is to be able to save and load Pendulum DateTimes into and from Peewee. I don't exactly see how to accomplish this using partial with json.dumps. My understanding is that I would need to subclass json.JSONEncoder and json.JSONDecoder. **Could you provide some additional direction in how to accomplish this or a more efficient route? Maybe I am missing something about using json.dumps. Thank you in advance.** ### Error > Traceback (most recent call last): > ... > File "/home/----/Projects/rf/.venv/lib/python3.8/site-packages/playhouse/sqlite_ext.py", line 142, in db_value > value = fn.json(self._json_dumps(value)) > File "/usr/lib/python3.8/json/__init__.py", line 231, in dumps > return _default_encoder.encode(obj) > File "/usr/lib/python3.8/json/encoder.py", line 199, in encode > chunks = self.iterencode(o, _one_shot=True) > File "/usr/lib/python3.8/json/encoder.py", line 257, in iterencode > return _iterencode(o, 0) > File "/usr/lib/python3.8/json/encoder.py", line 179, in default > raise TypeError(f'Object of type {o.__class__.__name__} ' > TypeError: Object of type DateTime is not JSON serializable > python-BaseException
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2487/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2487/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2486
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2486/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2486/comments
https://api.github.com/repos/coleifer/peewee/issues/2486/events
https://github.com/coleifer/peewee/issues/2486
1,051,493,025
I_kwDOAA7yGM4-rIKh
2,486
`get_or_create` isn't automatically including the primary key
{ "login": "rbracco", "id": 47190785, "node_id": "MDQ6VXNlcjQ3MTkwNzg1", "avatar_url": "https://avatars.githubusercontent.com/u/47190785?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rbracco", "html_url": "https://github.com/rbracco", "followers_url": "https://api.github.com/users/rbracco/followers", "following_url": "https://api.github.com/users/rbracco/following{/other_user}", "gists_url": "https://api.github.com/users/rbracco/gists{/gist_id}", "starred_url": "https://api.github.com/users/rbracco/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/rbracco/subscriptions", "organizations_url": "https://api.github.com/users/rbracco/orgs", "repos_url": "https://api.github.com/users/rbracco/repos", "events_url": "https://api.github.com/users/rbracco/events{/privacy}", "received_events_url": "https://api.github.com/users/rbracco/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I'm getting some really weird results. I tried to hack around it by calling `ModelResult.get(audio_id=af.id, model_id=tm.id)` and manually updating if it existed, and if it didnt, catching the `DoesNotExist` and manually creating the `ModelResult` with\r\n\r\n```\r\nmax_id = ModelResult.select(fn.MAX(ModelResult.id)).scalar() + 1\r\nmr = ModelResult(\r\n id=max_id,\r\n prediction=pred,\r\n loss=loss,\r\n audio_id=af.id,\r\n model_id=tm.id,\r\n )\r\nmr.save()\r\n```\r\nIt worked for like 20 rows and then the rows stopped getting added and the ID stopped incrementing. I set a breakpoint and everything looks good in the ModelResult instance but mr.save() is returning 0. I searched and found #1198 which makes me think something is really wrong. \r\n\r\n**Edit:** I started using `mr.save(force_insert=True)` which worked and then checked the docs and saw that it happens when you create your own non-integer primary key which I don't think I did. My schema doesn't use `PrimaryKey` anywhere but I checked the schema in SQLite browser and I now see that the `id` field for `ModelResult` is `NUMERIC` not `INTEGER`. I'm not sure how this happened but I believe that's the issue. I'm going to explore a bit more and I'll post here and close out if I confirm. ", "There's nothing wrong with peewee, surely this is something in your code or schema.\r\n\r\n```python\r\nclass Audio(Base):\r\n name = TextField()\r\n\r\nclass Note(Base):\r\n name = TextField()\r\n\r\nclass Result(Base):\r\n audio = ForeignKeyField(Audio)\r\n note = ForeignKeyField(Note, null=True)\r\n\r\ndb.create_tables([Audio, Note, Result])\r\n\r\na1 = Audio.create(name='a1')\r\na2 = Audio.create(name='a2')\r\nn1 = Note.create(name='n1')\r\nn2 = Note.create(name='n2')\r\n\r\nr1a, created = Result.get_or_create(audio=a1)\r\nassert created\r\nr1b, created = Result.get_or_create(audio=a1)\r\nassert not created\r\nassert r1a.id == r1b.id\r\n\r\nr2a, created = Result.get_or_create(audio=a2, note=n2)\r\nassert created\r\nr2b, created = Result.get_or_create(audio=a2, note=n2)\r\nassert not created\r\nassert r2a.id == r2b.id\r\n```" ]
"2021-11-12T01:21:19"
"2021-11-12T13:47:52"
"2021-11-12T13:47:52"
NONE
null
I have a field `ModelResult` with a foreignkey for an `Audio` class. I recently updated `ModelResult` to take a 2nd foreign key for a `Model` class. When I call `get_or_create` I get the error `IntegrityError: NOT NULL constraint failed: modelresult.id`. For some reason the id isn't being automatically included/incremented, but if I include it manually in the defaults (e.g. {"id": 10000000}, it works). My intention is that if a row exists with that combination of audio and model, update it, otherwise create it. Any idea where I might be going wrong? Thank you by the way, peewee is a pleasure to work with. ``` audio = AudioFile.get(AudioFile.filename == fname) tm, created = TrainModel.get_or_create(name=model_name) defaults = { "prediction": pred, "loss": loss, } mr, created = ModelResult.get_or_create(audio_id=audio, model_id=tm, defaults=defaults) # if mr already exists, update it with new values ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2486/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2486/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2485
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2485/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2485/comments
https://api.github.com/repos/coleifer/peewee/issues/2485/events
https://github.com/coleifer/peewee/pull/2485
1,051,246,165
PR_kwDOAA7yGM4ua5_k
2,485
Document example usage of object_id_name
{ "login": "pylipp", "id": 10617122, "node_id": "MDQ6VXNlcjEwNjE3MTIy", "avatar_url": "https://avatars.githubusercontent.com/u/10617122?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pylipp", "html_url": "https://github.com/pylipp", "followers_url": "https://api.github.com/users/pylipp/followers", "following_url": "https://api.github.com/users/pylipp/following{/other_user}", "gists_url": "https://api.github.com/users/pylipp/gists{/gist_id}", "starred_url": "https://api.github.com/users/pylipp/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pylipp/subscriptions", "organizations_url": "https://api.github.com/users/pylipp/orgs", "repos_url": "https://api.github.com/users/pylipp/repos", "events_url": "https://api.github.com/users/pylipp/events{/privacy}", "received_events_url": "https://api.github.com/users/pylipp/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I found the text and admonition confusing, even in context. I'll pass for now but will think about whether some improvement to the docs is needed for this situation." ]
"2021-11-11T18:22:13"
"2021-11-11T22:46:25"
"2021-11-11T22:46:24"
NONE
null
In combination with `column_name` defined, an AttributeError is raised otherwise. This behavior was confusing me at first, until I found #1302. The docs (apart from the ForeignKeyField API) don't mention object_id_name
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2485/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2485/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2485", "html_url": "https://github.com/coleifer/peewee/pull/2485", "diff_url": "https://github.com/coleifer/peewee/pull/2485.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2485.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2484
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2484/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2484/comments
https://api.github.com/repos/coleifer/peewee/issues/2484/events
https://github.com/coleifer/peewee/issues/2484
1,046,845,023
I_kwDOAA7yGM4-ZZZf
2,484
Insert doesn't return lastrowid for Postgresql
{ "login": "sebo-b", "id": 51707418, "node_id": "MDQ6VXNlcjUxNzA3NDE4", "avatar_url": "https://avatars.githubusercontent.com/u/51707418?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sebo-b", "html_url": "https://github.com/sebo-b", "followers_url": "https://api.github.com/users/sebo-b/followers", "following_url": "https://api.github.com/users/sebo-b/following{/other_user}", "gists_url": "https://api.github.com/users/sebo-b/gists{/gist_id}", "starred_url": "https://api.github.com/users/sebo-b/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sebo-b/subscriptions", "organizations_url": "https://api.github.com/users/sebo-b/orgs", "repos_url": "https://api.github.com/users/sebo-b/repos", "events_url": "https://api.github.com/users/sebo-b/events{/privacy}", "received_events_url": "https://api.github.com/users/sebo-b/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Postgres/psycopg2 does not provide a lastrowid db-api method -- you need to use RETURNING with postgres. This is handled automatically for you when you use the models APIs, but **not** when you are using the low-level query builder (e.g. `Table`)." ]
"2021-11-07T20:44:01"
"2021-11-08T14:22:51"
"2021-11-08T14:22:51"
NONE
null
Minimal test case: ``` test = Table('test',('id','value'),primary_key='id') test.bind(db) lastRowId = test.insert({test.value: 10}).execute() ``` Actual: lastRowId is None Expected: lastRowId contains generated primary key In Postgresql implementation, for receiving inserted id RETURNING clause is used. The following code works correctly: `lastRowId = test.insert({test.value: 10}).returning(test.id).execute()` however in the current implementation equivalent to this is used: `lastRowId = test.insert({test.value: 10}).returning('id').execute()` Postrgres returns something like this: `Column(name='?column?', type_code=25)` which then is wrapped and looks like this `{'?column?': 'id'}` then `last_insert_id` is trying to access `cursor[0][0]` which raises an error so None is returned
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2484/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2484/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2483
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2483/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2483/comments
https://api.github.com/repos/coleifer/peewee/issues/2483/events
https://github.com/coleifer/peewee/pull/2483
1,044,521,423
PR_kwDOAA7yGM4uFWu9
2,483
Insert link to gevent docs for monkey-patching
{ "login": "pylipp", "id": 10617122, "node_id": "MDQ6VXNlcjEwNjE3MTIy", "avatar_url": "https://avatars.githubusercontent.com/u/10617122?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pylipp", "html_url": "https://github.com/pylipp", "followers_url": "https://api.github.com/users/pylipp/followers", "following_url": "https://api.github.com/users/pylipp/following{/other_user}", "gists_url": "https://api.github.com/users/pylipp/gists{/gist_id}", "starred_url": "https://api.github.com/users/pylipp/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pylipp/subscriptions", "organizations_url": "https://api.github.com/users/pylipp/orgs", "repos_url": "https://api.github.com/users/pylipp/repos", "events_url": "https://api.github.com/users/pylipp/events{/privacy}", "received_events_url": "https://api.github.com/users/pylipp/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Not a fan of links becoming stale?\r\n" ]
"2021-11-04T09:34:33"
"2021-11-04T15:08:27"
"2021-11-04T12:55:15"
NONE
null
null
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2483/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2483/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2483", "html_url": "https://github.com/coleifer/peewee/pull/2483", "diff_url": "https://github.com/coleifer/peewee/pull/2483.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2483.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2482
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2482/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2482/comments
https://api.github.com/repos/coleifer/peewee/issues/2482/events
https://github.com/coleifer/peewee/issues/2482
1,043,929,853
I_kwDOAA7yGM4-ORr9
2,482
Postgres: db.create_tables() throws table already exists error
{ "login": "dhrumilp31", "id": 87986360, "node_id": "MDQ6VXNlcjg3OTg2MzYw", "avatar_url": "https://avatars.githubusercontent.com/u/87986360?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dhrumilp31", "html_url": "https://github.com/dhrumilp31", "followers_url": "https://api.github.com/users/dhrumilp31/followers", "following_url": "https://api.github.com/users/dhrumilp31/following{/other_user}", "gists_url": "https://api.github.com/users/dhrumilp31/gists{/gist_id}", "starred_url": "https://api.github.com/users/dhrumilp31/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/dhrumilp31/subscriptions", "organizations_url": "https://api.github.com/users/dhrumilp31/orgs", "repos_url": "https://api.github.com/users/dhrumilp31/repos", "events_url": "https://api.github.com/users/dhrumilp31/events{/privacy}", "received_events_url": "https://api.github.com/users/dhrumilp31/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "You need to re-bind the models to the postgres database in order for them to be created on the Pg database:\r\n\r\n```python\r\ndb = PostgresqlDatabase('prodigy_db', user='usr', password='pwd', host='host', port=5432)\r\ndb.bind(models.values())\r\ndb.create_tables(list(models.values()), safe=False)\r\n```", "Thanks @coleifer, it worked. Appreciate the quick response. " ]
"2021-11-03T18:22:09"
"2021-11-03T19:21:28"
"2021-11-03T19:13:39"
NONE
null
I am trying to load schema from existing SQLite database and using those definitions, would like to create three new tables in Postgres db. Following is the code that I am using: ```` from peewee import SqliteDatabase, PostgresqlDatabase, Model from playhouse.reflection import generate_models, print_model, print_table_sql # Load existing prodigy database sqlite_db = SqliteDatabase("prodigy.db") models = generate_models(sqlite_db) print('Models: {}'.format(models)) # Connect to a Postgres database and create tables db = PostgresqlDatabase('prodigy_db', user='usr', password='pwd', host='host', port=5432) db.create_tables(list(models.values()), safe=False) print('DB: {}'.format(db.get_tables())) db.close() ```` Schema: ```` dataset id AUTO PK name VARCHAR created INT meta BLOB session INT index(es) name UNIQUE example id AUTO PK input_hash INT task_hash INT content BLOB link id AUTO PK example INT FK: example.id dataset INT FK: dataset.id index(es) example_id dataset_id ```` Which creates the models successfully, but fails to create new tables in Postgres based on those models. ``` Models: {'dataset': <Model: dataset>, 'example': <Model: example>, 'link': <Model: link>} Traceback (most recent call last): File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 3160, in execute_sql cursor.execute(sql, params or ()) sqlite3.OperationalError: table "dataset" already exists During handling of the above exception, another exception occurred: Traceback (most recent call last): File "setup_db.py", line 15, in <module> db.create_tables(list(models.values()), safe=False) File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 3348, in create_tables model.create_table(**options) File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 6721, in create_table cls._schema.create_all(safe, **options) File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 5828, in create_all self.create_table(safe, **table_options) File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 5683, in create_table self.database.execute(self._create_table(safe=safe, **options)) File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 3173, in execute return self.execute_sql(sql, params, commit=commit) File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 3167, in execute_sql self.commit() File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 2933, in __exit__ reraise(new_type, new_type(exc_value, *exc_args), traceback) File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 191, in reraise raise value.with_traceback(tb) File "/home/ubuntu/prodigy/venv/lib/python3.6/site-packages/peewee.py", line 3160, in execute_sql cursor.execute(sql, params or ()) peewee.OperationalError: table "dataset" already exists ``` The error says table 'dataset' already exists, but here is the describe table command on prodigy_db: <img width="709" alt="Screen Shot 2021-11-03 at 2 14 28 PM" src="https://user-images.githubusercontent.com/87986360/140168892-5ff90270-72f9-4599-be92-7ffec2f1f211.png"> Am I missing something here? Thanks in advance.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2482/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2482/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2481
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2481/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2481/comments
https://api.github.com/repos/coleifer/peewee/issues/2481/events
https://github.com/coleifer/peewee/pull/2481
1,039,103,926
PR_kwDOAA7yGM4t07z6
2,481
Fix type in querying.rst
{ "login": "stenci", "id": 5955495, "node_id": "MDQ6VXNlcjU5NTU0OTU=", "avatar_url": "https://avatars.githubusercontent.com/u/5955495?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stenci", "html_url": "https://github.com/stenci", "followers_url": "https://api.github.com/users/stenci/followers", "following_url": "https://api.github.com/users/stenci/following{/other_user}", "gists_url": "https://api.github.com/users/stenci/gists{/gist_id}", "starred_url": "https://api.github.com/users/stenci/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stenci/subscriptions", "organizations_url": "https://api.github.com/users/stenci/orgs", "repos_url": "https://api.github.com/users/stenci/repos", "events_url": "https://api.github.com/users/stenci/events{/privacy}", "received_events_url": "https://api.github.com/users/stenci/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[]
"2021-10-29T02:28:32"
"2021-10-29T11:09:10"
"2021-10-29T11:09:10"
CONTRIBUTOR
null
null
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2481/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2481/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2481", "html_url": "https://github.com/coleifer/peewee/pull/2481", "diff_url": "https://github.com/coleifer/peewee/pull/2481.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2481.patch", "merged_at": "2021-10-29T11:09:10" }
https://api.github.com/repos/coleifer/peewee/issues/2480
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2480/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2480/comments
https://api.github.com/repos/coleifer/peewee/issues/2480/events
https://github.com/coleifer/peewee/issues/2480
1,038,288,678
I_kwDOAA7yGM494wcm
2,480
ReconnectMixin with atomic cause exception
{ "login": "mattangus", "id": 5302804, "node_id": "MDQ6VXNlcjUzMDI4MDQ=", "avatar_url": "https://avatars.githubusercontent.com/u/5302804?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mattangus", "html_url": "https://github.com/mattangus", "followers_url": "https://api.github.com/users/mattangus/followers", "following_url": "https://api.github.com/users/mattangus/following{/other_user}", "gists_url": "https://api.github.com/users/mattangus/gists{/gist_id}", "starred_url": "https://api.github.com/users/mattangus/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/mattangus/subscriptions", "organizations_url": "https://api.github.com/users/mattangus/orgs", "repos_url": "https://api.github.com/users/mattangus/repos", "events_url": "https://api.github.com/users/mattangus/events{/privacy}", "received_events_url": "https://api.github.com/users/mattangus/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "The transaction details are wrapped by the driver, so I can't really comment on why this is the case with pymysql. I'd suggest possibly trying to ping the db before beginning transactions if its possible your connection has been idle for quite some time. You can do this quite easily.\r\n\r\nhttps://pymysql.readthedocs.io/en/latest/modules/connections.html#pymysql.connections.Connection.ping\r\n\r\n```python\r\n\r\ndb = ReconnectMySQLDatabase(...)\r\n\r\ntry:\r\n db.connection().ping(reconnect=False)\r\nexcept:\r\n db.close()\r\n db.connect()\r\n```\r\n\r\nCouple things to consider, though. Why are you leaving open connections sitting idle for a long time? This may indicate a problematic design. If this is a web app, you should probably be using the PooledMySQLDatabase and opening/closing connections each request. If this is something else, however, I'd suggest either a) connect/close when you need to do work, or b) use something like the code above to ensure connection liveness." ]
"2021-10-28T09:23:19"
"2021-10-28T13:01:01"
"2021-10-28T13:01:01"
NONE
null
Using the `ReconnectMixin` and executing a query after a long wait inside an `atomic` block causes an exception. Here is some sample code: ```python class ReconnectMySQLDatabase(ReconnectMixin, pw.MySQLDatabase): """Class to make MySQL automatically reconnect after long periods without querying the database """ pass DB = ReconnectMySQLDatabase(...) # wait a long time tweets = Tweet.select() with DB.atomic(): for t in tweets: t.user = "me" Tweet.bulk_update(tweets, fields=["user"], batch_size=100) ``` This will cause an exception saying that the connection was reset inside a transaction. Shouldn't the transaction initialization cause the reconnect.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2480/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2480/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2479
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2479/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2479/comments
https://api.github.com/repos/coleifer/peewee/issues/2479/events
https://github.com/coleifer/peewee/issues/2479
1,037,145,443
I_kwDOAA7yGM490ZVj
2,479
tests.apsw_ext.TestAPSWExtension tests failing (LockedError) after 10be40b2ea3d76a9 (RETURNING support)
{ "login": "mgorny", "id": 110765, "node_id": "MDQ6VXNlcjExMDc2NQ==", "avatar_url": "https://avatars.githubusercontent.com/u/110765?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mgorny", "html_url": "https://github.com/mgorny", "followers_url": "https://api.github.com/users/mgorny/followers", "following_url": "https://api.github.com/users/mgorny/following{/other_user}", "gists_url": "https://api.github.com/users/mgorny/gists{/gist_id}", "starred_url": "https://api.github.com/users/mgorny/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/mgorny/subscriptions", "organizations_url": "https://api.github.com/users/mgorny/orgs", "repos_url": "https://api.github.com/users/mgorny/repos", "events_url": "https://api.github.com/users/mgorny/events{/privacy}", "received_events_url": "https://api.github.com/users/mgorny/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Thanks for reporting and bisecting, I've got a fix and will push a new release later today.", "Thank you, I can confirm that tests pass on master now." ]
"2021-10-27T08:47:59"
"2021-10-27T14:01:59"
"2021-10-27T12:45:32"
NONE
null
I'm getting 12 test errors with the git master (and 3.14.6) release that are regressions compared to 3.14.4. An example test error looks like the following: ``` ====================================================================== ERROR: test_update_delete (tests.apsw_ext.TestAPSWExtension) ---------------------------------------------------------------------- Traceback (most recent call last): File "/tmp/peewee/tests/base.py", line 246, in tearDown self.database.drop_tables(self.requires, safe=True) File "/tmp/peewee/peewee.py", line 3352, in drop_tables model.drop_table(**kwargs) File "/tmp/peewee/peewee.py", line 6730, in drop_table cls._schema.drop_all(safe, drop_sequences, **options) File "/tmp/peewee/peewee.py", line 5838, in drop_all self.drop_table(safe, **options) File "/tmp/peewee/peewee.py", line 5711, in drop_table self.database.execute(self._drop_table(safe=safe, **options)) File "/tmp/peewee/peewee.py", line 3173, in execute return self.execute_sql(sql, params, commit=commit) File "/tmp/peewee/playhouse/apsw_ext.py", line 122, in execute_sql cursor.execute(sql, params or ()) apsw.LockedError: LockedError: database table is locked ``` Full build and test log: [test.log](https://github.com/coleifer/peewee/files/7424610/test.log) I have reproduced the problem with Python 3.8.12 and 3.10.0. A quick bisect points to the following commit as the first bad one: ``` commit 10be40b2ea3d76a9e410e1e5d0a7b5c1c58c8467 (refs/bisect/bad) Author: Charles Leifer <coleifer@gmail.com> Date: 2021-10-21 15:52:55 +0200 Add provisional support for RETURNING for sqlite_ext/mariadb. Sqlite has kind-of unusual behavior with regards to lastrowid and the rowid in general, so I'm retaining the original behavior for the standard SqliteDatabase, but using SqliteExtDatabase will now support RETURNING for Sqlite 3.35 or newer. Similarly, MariaDBConnectorDatabase will also support RETURNING for MariaDB 10.5 or newer. ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2479/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2479/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2478
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2478/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2478/comments
https://api.github.com/repos/coleifer/peewee/issues/2478/events
https://github.com/coleifer/peewee/issues/2478
1,034,798,619
I_kwDOAA7yGM49rcYb
2,478
Implementation of Modulo operator in documentation
{ "login": "gaspardbb", "id": 40466739, "node_id": "MDQ6VXNlcjQwNDY2NzM5", "avatar_url": "https://avatars.githubusercontent.com/u/40466739?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gaspardbb", "html_url": "https://github.com/gaspardbb", "followers_url": "https://api.github.com/users/gaspardbb/followers", "following_url": "https://api.github.com/users/gaspardbb/following{/other_user}", "gists_url": "https://api.github.com/users/gaspardbb/gists{/gist_id}", "starred_url": "https://api.github.com/users/gaspardbb/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gaspardbb/subscriptions", "organizations_url": "https://api.github.com/users/gaspardbb/orgs", "repos_url": "https://api.github.com/users/gaspardbb/repos", "events_url": "https://api.github.com/users/gaspardbb/events{/privacy}", "received_events_url": "https://api.github.com/users/gaspardbb/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "What database driver are you using this with, out of curiosity? The example works fine with Sqlite but I'm wondering if this is a result of psycopg2 doing its mogrifying client-side.", "This was with Postgre. Indeed, the issue was raised on l. 3144, `cursor.execute(sql, params or ())`." ]
"2021-10-25T07:55:13"
"2021-10-25T17:54:53"
"2021-10-25T14:32:07"
NONE
null
## Problem In the [documentation](https://docs.peewee-orm.com/en/latest/peewee/query_operators.html?#adding-user-defined-operators), it is said that the modulo operator can be implemented with ```python from peewee import * from peewee import Expression # the building block for expressions def mod(lhs, rhs): return Expression(lhs, '%', rhs) ``` This gives a `IndexError: list index out of range`, due to the `%` causing string formatting issues. ## Fix Replacing `%` with `%%`. ```python def mod(lhs, rhs): return Expression(lhs, '%%', rhs) ``` That's not a big issue, but when we know nothing about Python 2 string formatting and SQL, it's a bit unsettling :)
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2478/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2478/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2477
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2477/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2477/comments
https://api.github.com/repos/coleifer/peewee/issues/2477/events
https://github.com/coleifer/peewee/pull/2477
1,034,480,322
PR_kwDOAA7yGM4tl8nk
2,477
removed a consider-using-in pitfall case
{ "login": "NaelsonDouglas", "id": 8750259, "node_id": "MDQ6VXNlcjg3NTAyNTk=", "avatar_url": "https://avatars.githubusercontent.com/u/8750259?v=4", "gravatar_id": "", "url": "https://api.github.com/users/NaelsonDouglas", "html_url": "https://github.com/NaelsonDouglas", "followers_url": "https://api.github.com/users/NaelsonDouglas/followers", "following_url": "https://api.github.com/users/NaelsonDouglas/following{/other_user}", "gists_url": "https://api.github.com/users/NaelsonDouglas/gists{/gist_id}", "starred_url": "https://api.github.com/users/NaelsonDouglas/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/NaelsonDouglas/subscriptions", "organizations_url": "https://api.github.com/users/NaelsonDouglas/orgs", "repos_url": "https://api.github.com/users/NaelsonDouglas/repos", "events_url": "https://api.github.com/users/NaelsonDouglas/events{/privacy}", "received_events_url": "https://api.github.com/users/NaelsonDouglas/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I'll pass." ]
"2021-10-24T18:15:28"
"2021-10-24T20:03:29"
"2021-10-24T20:03:29"
NONE
null
**Problem**: The code was incurring into the consider-using-in pitfall case, as described by Pylint documentation here https://github.com/vald-phoenix/pylint-errors/blob/master/plerr/errors/refactoring/R1714.md **Solution**: Applied adequate refactoring. Only a single line of code was changed
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2477/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2477/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2477", "html_url": "https://github.com/coleifer/peewee/pull/2477", "diff_url": "https://github.com/coleifer/peewee/pull/2477.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2477.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2476
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2476/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2476/comments
https://api.github.com/repos/coleifer/peewee/issues/2476/events
https://github.com/coleifer/peewee/issues/2476
1,033,428,716
I_kwDOAA7yGM49mN7s
2,476
Just warrning when insert_many NULL value into a Not Allow Null field
{ "login": "ThaoPN", "id": 19179219, "node_id": "MDQ6VXNlcjE5MTc5MjE5", "avatar_url": "https://avatars.githubusercontent.com/u/19179219?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ThaoPN", "html_url": "https://github.com/ThaoPN", "followers_url": "https://api.github.com/users/ThaoPN/followers", "following_url": "https://api.github.com/users/ThaoPN/following{/other_user}", "gists_url": "https://api.github.com/users/ThaoPN/gists{/gist_id}", "starred_url": "https://api.github.com/users/ThaoPN/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ThaoPN/subscriptions", "organizations_url": "https://api.github.com/users/ThaoPN/orgs", "repos_url": "https://api.github.com/users/ThaoPN/repos", "events_url": "https://api.github.com/users/ThaoPN/events{/privacy}", "received_events_url": "https://api.github.com/users/ThaoPN/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Your first query \"add_one_null_row()\" is not specifying any value at all for `not_null_field`, while your second query \"add_multiple_null_row()\" is explicitly specifying NULL for `not_null_field`. I am guessing that older MySQL (with its many quirks) handles these situations differently, raising an IntegrityError for the first case, but possibly silently converting the second case to an empty string or something like that?\r\n\r\nIn short: I think this is a MySQL thing. Both the error *and* the warning are generated by MySQL. I suspect there is a MySQL setting you can enable which will cause the IntegrityError to be raised in both cases -- possibly you are on an old MySQL or running with relaxed settings?\r\n\r\nIn any case, I suggest consulting the MySQL docs or checking out StackOverflow.\r\n\r\nWhen I run this on MariaDB 10.4, I get an IntegrityError for *both* queries -- as expected. The first query gives the integrityerror for Column \"text\" cannot be null, and the second query gives the integrityerror for Column \"not_null_field\" cannot be null." ]
"2021-10-22T10:29:27"
"2021-10-22T15:30:58"
"2021-10-22T15:30:19"
NONE
null
I have a table like below ``` CREATE TABLE `tests` ( `id` int(11) NOT NULL AUTO_INCREMENT, `text` varchar(45) NOT NULL, `not_null_field` varchar(45) NOT NULL, PRIMARY KEY (`id`) ) ENGINE=InnoDB AUTO_INCREMENT=21 DEFAULT CHARSET=latin1; ``` and the code I used to insert data: ``` class Test(BaseModel): id = AutoField() text = CharField(null=False) not_null_field = CharField(null=False) class Meta: table_name = "tests" def add_one_null_row(): Test.insert_many([{Test.text: None}]).execute() def add_multiple_null_row(): Test.insert_many( [ {Test.text: "abc", Test.not_null_field: None} ] ).execute() ``` When I `insert_many` with one record like above, I got the error `peewee.IntegrityError: (1048, "Column 'not_null_field' cannot be null")` But when I `insert_many` with more than one record, I just got a warning `.../python3.7/site-packages/pymysql/cursors.py:170: Warning: (1048, "Column 'not_null_field' cannot be null")` and the data be inserted successfully with `not_null_field` is `empty`. The query I got: ``` INSERT INTO `tests` (`text`, `not_null_field`) VALUES ('abc', NULL), ('abc', NULL) ``` I don't know why I just got a warning but an error here, somebody help to correct me.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2476/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2476/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2475
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2475/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2475/comments
https://api.github.com/repos/coleifer/peewee/issues/2475/events
https://github.com/coleifer/peewee/issues/2475
1,033,356,408
I_kwDOAA7yGM49l8R4
2,475
Join with subquery column alias not included in result row
{ "login": "larsch", "id": 15712, "node_id": "MDQ6VXNlcjE1NzEy", "avatar_url": "https://avatars.githubusercontent.com/u/15712?v=4", "gravatar_id": "", "url": "https://api.github.com/users/larsch", "html_url": "https://github.com/larsch", "followers_url": "https://api.github.com/users/larsch/followers", "following_url": "https://api.github.com/users/larsch/following{/other_user}", "gists_url": "https://api.github.com/users/larsch/gists{/gist_id}", "starred_url": "https://api.github.com/users/larsch/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/larsch/subscriptions", "organizations_url": "https://api.github.com/users/larsch/orgs", "repos_url": "https://api.github.com/users/larsch/repos", "events_url": "https://api.github.com/users/larsch/events{/privacy}", "received_events_url": "https://api.github.com/users/larsch/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "The reason for this is that Peewee is attempting to reconstruct the model-graph, so the \"sum\" attribute (which logically belongs to Bar) is placed in `row.bar.sum`.\r\n\r\nHere:\r\n\r\n```python\r\n\r\nsq = (Bar.select(Bar.foo_id, fn.SUM(Bar.value).alias('sum'))\r\n .group_by(Bar.foo_id))\r\nquery = (Foo.select(Foo.name, sq.c.sum.alias('sum'))\r\n .join(sq, on=(Foo.id == sq.c.foo_id))\r\n .order_by(Foo.name))\r\n\r\nfor foo in query:\r\n print(foo.name, foo.bar.sum)\r\n\r\n# BarBaz 13\r\n# BazFoo 13\r\n# FooBar 13\r\n```\r\n\r\nIf you want all attributes simply put on the model being selected, specify `.objects()` in your outer query:\r\n\r\n```python\r\nfor foo in query.objects():\r\n print(foo.name, foo.sum)\r\n```\r\n\r\nDetails here: http://docs.peewee-orm.com/en/latest/peewee/relationships.html#selecting-from-multiple-sources" ]
"2021-10-22T09:10:28"
"2021-10-22T15:16:21"
"2021-10-22T15:16:21"
NONE
null
peewee version: 3.14.4 I can't figure out how to get the aliased column of the subquery out of the result. It is not present in `row.__dict__` and accessing gives `AttributeError: object has no attribute`. Below is a distilled reproduction of the issue. Running the `query.sql()` manually with sqlite3 actually produces the expected values and columns, so the query generator works, but the column is not accessible in the result. ```python from peewee import * database = SqliteDatabase('join_alias.db') class BaseModel(Model): class Meta: database = database class Foo(BaseModel): id = AutoField(unique=True) name = TextField() class Bar(BaseModel): id = AutoField(unique=True) foo_id = IntegerField() value = IntegerField(default=0) with database: database.create_tables([Foo, Bar]) f1 = Foo.create(name='FooBar') f2 = Foo.create(name='BarBaz') f3 = Foo.create(name='BazFoo') b1a = Bar.create(foo_id=f1.id, value=7) b1b = Bar.create(foo_id=f1.id, value=6) b2a = Bar.create(foo_id=f2.id, value=5) b2c = Bar.create(foo_id=f2.id, value=8) b3a = Bar.create(foo_id=f3.id, value=9) b3b = Bar.create(foo_id=f3.id, value=4) subquery = ( Bar .select(Bar.foo_id, fn.SUM(Bar.value).alias('sum')) .group_by(Bar.foo_id)) query = ( Foo .select(Foo.name, subquery.c.sum.alias('sum')) .join(subquery, on=(Foo.id == subquery.c.foo_id))) for row in query.execute(): print(row.name) #print(row.sum) # Foo object has no attribute 'sum' #row.__dict__ doesn't contain sum ``` Attempting to access `row.sum`: ``` Traceback (most recent call last): File "...\join_alias.py", line 47, in <module> print(row.sum) AttributeError: 'Foo' object has no attribute 'sum' ``` The `Foo.__dict__` objects returned, missing the `sum` field are: ``` {'__data__': {'name': 'FooBar'}, '_dirty': set(), '__rel__': {}, 'bar': <Bar: None>} {'__data__': {'name': 'BarBaz'}, '_dirty': set(), '__rel__': {}, 'bar': <Bar: None>} {'__data__': {'name': 'BazFoo'}, '_dirty': set(), '__rel__': {}, 'bar': <Bar: None>} ``` The generated query is: ```sql SELECT t1.name, t2.sum AS sum FROM foo AS t1 INNER JOIN ( SELECT t3.foo_id, SUM(t3.value) AS sum FROM bar AS t3 GROUP BY t3.foo_id ) AS t2 ON (t1.id = t2.foo_id) ``` and when run directly with sqlite3, gives exactly the expected result: |name|sum| |-|-| |FooBar|13| |BarBaz|13| |BazFoo|13|
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2475/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2475/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2474
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2474/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2474/comments
https://api.github.com/repos/coleifer/peewee/issues/2474/events
https://github.com/coleifer/peewee/issues/2474
1,027,995,383
I_kwDOAA7yGM49Rfb3
2,474
Union queries containing order_by
{ "login": "SilvanVerhoeven", "id": 44174681, "node_id": "MDQ6VXNlcjQ0MTc0Njgx", "avatar_url": "https://avatars.githubusercontent.com/u/44174681?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SilvanVerhoeven", "html_url": "https://github.com/SilvanVerhoeven", "followers_url": "https://api.github.com/users/SilvanVerhoeven/followers", "following_url": "https://api.github.com/users/SilvanVerhoeven/following{/other_user}", "gists_url": "https://api.github.com/users/SilvanVerhoeven/gists{/gist_id}", "starred_url": "https://api.github.com/users/SilvanVerhoeven/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SilvanVerhoeven/subscriptions", "organizations_url": "https://api.github.com/users/SilvanVerhoeven/orgs", "repos_url": "https://api.github.com/users/SilvanVerhoeven/repos", "events_url": "https://api.github.com/users/SilvanVerhoeven/events{/privacy}", "received_events_url": "https://api.github.com/users/SilvanVerhoeven/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "There's a little helper you can use to wrap a select query:\r\n\r\n```python\r\nclass A(Base):\r\n a = TextField()\r\n\r\nclass B(Base):\r\n b = TextField()\r\n\r\ndb.create_tables([A, B])\r\n\r\nA.insert_many([('a%d' % i,) for i in range(10)]).execute()\r\nB.insert_many([('b%d' % i,) for i in range(10)]).execute()\r\n\r\nq1 = A.select(A.a).order_by(A.a).limit(5)\r\nq2 = B.select(B.b).order_by(B.b).limit(5)\r\n\r\n# Wrap the queries in the form SELECT * FROM (SELECT ... LIMIT 5)\r\nwq1 = q1.select_from(SQL('*'))\r\nwq2 = q2.select_from(SQL('*'))\r\n\r\n# Create the union - note that it may be necessary to re-bind the\r\n# query to the database. I've fixed this, but it is not contained in the\r\n# current release on PyPI.\r\nu = (wq1 | wq2).bind(db)\r\n\r\nfor row in u:\r\n print(row)\r\n```\r\n\r\nThis prints:\r\n\r\n```\r\n{'a': 'a0'}\r\n{'a': 'a1'}\r\n{'a': 'a2'}\r\n{'a': 'a3'}\r\n{'a': 'a4'}\r\n{'a': 'b0'}\r\n{'a': 'b1'}\r\n{'a': 'b2'}\r\n{'a': 'b3'}\r\n{'a': 'b4'}\r\n```" ]
"2021-10-16T08:47:41"
"2021-10-18T14:06:06"
"2021-10-18T13:37:39"
NONE
null
I receive two queries from separate modules and want to combine the queries using `UNION` (or `UNION ALL`). The following works like a charm in MySQL: ```python def fetch_orders_query(limit, **kwargs): return ModuleA.fetch_orders_query(limit // 2, **kwargs) | ModuleB.fetch_orders_query(limit // 2, **kwargs) # Resulting SQL query ('(SELECT *fields* FROM *table* AS `t1` WHERE (*predicate1*) ORDER BY *field_1* LIMIT *x*) UNION (SELECT *fields* FROM *table* AS `t2` WHERE (*predicate2*) ORDER BY *field_2* LIMIT *y*)') ``` Yet in SQLite, it throws the error `sqlite3.OperationalError: ORDER BY clause should come after UNION not before`. I couldn't find a working solution for peewee, yet I was trying to write a query like proposed in this StackOverflow post: https://stackoverflow.com/questions/10812910/combine-two-statements-with-limits-using-union, without success. Is there a clean way to solve this issue in peewee? Executing both queries separately and combining the results is not a preferred solution, as the database is on an external server and network operations are extremely expensive performance-wise.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2474/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2474/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2473
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2473/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2473/comments
https://api.github.com/repos/coleifer/peewee/issues/2473/events
https://github.com/coleifer/peewee/issues/2473
1,025,972,748
I_kwDOAA7yGM49JxoM
2,473
Support for psycopg3
{ "login": "elderlabs", "id": 30255476, "node_id": "MDQ6VXNlcjMwMjU1NDc2", "avatar_url": "https://avatars.githubusercontent.com/u/30255476?v=4", "gravatar_id": "", "url": "https://api.github.com/users/elderlabs", "html_url": "https://github.com/elderlabs", "followers_url": "https://api.github.com/users/elderlabs/followers", "following_url": "https://api.github.com/users/elderlabs/following{/other_user}", "gists_url": "https://api.github.com/users/elderlabs/gists{/gist_id}", "starred_url": "https://api.github.com/users/elderlabs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/elderlabs/subscriptions", "organizations_url": "https://api.github.com/users/elderlabs/orgs", "repos_url": "https://api.github.com/users/elderlabs/repos", "events_url": "https://api.github.com/users/elderlabs/events{/privacy}", "received_events_url": "https://api.github.com/users/elderlabs/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "There are a number of psycopg3 changes that look to be backwards-incompatible (which is understandable) but will likely be a pain to identify and work around. Parameter binding, especially, looks like it has changed quite a bit.\r\n\r\nIt's definitely on my radar and will be added to Peewee, but I will probably wait a bit to add support. If anyone is interested in the meantime, feel free to implement this in a standalone/extension module.", "Hello\r\n\r\nI looked at your extension but I wondering, can I use this in conjuction with PostgresqlExtDatabase?\r\n\r\nAny plans on built-in support for psycopg3? Considering is also adding it in their next release and it is a stable release..", "Are there any specific features of `PostgresqlExtDatabase` you were intending to use - e.g. jsonb, arrays? I'll have to spend some time seeing what works and what doesn't, since we use some helpers from `psycopg2.extras` (e.g. `Json`).", "I'm going through all my raw SQL code currently, but if I'm not mistaken I will have some jsonb-specifics.\r\n\r\nI will use psycopg2 for now though, but it would be great to see support for 3.", "No, I agree with you, things are going to be moving towards psycopg3 in the coming years so I'll want to be ready for that. Thanks for the ping on this, I'll be aiming to getting parity for psycopg3 with the current PostgresqlExtDatabase. No definitive timeline but it's a priority for me." ]
"2021-10-14T05:38:37"
"2023-03-30T15:35:45"
"2021-10-14T13:08:32"
NONE
null
Preemptive ticket here. The maintainer of psycopg2 has busied themself with its next iteration, [psycopg3](https://github.com/psycopg/psycopg). Are there plans to add support for the project; is there already native support by design? This ticket's more to see if the project's on your radar as it has officially released, and support for psycopg2 might be in decline fairly soon. Thanks.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2473/reactions", "total_count": 5, "+1": 5, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2473/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2472
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2472/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2472/comments
https://api.github.com/repos/coleifer/peewee/issues/2472/events
https://github.com/coleifer/peewee/issues/2472
1,025,820,130
I_kwDOAA7yGM49JMXi
2,472
Model.bulk_create() ignores id of instances
{ "login": "balloonz", "id": 6794239, "node_id": "MDQ6VXNlcjY3OTQyMzk=", "avatar_url": "https://avatars.githubusercontent.com/u/6794239?v=4", "gravatar_id": "", "url": "https://api.github.com/users/balloonz", "html_url": "https://github.com/balloonz", "followers_url": "https://api.github.com/users/balloonz/followers", "following_url": "https://api.github.com/users/balloonz/following{/other_user}", "gists_url": "https://api.github.com/users/balloonz/gists{/gist_id}", "starred_url": "https://api.github.com/users/balloonz/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/balloonz/subscriptions", "organizations_url": "https://api.github.com/users/balloonz/orgs", "repos_url": "https://api.github.com/users/balloonz/repos", "events_url": "https://api.github.com/users/balloonz/events{/privacy}", "received_events_url": "https://api.github.com/users/balloonz/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Why are you manually specifying the ID for an auto-incrementing (opaque) primary key? The `bulk_create()` API is a shorthand for generating an `insert_many()`. Just use `insert_many()` if you really want to explicitly set these, I suppose.", "I was importing some data into the database which involve relational data.\r\nI thought `bulk_create()` as an alternative to `insert_many()` may also work.\r\nIf it is intended behavior, I will stick to `insert_many()`. \r\nThanks for the quick response. \r\n\r\n", "If you want, you can disable the primary keys `auto_increment` attribute during bulk loading, but (especially if you're just creating the models to insert the data only) insert many is probably a fine choice." ]
"2021-10-14T00:50:11"
"2021-10-14T02:31:06"
"2021-10-14T00:53:01"
NONE
null
It seems that Model.create() method respects id while Model.bulk_create ignores id of instances. my model class: ``` class Book(Model): title = CharField(default="") author = CharField(default="") created_date = DateTimeField(default=datetime.now) class Meta: database = SqliteDatabase(":memory:") ``` The following code prints [(1, 'A'), (2, 'B'), (4, 'D'), (5, 'C')]. ``` with db.atomic(): Book.create(id=1, title="A") Book.create(id=2, title="B") Book.create(id=5, title="C") Book.create(id=4, title="D") print([(book.id, book.title) for book in Book.select()]) ``` The following code prints: [(1, 'A'), (2, 'B'), (5, 'C'), (4, 'D')] [(1, 'A'), (2, 'B'), (3, 'C'), (4, 'D')] ``` book_items = [ Book(id=1, title="A"), Book(id=2, title="B"), Book(id=5, title="C"), Book(id=4, title="D"), ] with db.atomic(): Book.bulk_create(book_items, batch_size=50) print([(book.id, book.title) for book in book_items]) print([(book.id, book.title) for book in Book.select()]) ``` Also, the unsaved instances are kept unchanged and therefore inconsistent with the instances returned by Model.select().
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2472/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2472/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2471
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2471/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2471/comments
https://api.github.com/repos/coleifer/peewee/issues/2471/events
https://github.com/coleifer/peewee/issues/2471
1,024,681,147
I_kwDOAA7yGM49E2S7
2,471
Mariadb
{ "login": "chrisemke", "id": 63082743, "node_id": "MDQ6VXNlcjYzMDgyNzQz", "avatar_url": "https://avatars.githubusercontent.com/u/63082743?v=4", "gravatar_id": "", "url": "https://api.github.com/users/chrisemke", "html_url": "https://github.com/chrisemke", "followers_url": "https://api.github.com/users/chrisemke/followers", "following_url": "https://api.github.com/users/chrisemke/following{/other_user}", "gists_url": "https://api.github.com/users/chrisemke/gists{/gist_id}", "starred_url": "https://api.github.com/users/chrisemke/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/chrisemke/subscriptions", "organizations_url": "https://api.github.com/users/chrisemke/orgs", "repos_url": "https://api.github.com/users/chrisemke/repos", "events_url": "https://api.github.com/users/chrisemke/events{/privacy}", "received_events_url": "https://api.github.com/users/chrisemke/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Added basic support in: `playhouse.mysql_ext.MariaDBConnectorDatabase`" ]
"2021-10-13T02:00:36"
"2021-10-13T13:12:14"
"2021-10-13T13:09:23"
NONE
null
Analyze the possibility of adding support to the mariadb database using the [official connector](https://pypi.org/project/mariadb/)
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2471/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2471/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2470
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2470/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2470/comments
https://api.github.com/repos/coleifer/peewee/issues/2470/events
https://github.com/coleifer/peewee/issues/2470
1,023,317,868
I_kwDOAA7yGM48_pds
2,470
Potential "list index out of range" in FunctionTable SQLite Extension
{ "login": "ecpost", "id": 16603401, "node_id": "MDQ6VXNlcjE2NjAzNDAx", "avatar_url": "https://avatars.githubusercontent.com/u/16603401?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ecpost", "html_url": "https://github.com/ecpost", "followers_url": "https://api.github.com/users/ecpost/followers", "following_url": "https://api.github.com/users/ecpost/following{/other_user}", "gists_url": "https://api.github.com/users/ecpost/gists{/gist_id}", "starred_url": "https://api.github.com/users/ecpost/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ecpost/subscriptions", "organizations_url": "https://api.github.com/users/ecpost/orgs", "repos_url": "https://api.github.com/users/ecpost/repos", "events_url": "https://api.github.com/users/ecpost/events{/privacy}", "received_events_url": "https://api.github.com/users/ecpost/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "The columns you can filter on are added as HIDDEN columns *after* the exported columns. So you would have:\r\n\r\n```\r\ntable func: series(<start=0>, <stop=None>, <step=1>)\r\n\r\nclass Series(TableFunction):\r\n params = ['start', 'stop', 'step']\r\n columns = ['value']\r\n\r\n...\r\n\r\nCREATE VIRTUAL TABLE series(start, stop, step, value HIDDEN)\r\n```\r\n\r\nWhen looking up the constrained column, we subtract from it the number of \"non-hidden\" columns to get the index of the hidden/parameterized column.\r\n\r\nI've added a check that this value does not drop below zero: 56fe72362a65199eb75bf25684eb290c57e3a37f", "Ah, ok. So that line is only meant to apply to those hidden columns, which would only be the non-negative cases. Thanks for the quick response and fix!" ]
"2021-10-12T03:27:06"
"2021-10-12T18:05:32"
"2021-10-12T13:05:08"
NONE
null
Hi, I ran into an issue with this line in FunctionTable's pwBestIndex() function in _sqlite_ext.pyx: ``` columns.append(table_func_cls.params[pConstraint.iColumn - table_func_cls._ncols]) ``` The value of iColumn is some number less than _ncols, so the index into the list is always negative. If it's _too_ negative, it goes past the start of params, causing an "IndexError: list index out of range" exception. If you need some sample code demonstrating the problem, I'll put something together. I'm happy to work on a fix and submit a pull request, but I'm having trouble figuring out the intended logic. (I'm wondering why column positions are involved in indexing into the params list. Although I do see that params are relevant here and in the filter step.) Maybe it was partly to deal with the constraints of SQLite's virtual table API. But anyway, if I can help at all, let me know.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2470/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2470/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2469
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2469/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2469/comments
https://api.github.com/repos/coleifer/peewee/issues/2469/events
https://github.com/coleifer/peewee/issues/2469
1,015,445,089
I_kwDOAA7yGM48hnZh
2,469
UPDATE queries don't work because Value objects sqlify as ?
{ "login": "Gaming32", "id": 42721887, "node_id": "MDQ6VXNlcjQyNzIxODg3", "avatar_url": "https://avatars.githubusercontent.com/u/42721887?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Gaming32", "html_url": "https://github.com/Gaming32", "followers_url": "https://api.github.com/users/Gaming32/followers", "following_url": "https://api.github.com/users/Gaming32/following{/other_user}", "gists_url": "https://api.github.com/users/Gaming32/gists{/gist_id}", "starred_url": "https://api.github.com/users/Gaming32/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Gaming32/subscriptions", "organizations_url": "https://api.github.com/users/Gaming32/orgs", "repos_url": "https://api.github.com/users/Gaming32/repos", "events_url": "https://api.github.com/users/Gaming32/events{/privacy}", "received_events_url": "https://api.github.com/users/Gaming32/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Detailed walkthrough of `Context.value`:\r\n```python\r\n def value(self, value, converter=None, add_param=True):\r\n # add_param = True\r\n # converter = <bound method Field.db_value of <CharField: User.token>>\r\n # value = 'ba8a310b920ece30e4e28e6f67d40a53'\r\n if converter:\r\n value = converter(value)\r\n elif converter is None and self.state.converter:\r\n # Explicitly check for None so that \"False\" can be used to signify\r\n # that no conversion should be applied.\r\n value = self.state.converter(value)\r\n\r\n # value is still 'ba8a310b920ece30e4e28e6f67d40a53'\r\n if isinstance(value, Node):\r\n with self(converter=None):\r\n return self.sql(value)\r\n elif is_model(value):\r\n # Under certain circumstances, we could end-up treating a model-\r\n # class itself as a value. This check ensures that we drop the\r\n # table alias into the query instead of trying to parameterize a\r\n # model (for instance, passing a model as a function argument).\r\n with self.scope_column():\r\n return self.sql(value)\r\n\r\n self._values.append(value) # Values obtains the value 'ba8a310b920ece30e4e28e6f67d40a53'\r\n return self.literal(self.state.param or '?') if add_param else self # self.literal('?') is called, as add_param is True\r\n```", "The ? is the param placeholder. Parameters are bound the placeholders. This is to avoid sql injections.", "It prevents the database from being updated though", "How do I put a string into my database then?", "Do you mean to imply that the fundamental operation of updating data is completely broken?", "```python\r\ndb = SqliteDatabase(':memory:')\r\n\r\nclass User(Model):\r\n token = TextField()\r\n class Meta:\r\n database = db\r\n\r\nUser.create_table()\r\n\r\nuser = User.create(token='')\r\nuser.save()\r\n\r\nuser.token = 'new-token'\r\nuser.save()\r\n\r\nuser_db = User.get(User.id == user.id)\r\nprint(user_db.id, user_db.token)\r\n\r\nUser.update(token='new2').where(User.id == user.id).execute()\r\n\r\nuser_db = User.get(User.id == user.id)\r\nprint(user_db.id, user_db.token)\r\n```\r\n\r\nPrints\r\n\r\n```\r\n1 new-token\r\n1 new2\r\n```", "Why was my comment deleted? I need legitimate help here. " ]
"2021-10-04T17:35:00"
"2021-10-04T21:07:59"
"2021-10-04T17:56:02"
NONE
null
Whenever I update a user's token using one of the two following pieces of code: ```python # 1 user.token = token user.save(only=['token']) # 2 User.update({User.token: token}).where(User.unique_id == user_id).execute() ``` The following query is executed: ```sql UPDATE "user" SET "token" = ? WHERE ("user"."unique_id" = ?) ``` All my arguments are dropped! As a result, the database is not updated. I've found that this appears to be because the `add_param` argument in `Context.value` is `True` by default. However, simply changing this to `False` does not work either (I didn't think it would, but I still tried it so I could say I did).
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2469/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2469/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2468
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2468/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2468/comments
https://api.github.com/repos/coleifer/peewee/issues/2468/events
https://github.com/coleifer/peewee/issues/2468
1,014,019,009
I_kwDOAA7yGM48cLPB
2,468
Bulk update of all null DateTimeFields fails
{ "login": "milesgranger", "id": 13764397, "node_id": "MDQ6VXNlcjEzNzY0Mzk3", "avatar_url": "https://avatars.githubusercontent.com/u/13764397?v=4", "gravatar_id": "", "url": "https://api.github.com/users/milesgranger", "html_url": "https://github.com/milesgranger", "followers_url": "https://api.github.com/users/milesgranger/followers", "following_url": "https://api.github.com/users/milesgranger/following{/other_user}", "gists_url": "https://api.github.com/users/milesgranger/gists{/gist_id}", "starred_url": "https://api.github.com/users/milesgranger/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/milesgranger/subscriptions", "organizations_url": "https://api.github.com/users/milesgranger/orgs", "repos_url": "https://api.github.com/users/milesgranger/repos", "events_url": "https://api.github.com/users/milesgranger/events{/privacy}", "received_events_url": "https://api.github.com/users/milesgranger/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "This is a Postgres limitation - for some reason Postgres does not behave well with certain data-types in these expressions. Related issues:\r\n\r\n* #2457\r\n* #2308\r\n* #2292\r\n* https://dba.stackexchange.com/questions/228046/why-do-i-need-to-cast-null-to-column-type/228058#228058\r\n\r\nI cannot really offer a solution at this time, other than to suggest using individual updates / calls to `save()`. Doing updates in a loop inside a transaction may be just as fast (if not faster) than the `bulk_update` anyways.", "Thank you, appreciate the information! :pray: " ]
"2021-10-02T11:19:46"
"2021-10-02T14:39:29"
"2021-10-02T13:48:56"
NONE
null
Hello there! :wave: I believe I've encountered a bug in an otherwise terrific library. Summary --- When performing a bulk update of models, where the model has a nullable `DateTimeField` and every model in the list to `bulk_update` has this field set to `None` will result in a an error complaining about it being of type timestamp but expression is of type text. Example --- ```python def test_bulk_update_of_all_null_datetime_fields(postgresdb): class Foobar(peewee.Model): class Meta: database = db col1 = peewee.IntegerField(primary_key=True) col2 = peewee.DateTimeField(null=True,) Foobar.create_table() models = [Foobar(col1=i, col2=datetime.now()) for i in range(2)] Foobar.bulk_create(models) models = [Foobar(col1=i, col2=None) for i in range(2)] try: Foobar.bulk_update(models, fields=["col2"]) except: db.rollback() # Updating a single model in the list to non-null will work fine. models[0].col2 = datetime.now() Foobar.bulk_update(models, fields=["col2"]) ``` This is the SQL all nulls generate ```sql UPDATE "foobar" SET "col2" = CASE "foobar"."col1" WHEN 0 THEN NULL WHEN 1 THEN NULL END WHERE ("foobar"."col1" IN (0, 1)) ``` Which will result in ```bash 2021-10-02 11:07:19.084 UTC [60] ERROR: column "col2" is of type timestamp without time zone but expression is of type text at character 30 2021-10-02 11:07:19.084 UTC [60] HINT: You will need to rewrite or cast the expression. 2021-10-02 11:07:19.084 UTC [60] STATEMENT: UPDATE "foobar" SET "col2" = CASE "foobar"."col1" WHEN 0 THEN NULL WHEN 1 THEN NULL END WHERE ("foobar"."col1" IN (0, 1)) ``` manually modifying the query to cast the nulls to timestamps will work ```sql UPDATE "foobar" SET "col2" = CASE "foobar"."col1" WHEN 0 THEN NULL::timestamp WHEN 1 THEN NULL::timestamp END WHERE ("foobar"."col1" IN (0, 1)) ``` --- System ubuntu 18.04 peewee version 3.14.4 Python3.7 Postgresql 12
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2468/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2468/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2467
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2467/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2467/comments
https://api.github.com/repos/coleifer/peewee/issues/2467/events
https://github.com/coleifer/peewee/issues/2467
1,009,708,412
I_kwDOAA7yGM48Lu18
2,467
Question: Why Rollback do not work for Postgres?
{ "login": "iamthen0ise", "id": 1730030, "node_id": "MDQ6VXNlcjE3MzAwMzA=", "avatar_url": "https://avatars.githubusercontent.com/u/1730030?v=4", "gravatar_id": "", "url": "https://api.github.com/users/iamthen0ise", "html_url": "https://github.com/iamthen0ise", "followers_url": "https://api.github.com/users/iamthen0ise/followers", "following_url": "https://api.github.com/users/iamthen0ise/following{/other_user}", "gists_url": "https://api.github.com/users/iamthen0ise/gists{/gist_id}", "starred_url": "https://api.github.com/users/iamthen0ise/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/iamthen0ise/subscriptions", "organizations_url": "https://api.github.com/users/iamthen0ise/orgs", "repos_url": "https://api.github.com/users/iamthen0ise/repos", "events_url": "https://api.github.com/users/iamthen0ise/events{/privacy}", "received_events_url": "https://api.github.com/users/iamthen0ise/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "You are using the transaction bit incorrectly I would say. Here's how I would do it:\r\n\r\n```python\r\n\r\nfor item in insert_items:\r\n with db.atomic() as txn:\r\n try:\r\n model.create(**item)\r\n except IntegrityError:\r\n txn.rollback()\r\n else:\r\n txn.commit()\r\n```", "Still getting `current transaction is aborted, commands ignored until end of transaction block` using described method.", "I'm sure you have some bug in your code. This works just fine:\r\n\r\n```python\r\ndb = PostgresqlDatabase('peewee_test')\r\n\r\nclass Reg(Model):\r\n key = CharField(unique=True)\r\n class Meta:\r\n database = db\r\n\r\ndb.create_tables([Reg])\r\nfor k in 'abcadaec':\r\n with db.atomic() as txn:\r\n try:\r\n Reg.create(key=k)\r\n except IntegrityError:\r\n txn.rollback()\r\n else:\r\n txn.commit()\r\n\r\nassert [r.key for r in Reg.select().order_by(Reg.key)] == list('abcde')\r\ndb.drop_tables([Reg])\r\n\r\n```", "Hmm, it works for soft inserts like `model.get_or_create()`, but causes block for strict `create` operation.\r\n\r\nDon't know if it good by design, but it helps me and all working like magic now. Will check the sources for reference.\r\n\r\nThank you for your help! :) " ]
"2021-09-28T12:47:50"
"2021-09-28T14:07:05"
"2021-09-28T13:14:53"
NONE
null
Hi! Explicit rollback do not work. Moreover, it block all the transactions until the program end. Peewee Versions is: 3.14.4 Code: ```python from playhouse.pool import PooledPostgresqlExtDatabase from playhouse.postgres_ext import ( Model, CharField, ) def get_db(): return PooledPostgresqlExtDatabase( DB_NAME, user=DB_USER, password=DB_PASS, host=DB_HOST, port=DB_PORT, max_connections=32, autocommit=False, autorollback=False, stale_timeout=300, ) class ItemsModel(Model): name = CharField() model = ItemsModel() insert_items = [] for item in insert_items: db = get_db() with db.transaction(): try: model.create(**item) db.commit() except Exception as e: db.rollback() ``` Any help is highly appreciated! Thank you!
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2467/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2467/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2466
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2466/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2466/comments
https://api.github.com/repos/coleifer/peewee/issues/2466/events
https://github.com/coleifer/peewee/issues/2466
1,006,237,428
I_kwDOAA7yGM47-fb0
2,466
Peewee cleanups all Model.dirty_fields with only params
{ "login": "penja", "id": 2717390, "node_id": "MDQ6VXNlcjI3MTczOTA=", "avatar_url": "https://avatars.githubusercontent.com/u/2717390?v=4", "gravatar_id": "", "url": "https://api.github.com/users/penja", "html_url": "https://github.com/penja", "followers_url": "https://api.github.com/users/penja/followers", "following_url": "https://api.github.com/users/penja/following{/other_user}", "gists_url": "https://api.github.com/users/penja/gists{/gist_id}", "starred_url": "https://api.github.com/users/penja/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/penja/subscriptions", "organizations_url": "https://api.github.com/users/penja/orgs", "repos_url": "https://api.github.com/users/penja/repos", "events_url": "https://api.github.com/users/penja/events{/privacy}", "received_events_url": "https://api.github.com/users/penja/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Thanks, this should be fixed now.", "@coleifer Thank you for so hot fix." ]
"2021-09-24T08:54:36"
"2021-09-24T13:03:01"
"2021-09-24T12:44:38"
NONE
null
For me was unexpected behavior with cleanup model dirty fields if we save model with `only` params. https://github.com/coleifer/peewee/blob/master/peewee.py#L6600-L6601 In my opinion, The saved field should be just removed from dirty fields. Please see invalid behavior in the following code snippet. ``` import peewee from peewee import * import logging import sys log = logging.getLogger() log.setLevel(logging.DEBUG) handler = logging.StreamHandler(sys.stdout) log.addHandler(handler) print(peewee.__version__) db = SqliteDatabase(":memory:") class Example(Model): class Meta: database = db name = CharField() description = CharField(null=True) db.create_tables([Example]) model = Example.create(name="foo", description="bar") model.name = "bazz" model.description = "example bar" model.save(only=[Example.name]) #some other business logic if model.dirty_fields: model.save() model = Example.get_by_id(model) print(model.name, model.description) assert model.name == "bazz" assert model.description == "example bar" ``` Output ``` 3.14.4 ('CREATE TABLE IF NOT EXISTS "example" ("id" INTEGER NOT NULL PRIMARY KEY, "name" VARCHAR(255) NOT NULL, "description" VARCHAR(255))', []) ('INSERT INTO "example" ("name", "description") VALUES (?, ?)', ['foo', 'bar']) ('UPDATE "example" SET "name" = ? WHERE ("example"."id" = ?)', ['bazz', 1]) ('SELECT "t1"."id", "t1"."name", "t1"."description" FROM "example" AS "t1" WHERE ("t1"."id" = ?) LIMIT ? OFFSET ?', [1, 1, 0]) bazz bar Traceback (most recent call last): File "dummy_sqlite.py", line 36, in <module> assert model.description == "example bar" AssertionError ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2466/reactions", "total_count": 2, "+1": 2, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2466/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2465
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2465/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2465/comments
https://api.github.com/repos/coleifer/peewee/issues/2465/events
https://github.com/coleifer/peewee/issues/2465
1,000,010,798
I_kwDOAA7yGM47mvQu
2,465
UNSIGNED BIGINT AUTO_INCREMENT support in create_tables?
{ "login": "sherrrrr", "id": 31510228, "node_id": "MDQ6VXNlcjMxNTEwMjI4", "avatar_url": "https://avatars.githubusercontent.com/u/31510228?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sherrrrr", "html_url": "https://github.com/sherrrrr", "followers_url": "https://api.github.com/users/sherrrrr/followers", "following_url": "https://api.github.com/users/sherrrrr/following{/other_user}", "gists_url": "https://api.github.com/users/sherrrrr/gists{/gist_id}", "starred_url": "https://api.github.com/users/sherrrrr/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sherrrrr/subscriptions", "organizations_url": "https://api.github.com/users/sherrrrr/orgs", "repos_url": "https://api.github.com/users/sherrrrr/repos", "events_url": "https://api.github.com/users/sherrrrr/events{/privacy}", "received_events_url": "https://api.github.com/users/sherrrrr/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "You overrode the pk, now just subclass `ForeignKeyField` and override the field_type property:\r\n\r\n```python\r\nclass UnsignedBigAutoField(BigAutoField):\r\n field_type = 'BIGINT UNSIGNED AUTO_INCREMENT'\r\n\r\nclass UnsignedForeignKeyField(ForeignKeyField):\r\n field_type = 'BIGINT UNSIGNED'\r\n\r\nclass User(Base):\r\n id = UnsignedBigAutoField()\r\n\r\nclass Tweet(Base):\r\n id = UnsignedBigAutoField()\r\n user = UnsignedForeignKeyField(User)\r\n```" ]
"2021-09-18T11:28:21"
"2021-09-18T18:31:02"
"2021-09-18T18:31:02"
NONE
null
I am using mysql and use UNSIGNED BIGINT as primary key. According to as [issue542](https://github.com/coleifer/peewee/issues/542), I try to use BIGINT UNSIGNED like this ```python class UnsignedBigAutoField(BigAutoField): field_type = 'BIGINT UNSIGNED' ``` and I try to use `ForeignKeyField` in models. ``` class User(Model): id = UnsignedBigAutoField(primary_key=True) class Tweet(Model): id = UnsignedBigAutoField(primary_key=True) user = ForeignKeyField(User, backref='tweets') ``` I want to use peewee to create tables, later I realize that inherite `AutoField` with `auto_increment = True` does't affect sql, there is no AUTO_INCREMENT in create table sql. What's more, when try to create `Tweet`, user_id's type is BIGINT without UNSIGNED beacuse of ForeignKeyField's [field_type](https://github.com/coleifer/peewee/blob/master/peewee.py#L5213). so I write this ```python class UnsignedBigAutoField(IntegerField): auto_increment = True field_type = 'BIGINT UNSIGNED AUTO_INCREMENT' def __init__(self, *args, **kwargs): if kwargs.get('primary_key') is False: raise ValueError('%s must always be a primary key.' % type(self)) kwargs['primary_key'] = True super(UnsignedBigAutoField, self).__init__(*args, **kwargs) ``` and this would cause `user_id` in Tweet have `AUTO_INCREMENT`. So is there a solution to use UNSIGNED BIGINT as primary key and foreign key when I want to create table using peewee?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2465/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2465/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2464
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2464/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2464/comments
https://api.github.com/repos/coleifer/peewee/issues/2464/events
https://github.com/coleifer/peewee/issues/2464
997,743,671
I_kwDOAA7yGM47eFw3
2,464
Why store _database in ThreadSafeDatabaseMetadata?
{ "login": "sherrrrr", "id": 31510228, "node_id": "MDQ6VXNlcjMxNTEwMjI4", "avatar_url": "https://avatars.githubusercontent.com/u/31510228?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sherrrrr", "html_url": "https://github.com/sherrrrr", "followers_url": "https://api.github.com/users/sherrrrr/followers", "following_url": "https://api.github.com/users/sherrrrr/following{/other_user}", "gists_url": "https://api.github.com/users/sherrrrr/gists{/gist_id}", "starred_url": "https://api.github.com/users/sherrrrr/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sherrrrr/subscriptions", "organizations_url": "https://api.github.com/users/sherrrrr/orgs", "repos_url": "https://api.github.com/users/sherrrrr/repos", "events_url": "https://api.github.com/users/sherrrrr/events{/privacy}", "received_events_url": "https://api.github.com/users/sherrrrr/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ ">At first I thought every request is default to master, then I found that if one request switch to slave, _database in User became slave, then if gevent switch to another greenlet with a new request, then User's database is default to slave.\r\n\r\nYeah, I think you're right actually, and the implementation needs a change. Originally the idea I had was to just make sure that binding was a thread-safe operation, and that within a running thread, a different thread cannot alter the db. But, as you indicate, it also makes sense to me to preserve the \"default\" db.\r\n\r\nYou can use your patch or you can try out this commit: 00498bfc073569a63cb30fa43d706fdcc3e14dc0" ]
"2021-09-16T04:00:42"
"2021-09-16T13:29:17"
"2021-09-16T13:29:02"
NONE
null
```python class ThreadSafeDatabaseMetadata(Metadata): """ Metadata class to allow swapping database at run-time in a multi-threaded application. To use: class Base(Model): class Meta: model_metadata_class = ThreadSafeDatabaseMetadata """ def __init__(self, *args, **kwargs): # The database attribute is stored in a thread-local. self._database = None self._local = threading.local() super(ThreadSafeDatabaseMetadata, self).__init__(*args, **kwargs) def _get_db(self): return getattr(self._local, 'database', self._database) def _set_db(self, db): self._local.database = self._database = db database = property(_get_db, _set_db) ``` I'm using peewee with gevent, and try to use `ThreadSafeDatabaseMetadata` to make sure every request can set its db to master/slave. Here is how I define my model. ```python class User(Model): id = BigAutoField(primary_key=True) class Meta: database = master model_metadata_class = ThreadSafeDatabaseMetadata ``` At first I thought every request is default to master, then I found that if one request switch to slave, `_database` in `User` became slave, then if gevent switch to another greenlet with a new request, then User's database is default to slave. Of course this can be solved by every request first call something like `master.bind([ALLMODELS])`, but if that so, why is `_database` necessary? Can I remove `_database` in `ThreadSafeDatabaseMetadata` and use like this? ```python class ThreadSafeDatabaseMetadata(Metadata): def __init__(self, *args, **kwargs): # database attribute is stored in a thread-local. self._local = threading.local() super(ThreadSafeDatabaseMetadata, self).__init__(*args, **kwargs) def _get_db(self): return getattr(self._local, 'database', master) # default use master def _set_db(self, db): self._local.database = db database = property(_get_db, _set_db) ``` And I think `ThreadSafeDatabaseMetadata`'s behavior is a bit confusing, and didn't find it in the document.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2464/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2464/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2463
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2463/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2463/comments
https://api.github.com/repos/coleifer/peewee/issues/2463/events
https://github.com/coleifer/peewee/issues/2463
994,155,602
MDU6SXNzdWU5OTQxNTU2MDI=
2,463
Help on select() and get() methods
{ "login": "Sonotoki-da", "id": 34913890, "node_id": "MDQ6VXNlcjM0OTEzODkw", "avatar_url": "https://avatars.githubusercontent.com/u/34913890?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Sonotoki-da", "html_url": "https://github.com/Sonotoki-da", "followers_url": "https://api.github.com/users/Sonotoki-da/followers", "following_url": "https://api.github.com/users/Sonotoki-da/following{/other_user}", "gists_url": "https://api.github.com/users/Sonotoki-da/gists{/gist_id}", "starred_url": "https://api.github.com/users/Sonotoki-da/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Sonotoki-da/subscriptions", "organizations_url": "https://api.github.com/users/Sonotoki-da/orgs", "repos_url": "https://api.github.com/users/Sonotoki-da/repos", "events_url": "https://api.github.com/users/Sonotoki-da/events{/privacy}", "received_events_url": "https://api.github.com/users/Sonotoki-da/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "The problem is here. Calling `query.get()` returns a model instance (if one exists) or raises a `DoesNotExist` if not. You are comparing the return value to `is True` -- and while the result (a model instance) is \"truthy\", it is *not* `True` - it is a model instance.\r\n\r\n```python\r\nif q.get_user_status(user_id=123456) is True:\r\n print(\"Subscribed\")\r\nelse:\r\n print(\"Not subscribed yet\")\r\n```\r\n\r\nYou should instead write:\r\n\r\n```python\r\ntry:\r\n user = q.get_user_status(user_id=123456)\r\nexcept q.TgUser.DoesNotExist:\r\n # user with id=123456 does not exist\r\n print('user does not exist')\r\nelse:\r\n if user.subscribed:\r\n print('subscribed')\r\n else:\r\n print('not subscribed')\r\n```", "omg... Thank you very much! > < It's working now!\r\nAnd sorry for taking your time to this kind of question :') I'll study further!" ]
"2021-09-12T13:11:21"
"2021-09-12T15:23:16"
"2021-09-12T13:53:01"
NONE
null
I've already searched for solutions in the Issues, but couldn't find an answer, nor understand properly :') OS: Windows Python: 3.6.8, 3.9.6 IDE: VSCode Tested purposely in 3.6.8 (saw the peewee is developed with that version). Thought maybe I can get the code work. The ```TgUser.select(TgUser.subscribed).where(TgUser.user_id == id)``` line is just doesn't work properly. It should be returning True (as in the db, the subscribed record is set to True for that user) Here's my rough code: `queries.py` ```python from datetime import date import peewee as p db = p.SqliteDatabase('database.db') class TgUser(p.Model): user_id = p.IntegerField(unique=True) username = p.CharField(null=True) first_name = p.CharField() last_name = p.CharField(null=True) msg_count = p.IntegerField(default=0) subscribed = p.BooleanField() joined = p.DateField() bot_status = p.CharField(default='member') class Meta: database = db def create_db(): db.create_tables([TgUser]) return db def register_user(values: dict, joined: date, subscribed: bool) -> bool: user = {} for x in ['user_id', 'username', 'first_name', 'last_name']: user.update({x: values.get(x, None)}) data, exist = TgUser.get_or_create(**user, joined=joined, subscribed=subscribed) return exist def get_user_status(user_id: int) -> bool: q = TgUser.select(TgUser.subscribed).where(TgUser.user_id == user_id) return q.get() ``` `test.py` ```python import queries as q from datetime import datetime db = q.create_db() values = { 'user_id': 123456, 'username': '@some_user', 'first_name': 'SomeUser', } q.register_user(values, joined=datetime.now(), subscribed=True) if q.get_user_status(user_id=123456) is True: print("Subscribed") else: print("Not subscribed yet") ``` Out: `Not subscribed yet` Please help, any suggestion is appreciated! I just wanted to stop writing raw SQL queries
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2463/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2463/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2462
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2462/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2462/comments
https://api.github.com/repos/coleifer/peewee/issues/2462/events
https://github.com/coleifer/peewee/issues/2462
991,707,639
MDU6SXNzdWU5OTE3MDc2Mzk=
2,462
ThreadSafeDatabaseMetadata in doc is not compatible with inheritance
{ "login": "sherrrrr", "id": 31510228, "node_id": "MDQ6VXNlcjMxNTEwMjI4", "avatar_url": "https://avatars.githubusercontent.com/u/31510228?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sherrrrr", "html_url": "https://github.com/sherrrrr", "followers_url": "https://api.github.com/users/sherrrrr/followers", "following_url": "https://api.github.com/users/sherrrrr/following{/other_user}", "gists_url": "https://api.github.com/users/sherrrrr/gists{/gist_id}", "starred_url": "https://api.github.com/users/sherrrrr/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sherrrrr/subscriptions", "organizations_url": "https://api.github.com/users/sherrrrr/orgs", "repos_url": "https://api.github.com/users/sherrrrr/repos", "events_url": "https://api.github.com/users/sherrrrr/events{/privacy}", "received_events_url": "https://api.github.com/users/sherrrrr/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "This is expected, if somewhat non-obvious. The whole purpose of using the `ThreadSafeDatabaseMetadata` is for when you plan to switch the database at run-time in a multi-threaded application, so the assumption is that at multiple points during runtime you will be explicitly binding the models to a database, e.g.:\r\n\r\n```python\r\n\r\n# BaseModel declares that it uses ThreadSafeDatabaseMetadata\r\n\r\nclass A(BaseModel):\r\n pass\r\n\r\nclass B(BaseModel):\r\n pass\r\n\r\nmaster = PostgresqlDatabase(...)\r\nreplica = PostgresqlDatabase(...)\r\n\r\n# Explicitly bind the models.\r\nmaster.bind([A, B])\r\n```\r\n\r\nWith that said, it *should* still inherit the \"default\" or original Meta.database, so I will put together a patch for this.", "b782657823a64fcdef6674cadccb2f2679933e6f" ]
"2021-09-09T03:05:37"
"2021-09-09T13:31:37"
"2021-09-09T13:31:36"
NONE
null
[ThreadSafeDatabaseMetadata in doc](http://docs.peewee-orm.com/en/latest/peewee/database.html#thread-safety-and-multiple-databases) ```python import threading from peewee import Metadata class ThreadSafeDatabaseMetadata(Metadata): def __init__(self, *args, **kwargs): # database attribute is stored in a thread-local. self._local = threading.local() super(ThreadSafeDatabaseMetadata, self).__init__(*args, **kwargs) def _get_db(self): return getattr(self._local, 'database', self._database) def _set_db(self, db): self._local.database = self._database = db database = property(_get_db, _set_db) class BaseModel(Model): class Meta: # Instruct peewee to use our thread-safe metadata implementation. model_metadata_class = ThreadSafeDatabaseMetadata ``` when you try to use a baseModel ```python class BaseModel(Model): class Meta: database = db model_metadata_class = ThreadSafeDatabaseMetadata class User(BaseModel): id = BigAutoField(primary_key=True) ``` `User._meta.database` will not correctly inherit `BaseMoel._meta.database` because User's meta_options init in this way. ``` for k in base_meta.__dict__: if k in all_inheritable and k not in meta_options: meta_options[k] = base_meta.__dict__[k] ``` where `database` is not in `base_meta.__dict__` but `_database`.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2462/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2462/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2461
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2461/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2461/comments
https://api.github.com/repos/coleifer/peewee/issues/2461/events
https://github.com/coleifer/peewee/issues/2461
987,336,488
MDU6SXNzdWU5ODczMzY0ODg=
2,461
Trying to do bulk_update AttributeError: 'dict' object has no attribute '_pk'
{ "login": "bgriffen", "id": 2667869, "node_id": "MDQ6VXNlcjI2Njc4Njk=", "avatar_url": "https://avatars.githubusercontent.com/u/2667869?v=4", "gravatar_id": "", "url": "https://api.github.com/users/bgriffen", "html_url": "https://github.com/bgriffen", "followers_url": "https://api.github.com/users/bgriffen/followers", "following_url": "https://api.github.com/users/bgriffen/following{/other_user}", "gists_url": "https://api.github.com/users/bgriffen/gists{/gist_id}", "starred_url": "https://api.github.com/users/bgriffen/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/bgriffen/subscriptions", "organizations_url": "https://api.github.com/users/bgriffen/orgs", "repos_url": "https://api.github.com/users/bgriffen/repos", "events_url": "https://api.github.com/users/bgriffen/events{/privacy}", "received_events_url": "https://api.github.com/users/bgriffen/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "It turns out you can't feed it a list of dictionaries like you can with `insert_many`. I worked around it by creating a list of instances. Not sure if this is the best solution to my situation but it works.\r\n\r\n```python\r\nupdates = []\r\nfor u in self.update_l:\r\n e = db.PaperDB(**u)\r\n updates.append(e)\r\n\r\nwith self.dbase.atomic():\r\n db.PaperDB.bulk_update(updates,fields=self.cols, batch_size=1000)\r\n```\r\n", "The docs should make this clear: http://docs.peewee-orm.com/en/latest/peewee/api.html#Model.bulk_update\r\n\r\nAlso note that the way `bulk_update()` is implemented, it uses a couple tricks to try to do it efficiently. **However** it may actually be more performant and simple to just issue an UPDATE for each model instance/row and do this within a transaction. I suggest going with the simplest possible thing until you find that it has become a significant bottleneck." ]
"2021-09-03T02:31:33"
"2021-09-03T12:11:44"
"2021-09-03T12:11:44"
NONE
null
I am trying to do a bulk_update using the example provided. ```python class BaseModel(pw.Model): class Meta: database = get_testdb('papers.db') pass class PaperDB(BaseModel): pmid = pw.IntegerField(unique=True) title = pw.CharField() abstract = pw.TextField() journal = pw.CharField() authors = pw.CharField() pubdate = pw.DateField() mesh_terms = pw.CharField() publication_types = pw.CharField() chemical_list = pw.CharField() keywords = pw.CharField() doi = pw.CharField() delete = pw.BooleanField() affiliations = pw.CharField() pmc = pw.IntegerField() other_id = pw.CharField() medline_ta = pw.CharField() nlm_unique_id = pw.IntegerField() issn_linking = pw.CharField() country = pw.CharField() pubyear = pw.IntegerField() pubmonth = pw.IntegerField() pubday = pw.IntegerField() .... # in code with self.dbase.atomic(): for idx in range(0, num_papers_update, n_batch): # self.update_l is a list of dictionaries containing info from dataframe # self.update_l = dfi.loc[self.update_pmid].reset_index().to_dict('records') db.PaperDB.bulk_update(self.update_l[idx:idx+n_batch],fields=[db.PaperDB.pmid, db.PaperDB.title, db.PaperDB.abstract, db.PaperDB.journal, db.PaperDB.authors, db.PaperDB.pubdate, db.PaperDB.mesh_terms, db.PaperDB.publication_types, db.PaperDB.chemical_list, db.PaperDB.keywords, db.PaperDB.doi, db.PaperDB.delete, db.PaperDB.affiliations, db.PaperDB.pmc, db.PaperDB.other_id, db.PaperDB.medline_ta, db.PaperDB.nlm_unique_id, db.PaperDB.issn_linking, db.PaperDB.country, db.PaperDB.pubyear, db.PaperDB.pubmonth, db.PaperDB.pubday]) ``` but I am getting this error: ```bash 264 if ~os.path.exists(self.fname_excludes): in update_db(self, current_pmids) 138 db.PaperDB.publication_types, 139 db.PaperDB.chemical_list, --> 140 db.PaperDB.keywords, 141 db.PaperDB.doi, 142 db.PaperDB.delete, ~/anaconda3/lib/python3.8/site-packages/peewee.py in bulk_update(cls, model_list, fields, batch_size) 6350 6351 for batch in batches: -> 6352 id_list = [model._pk for model in batch] 6353 update = {} 6354 for field, attr in zip(fields, attrs): ~/anaconda3/lib/python3.8/site-packages/peewee.py in <listcomp>(.0) 6350 6351 for batch in batches: -> 6352 id_list = [model._pk for model in batch] 6353 update = {} 6354 for field, attr in zip(fields, attrs): AttributeError: 'dict' object has no attribute '_pk' ``` Also do I really have to supply all the fields from the model? Is there a way to just do it on all fields or say the ones being ingested in that item?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2461/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2461/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2460
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2460/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2460/comments
https://api.github.com/repos/coleifer/peewee/issues/2460/events
https://github.com/coleifer/peewee/issues/2460
979,798,538
MDU6SXNzdWU5Nzk3OTg1Mzg=
2,460
Multiple condition splicing
{ "login": "Baloneo", "id": 16448666, "node_id": "MDQ6VXNlcjE2NDQ4NjY2", "avatar_url": "https://avatars.githubusercontent.com/u/16448666?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Baloneo", "html_url": "https://github.com/Baloneo", "followers_url": "https://api.github.com/users/Baloneo/followers", "following_url": "https://api.github.com/users/Baloneo/following{/other_user}", "gists_url": "https://api.github.com/users/Baloneo/gists{/gist_id}", "starred_url": "https://api.github.com/users/Baloneo/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Baloneo/subscriptions", "organizations_url": "https://api.github.com/users/Baloneo/orgs", "repos_url": "https://api.github.com/users/Baloneo/repos", "events_url": "https://api.github.com/users/Baloneo/events{/privacy}", "received_events_url": "https://api.github.com/users/Baloneo/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "There are a couple ways, but probably the one I recommend for dynamically building up a query is to use reduce:\r\n\r\n```python\r\n\r\nexprs = []\r\nif name:\r\n exprs.append(ProductType.name == name)\r\nif other:\r\n exprs.append(ProductType.other == other)\r\n\r\nif exprs:\r\n query = query.where(reduce(operator.and_, exprs))\r\n```\r\n\r\nSince you're using the same operator (AND) you can also just chain the calls:\r\n\r\n```python\r\n\r\nquery = ProductType.select()\r\nif name:\r\n query = query.where(ProductType.name == name)\r\nif other:\r\n query = query.where(ProductType.other == other)\r\n...\r\n```", "thanks!" ]
"2021-08-26T03:36:40"
"2021-08-26T13:25:49"
"2021-08-26T13:05:23"
NONE
null
Sometimes the front end will pass an empty string, if where condition like this: ``` ProductType.demand_side == "" ``` will not find results, so I used the following code ```python name = current_schema_data.get('name') demand_side = current_schema_data.get('demand_side') expression = 1 if name: expression &= (ProductType.product_name == name) if demand_side: expression &= (ProductType.demand_side == demand_side) sql = ProductType.select().where(expression) print(sql) ``` ``` SELECT `t1`.`id`, `t1`.`create_time`, `t1`.`update_time`, `t1`.`product_name`, `t1`.`fk_organization_id`, `t1`.`unit`, `t1`.`type`, `t1`.`calc_type`, `t1`.`demand_side` FROM `product_type` AS `t1` WHERE (1 AND (`t1`.`demand_side` = '34')) LIMIT 100 OFFSET 0 ``` It always has `Where(1 and FOO)` or `Where 1`, How to remove it? I have try using `expression = None`, But the results it found are all empty
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2460/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2460/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2459
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2459/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2459/comments
https://api.github.com/repos/coleifer/peewee/issues/2459/events
https://github.com/coleifer/peewee/issues/2459
979,239,136
MDU6SXNzdWU5NzkyMzkxMzY=
2,459
BETWEEN does not work with datetime objects in SQlite
{ "login": "LVerneyPEReN", "id": 58298410, "node_id": "MDQ6VXNlcjU4Mjk4NDEw", "avatar_url": "https://avatars.githubusercontent.com/u/58298410?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LVerneyPEReN", "html_url": "https://github.com/LVerneyPEReN", "followers_url": "https://api.github.com/users/LVerneyPEReN/followers", "following_url": "https://api.github.com/users/LVerneyPEReN/following{/other_user}", "gists_url": "https://api.github.com/users/LVerneyPEReN/gists{/gist_id}", "starred_url": "https://api.github.com/users/LVerneyPEReN/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LVerneyPEReN/subscriptions", "organizations_url": "https://api.github.com/users/LVerneyPEReN/orgs", "repos_url": "https://api.github.com/users/LVerneyPEReN/repos", "events_url": "https://api.github.com/users/LVerneyPEReN/events{/privacy}", "received_events_url": "https://api.github.com/users/LVerneyPEReN/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Sqlite does not have a dedicated date/datetime type. It is up to you, the developer, to ensure that your datetimes are stored in a way that supports correct lexicographic ordering. Additionally, Peewee does not apply any conversion at all to datetime objects - it hands them as-is to the db driver. It is actually the sqlite driver that is handling the conversion for sqlite: (`import sqlite3; sqlite3.adapters`). This is consistent and supports lexicographic ordering, but if you've stored your datetimes in ISO format in Sqlite, then you may have some work to do to get them working properly with python's sqlite3 module.\r\n\r\nYou can:\r\n\r\n* register a different adapter for `datetime.datetime` that converts them to string in iso format - this may be the easiest\r\n* write a custom `DateTimeField` subclass that implements a `db_value()` method that returns the datetime in isoformat\r\n* update your data to use the python format (probably undesirable, but mentioning it as a possibility)\r\n\r\nAll that said, at the end of the day this is not really a peewee issue -- this is a consequence of the behavior of the python sqlite driver and sqlite's own lack of a dedicated datetime type.", "Thanks for the very quick response, this is indeed an issue due to having iso formatted values in db!\r\n\r\nI tried to debug how they happened to get there, and it seems `DateTimeField` does not convert iso formatted datetime strings to a date object, but rather append them silently to the database.\r\n\r\nThis seems to be an easy mistake which could happen often, and quite silent.\r\n\r\nWhat about extending https://github.com/coleifer/peewee/blob/master/peewee.py#L4971-L4975 to also support isoformat formats?\r\n\r\nI think something like\r\n\r\n```python\r\nformats = [\r\n '%Y-%m-%dT%H:%M:%S.%f%z',\r\n '%Y-%m-%dT%H:%M:%S.%f',\r\n '%Y-%m-%d %H:%M:%S.%f',\r\n '%Y-%m-%dT%H:%M:%S%z',\r\n '%Y-%m-%dT%H:%M:%S',\r\n '%Y-%m-%d %H:%M:%S',\r\n '%Y-%m-%d',\r\n]\r\n```\r\n\r\nshould handle all these cases properly.\r\n\r\n**EDIT**: The same formats addition is probably required as well in `DateField` and `TimeField`. I may have missed something, but I don't see any reason why isoformatted strings would not get promoted to relevant date objects for proper addition by peewee.\r\n\r\nThanks!", "That will allow you to read them back but the problem is still that the data in your db was written in iso format.\r\n\r\nSubclassing and implementing your formats and the corresponding db_value() method should cover all the bases.", "To expand on my comment, there are 2 sides to the adaptation process. The side where you take a python value (a datetime in this example) and convert it to something the db understands, and the other side where you get a value from the db and convert it to something python understands.\r\n\r\nJust extending the `formats` won't really help because it only affects the 2nd side (going from db -> python). That's why I suggested implementing `db_value()` which will correctly convert the datetime to an isoformat string before sending it to the db.\r\n\r\n```python\r\n\r\n# minimal example\r\nclass ISODateTimeField(DateTimeField):\r\n def db_value(self, val):\r\n if val is not None:\r\n return val.isoformat()\r\n def python_value(self, val):\r\n if val is not None:\r\n return datetime.datetime.strptime(val, '%Y-%m-%dT%H:%M:%S.%f')\r\n```", "Sorry, I think my latest comment may not have been clear about the details of my error (and history) and my proposal.\r\n\r\nFirst, I should emphasize that BETWEEN operation is behaving as expected for datetime objects. The issue boils down to database insertion. Then, I should also emphasize that I do not have a pre-existent database but am creating a database from scratch using Peewee from the beginning and pushing data into it.\r\n\r\nMy mistake and the source of this issue boils down to how Peewee deals with incompatible values with compatible types for datetime fields.\r\n\r\nHere is a minimal illustration about upon the quickstart page of the docs :\r\n\r\n\r\n```python\r\nfrom datetime import datetime\r\nfrom peewee import *\r\n\r\ndb = SqliteDatabase('people.db')\r\n\r\nclass Person(Model):\r\n name = CharField()\r\n birthday = DateTimeField()\r\n\r\n class Meta:\r\n database = db # This model uses the \"people.db\" database.\r\n\r\ndb.connect()\r\ndb.create_tables([Person])\r\n\r\n\r\n# Create a person with a datetime object as birthday, this is the most typical use case\r\nuncle_bob = Person(name='Bob', birthday=datetime(1960, 1, 15, 1, 0, 0))\r\nuncle_bob.save()\r\n\r\n# Create a person with a string representing a date. String format matches one of https://github.com/coleifer/peewee/blob/master/peewee.py#L4971-L4975 so we hit https://github.com/coleifer/peewee/blob/master/peewee.py#L4948 and it is promoted to datetime object.\r\n# This is therefore **the same behavior** as before, as expected\r\nuncle_sam = Person(\r\n name='Sam',\r\n birthday=\"1960-01-15 01:00:00\"\r\n)\r\nuncle_sam.save()\r\n\r\n# Create a person with an isoformatted string. This format does not match any of the known formats for DateTimeField, so we hit https://github.com/coleifer/peewee/blob/master/peewee.py#L4949-L4951 which is completely silent and the value is inserted as is in the database.\r\n# This is how I ended up with isoformatted dates in my db, by (incorrectly) manipulating an isoformatted string at some point (while I thought I had a datetime object) and Peewee / sqlite3 not throwing any error at me.\r\nuncle_tom = Person(\r\n name='tom',\r\n birthday=\"1960-01-15T01:00:00Z\"\r\n)\r\nuncle_tom.save()\r\n\r\n# Note that when the object type is incompatible, sqlite3 will shout. But for any string which does not match a known date format, it gets inserted as is in the db.\r\nuncle_guy = Person(name='Guy', birthday=\"This is definitely not a date\")\r\nuncle_guy.save() #\r\n```\r\n\r\n\r\n\r\nI feel like the fact that Peewee supports some date string formats is giving the wrong message that passing string instead of datetime object is safe, and one is likely to assume isoformatted datetime is fine. This is not the case.\r\n\r\nSo, I would like Peewee to minimize the possibility to reproduce this error (while trying to reduce the extra work on Peewee to a minimum) by:\r\n* Adding new formats in https://github.com/coleifer/peewee/blob/master/peewee.py#L4971-L4975 to support isoformatted dates the same way as the other date format. See https://github.com/coleifer/peewee/issues/2459#issuecomment-905605324.\r\n* Probably add a `logging.warning` item in https://github.com/coleifer/peewee/blob/master/peewee.py#L4950 so that this promotion error is no longer silent and the user has some form of easy discovery that they are not using Peewee the right way.\r\n\r\nI can offer to handle a PR for this if this can be useful and you agree with my reading of this issue.", "I'd suggest doing stronger validation before insertion into the DB and keeping in mind Sqlite's loose typing -- for fiddly things like datetimes this can be a gotcha for Sqlite users as it is up to you to be consistent about how you insert them to preserve ordering. I don't plan to make changes on the Peewee side at this time." ]
"2021-08-25T14:47:27"
"2021-08-26T13:01:18"
"2021-08-25T14:58:55"
NONE
null
Hi, I have an issue where `BETWEEN` operation which does not work with a `DateTimeField` in SQLite when checking against Python `datetime.datetime` objects. It seems this is due to the default adapter being used for such formats in SQLite is `str` conversion, which is not the same as an `isoformat`. Therefore, lexicographic comparison is broken afterwards. See ```python >>> import datetime >>> str(datetime.datetime.now()) '2021-08-25 16:44:43.952660' >>> datetime.datetime.now().isoformat() '2021-08-25T16:44:56.216273' ``` And the extra "T" between date and time part. Thanks!
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2459/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2459/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2458
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2458/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2458/comments
https://api.github.com/repos/coleifer/peewee/issues/2458/events
https://github.com/coleifer/peewee/issues/2458
978,766,431
MDU6SXNzdWU5Nzg3NjY0MzE=
2,458
select (15 minutes) vs execute_sql (35 seconds)
{ "login": "netsamir", "id": 1165885, "node_id": "MDQ6VXNlcjExNjU4ODU=", "avatar_url": "https://avatars.githubusercontent.com/u/1165885?v=4", "gravatar_id": "", "url": "https://api.github.com/users/netsamir", "html_url": "https://github.com/netsamir", "followers_url": "https://api.github.com/users/netsamir/followers", "following_url": "https://api.github.com/users/netsamir/following{/other_user}", "gists_url": "https://api.github.com/users/netsamir/gists{/gist_id}", "starred_url": "https://api.github.com/users/netsamir/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/netsamir/subscriptions", "organizations_url": "https://api.github.com/users/netsamir/orgs", "repos_url": "https://api.github.com/users/netsamir/repos", "events_url": "https://api.github.com/users/netsamir/events{/privacy}", "received_events_url": "https://api.github.com/users/netsamir/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Hi,\r\n\r\nI think I have found the solution to my problem by logging with DEBUG.\r\n\r\nI have rewritten the query like that :\r\n\r\n for row in User.select(User.Service, fn.count(1).alias('count')).group_by(User.Service):\r\n print(row.Service, row.count)\r\n\r\n DEBUG:peewee:('SELECT \"t1\".\"Service\", count(?) AS \"count\" FROM \"sag\" AS \"t1\" GROUP BY \"t1\".\"Service\"', [1])\r\n\r\ninstead of \r\n\r\n DEBUG:peewee:('SELECT \"t1\".\"id\", \"t1\".\"LogSysTime\", \"t1\".\"LogCorrelationId\", \"t1\".\"ApplicationId\", \"t1\".\"UnitName\", \"t1\".\"Primitive\", \"t1\".\"Priority\", \"t1\".\"RequControl\", \"t1\".\"RequtRef\", \"t1\".\"RequType\", \"t1\".\"Requestor\", \"t1\".\"Resp\", \"t1\".\"ReserRef\", \"t1\".\"ResponseControl\", \"t1\".\"Service\", \"t1\".\"Status\", \"t1\".\"SwtRef\", \"t1\".\"SwRequestRef\", \"t1\".\"SwResponseRef\", \"t1\".\"SwTime\", \"t1\".\"TransferRef\", \"t1\".\"PayloadSize\", count(?) AS \"count\" FROM \"sag\" AS \"t1\" GROUP BY \"t1\".\"Service\"', [1])", "If you were using postgres your first query would not have even run because you were selecting all columns but only grouping by one of them.\r\n\r\nThe second query is correct:\r\n\r\n q = User.select(User.Service, fn.count(User.id).alias('count')).group_by(User.Service)" ]
"2021-08-25T06:25:21"
"2021-08-25T12:34:13"
"2021-08-25T12:34:13"
NONE
null
Hi, Please find below two very simple script that execute the exact same query, but: - The first one, `select`, takes 15 min to complete - The second one, `execute_sql`, takes 35 seconds to complete Do you know how to improve performance of the first script please ? ### Here are the details of the database: The database I am using is : sqlite3. It is 5.1GB. $ sqlite3 --version 3.7.17 2013-05-20 00:56:22 118a3b35693b134d56ebd780123b7fd6f1497668 [ins] In [7]: print_table_sql(Sag) CREATE TABLE IF NOT EXISTS "sag" ( "id" INTEGER NOT NULL PRIMARY KEY, "LogSysTime" DATETIME NOT NULL, "LogId" VARCHAR(255) NOT NULL, "Apd" VARCHAR(255) NOT NULL, "UnitName" VARCHAR(255) NOT NULL, "Primitive" VARCHAR(255), "Priority" VARCHAR(255), "RequestControl" VARCHAR(255), "RequestRef" VARCHAR(255), "RequestType" VARCHAR(255), "Sender" VARCHAR(255), "Receiver" VARCHAR(255), "ReceiverRef" VARCHAR(255), "ResponseControl" VARCHAR(255), "Service" VARCHAR(255), "Status" VARCHAR(255), "SwtRef" VARCHAR(255), "SwtRequestRef" VARCHAR(255), "SwtResponseRef" VARCHAR(255), "SwtTime" VARCHAR(255), "TransferRef" VARCHAR(255), "PayloadSize" INTEGER ) ### Here are the details of the scripts: 1. Script : `select` $ cat script1.py from parser_User_stats_all_traffic import User, db from peewee import * db = SqliteDatabase('user_db.sqlite') \# database schema and creation for row in User.select(User, fn.count(1).alias('count')).group_by(User.Service): print(row.Service, row.count) $ time python script1.py <output> real 15m7.341s user 1m10.136s sys 1m6.303s **Additional info:** This script create 2 temporary files with 4.4 GB and then it hanged for quiet some time without consuming CPU nor Memory. 2. Script : `execute_sql` $ cat script2.py from parser_User_stats_all_traffic import User, db from peewee import * db = SqliteDatabase('user_db.sqlite') \# database schema and creation for row in db.execute_sql('select Service, count(1) from User group by Service'): print(row) $ time python script2.py <output> real 0m35.093s user 0m19.478s sys 0m9.827s **Additional info:** It creates 2 temporary files of 147 MB and return.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2458/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2458/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2457
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2457/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2457/comments
https://api.github.com/repos/coleifer/peewee/issues/2457/events
https://github.com/coleifer/peewee/issues/2457
975,841,918
MDU6SXNzdWU5NzU4NDE5MTg=
2,457
UUIDField with bulk_update fails
{ "login": "milesgranger", "id": 13764397, "node_id": "MDQ6VXNlcjEzNzY0Mzk3", "avatar_url": "https://avatars.githubusercontent.com/u/13764397?v=4", "gravatar_id": "", "url": "https://api.github.com/users/milesgranger", "html_url": "https://github.com/milesgranger", "followers_url": "https://api.github.com/users/milesgranger/followers", "following_url": "https://api.github.com/users/milesgranger/following{/other_user}", "gists_url": "https://api.github.com/users/milesgranger/gists{/gist_id}", "starred_url": "https://api.github.com/users/milesgranger/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/milesgranger/subscriptions", "organizations_url": "https://api.github.com/users/milesgranger/orgs", "repos_url": "https://api.github.com/users/milesgranger/repos", "events_url": "https://api.github.com/users/milesgranger/events{/privacy}", "received_events_url": "https://api.github.com/users/milesgranger/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "See #2308" ]
"2021-08-20T18:47:28"
"2021-08-20T20:22:26"
"2021-08-20T20:22:26"
NONE
null
Summary --- It appears using `Model.bulk_update` with a model using a `UUIDField` fails. Minimal Example --- ```python def test_uuid_bulk_update(db): n_rows = 1_000 class Foo(peewee.Model): class Meta: database = db pk = peewee.IntegerField(primary_key=True) bar = peewee.UUIDField() if Foo.table_exists(): Foo.drop_table() Foo.create_table() data = [ {"pk": i, "bar": uuid.uuid3(uuid.NAMESPACE_URL, f"{i}")} for i in range(n_rows) ] with db.atomic(): Foo.bulk_create((Foo(**kwargs) for kwargs in data)) updates = [ {"pk": item["pk"], "bar": uuid.uuid3(uuid.NAMESPACE_URL, f"{i*1.5}")} for i, item in enumerate(data[:100]) ] # Saving them individually works fine. # for update in updates: # Foo(**update).save() # This fails. Foo.bulk_update([Foo(**kwargs) for kwargs in updates], fields=["bar"]) ``` Using a `TextField` instead will work fine in the code above. Error --- ``` > with __exception_wrapper__: cursor = self.cursor(commit) try: cursor.execute(sql, params or ()) E psycopg2.errors.DatatypeMismatch: column "bar" is of type uuid but expression is of type text E LINE 1: UPDATE "foo" SET "bar" = CASE "foo"."pk" WHEN 0 THEN 'e0a687... E ^ E HINT: You will need to rewrite or cast the expression. ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2457/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2457/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2456
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2456/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2456/comments
https://api.github.com/repos/coleifer/peewee/issues/2456/events
https://github.com/coleifer/peewee/issues/2456
970,796,314
MDU6SXNzdWU5NzA3OTYzMTQ=
2,456
Is there a module that implements Mock?
{ "login": "ekiim", "id": 6476691, "node_id": "MDQ6VXNlcjY0NzY2OTE=", "avatar_url": "https://avatars.githubusercontent.com/u/6476691?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ekiim", "html_url": "https://github.com/ekiim", "followers_url": "https://api.github.com/users/ekiim/followers", "following_url": "https://api.github.com/users/ekiim/following{/other_user}", "gists_url": "https://api.github.com/users/ekiim/gists{/gist_id}", "starred_url": "https://api.github.com/users/ekiim/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ekiim/subscriptions", "organizations_url": "https://api.github.com/users/ekiim/orgs", "repos_url": "https://api.github.com/users/ekiim/repos", "events_url": "https://api.github.com/users/ekiim/events{/privacy}", "received_events_url": "https://api.github.com/users/ekiim/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "No, mocking a sql database is an extremely complex affair. Most people use shallow mocks or just an in memory sqlite database for testing." ]
"2021-08-14T02:04:42"
"2021-08-14T15:11:04"
"2021-08-14T15:11:04"
NONE
null
Hello, I'm trying to find out if there is a module for mocking for models. I know it's a small project, I'm interested in finding one.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2456/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2456/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2455
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2455/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2455/comments
https://api.github.com/repos/coleifer/peewee/issues/2455/events
https://github.com/coleifer/peewee/pull/2455
963,324,836
MDExOlB1bGxSZXF1ZXN0NzA1OTYxMzQx
2,455
Update api.rst
{ "login": "aparcar", "id": 16000931, "node_id": "MDQ6VXNlcjE2MDAwOTMx", "avatar_url": "https://avatars.githubusercontent.com/u/16000931?v=4", "gravatar_id": "", "url": "https://api.github.com/users/aparcar", "html_url": "https://github.com/aparcar", "followers_url": "https://api.github.com/users/aparcar/followers", "following_url": "https://api.github.com/users/aparcar/following{/other_user}", "gists_url": "https://api.github.com/users/aparcar/gists{/gist_id}", "starred_url": "https://api.github.com/users/aparcar/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/aparcar/subscriptions", "organizations_url": "https://api.github.com/users/aparcar/orgs", "repos_url": "https://api.github.com/users/aparcar/repos", "events_url": "https://api.github.com/users/aparcar/events{/privacy}", "received_events_url": "https://api.github.com/users/aparcar/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Actually the solution is as simple as `(idlist.split(\",\") if idlist else [])`", "a643a6952372e22ce658c8ca9a23322bb4b2ab57" ]
"2021-08-08T00:12:24"
"2021-08-08T13:17:59"
"2021-08-08T13:17:59"
NONE
null
The example is wrong since it doesn't split the `idlist`. This is the same function as used in the test. However the example has a case with an empty list (`[]`) which won't work, since `None.split(",")` won't work. Any simple solutions here?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2455/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2455/timeline
null
null
true
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2455", "html_url": "https://github.com/coleifer/peewee/pull/2455", "diff_url": "https://github.com/coleifer/peewee/pull/2455.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2455.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2454
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2454/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2454/comments
https://api.github.com/repos/coleifer/peewee/issues/2454/events
https://github.com/coleifer/peewee/issues/2454
962,626,425
MDU6SXNzdWU5NjI2MjY0MjU=
2,454
Insert Many\Upsert (PostgreSQL) doesn't work with M2M relations
{ "login": "madMathematician971", "id": 63659861, "node_id": "MDQ6VXNlcjYzNjU5ODYx", "avatar_url": "https://avatars.githubusercontent.com/u/63659861?v=4", "gravatar_id": "", "url": "https://api.github.com/users/madMathematician971", "html_url": "https://github.com/madMathematician971", "followers_url": "https://api.github.com/users/madMathematician971/followers", "following_url": "https://api.github.com/users/madMathematician971/following{/other_user}", "gists_url": "https://api.github.com/users/madMathematician971/gists{/gist_id}", "starred_url": "https://api.github.com/users/madMathematician971/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/madMathematician971/subscriptions", "organizations_url": "https://api.github.com/users/madMathematician971/orgs", "repos_url": "https://api.github.com/users/madMathematician971/repos", "events_url": "https://api.github.com/users/madMathematician971/events{/privacy}", "received_events_url": "https://api.github.com/users/madMathematician971/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "How is an upsert supposed to work across multiple tables? I think you do not understand how many-to-many actually works -- there is a junction table. There is no such thing as website.regions - that information is stored in the through table." ]
"2021-08-06T10:51:31"
"2021-08-06T19:51:44"
"2021-08-06T19:51:44"
NONE
null
Hello everyone, Trying to utilize an upsert along with an insert_many query for a model with m2m relation. The error I get is `'ManyToManyField' object has no attribute '_sort_key'` The code example is provided below. Trying to figure out if this is expected behavior due to the fact that ManyToManyField is not a usual peewee field in the default sense or if it's a bug. ``` class Region(_BaseModel): name = pw.CharField(unique=True) class Website(_BaseModel): name = pw.CharField() url = pw.CharField(unique=True) regions = pw.ManyToManyField(Region, backref='websites') WebsiteRegion = Website.regions.get_through_model() ``` (_Base model has a UUID4 id assigned to it) However, when I'm trying to use an upsert which was implemented before M2M relation was added to Website Model and worked just fine, I get the error listed above ('ManyToManyField' object has no attribute '_sort_key') The upsert setup: ``` query = Website.insert_many(self.data).returning(*self.__RETURNING).on_conflict( conflict_target=[Website.url], preserve=[Website.url], update={ Website.name: EXCLUDED.name, Website.regions: EXCLUDED.regions, } ) ``` where self.data is a valid list of pydantic models converted to dicts and self.__RETURNING is a valid list of `Website` model attributes desired to be returned. Perhaps I am missing something required for m2m to work in the upsert? I already have a workaround utilizing a loop with `website = Website.create(<data>)` and a default `website.regions.add(regions: List[UUID])` for populating m2m, but still having a working upsert for m2m would be great if possible, as create and update in a loop is less efficient than an upsert. Anyone faced a similar issue with upsert and m2m before and what were your solutions \ workarounds? Appreciate your time!
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2454/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2454/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2453
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2453/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2453/comments
https://api.github.com/repos/coleifer/peewee/issues/2453/events
https://github.com/coleifer/peewee/issues/2453
962,161,857
MDU6SXNzdWU5NjIxNjE4NTc=
2,453
Problem with ended parentheses alias
{ "login": "wes9319", "id": 11861951, "node_id": "MDQ6VXNlcjExODYxOTUx", "avatar_url": "https://avatars.githubusercontent.com/u/11861951?v=4", "gravatar_id": "", "url": "https://api.github.com/users/wes9319", "html_url": "https://github.com/wes9319", "followers_url": "https://api.github.com/users/wes9319/followers", "following_url": "https://api.github.com/users/wes9319/following{/other_user}", "gists_url": "https://api.github.com/users/wes9319/gists{/gist_id}", "starred_url": "https://api.github.com/users/wes9319/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/wes9319/subscriptions", "organizations_url": "https://api.github.com/users/wes9319/orgs", "repos_url": "https://api.github.com/users/wes9319/repos", "events_url": "https://api.github.com/users/wes9319/events{/privacy}", "received_events_url": "https://api.github.com/users/wes9319/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I don't know if I can change this logic without introducing a possible regression for users who select arbitrary sql functions and do not alias them. The current logic looks if there's a \".\" within the column name, takes the stuff to the right of it and strips `\"` and `)`, e.g.:\r\n\r\nUPPER(\"t1\".\"name\") -> \"name\") -> name\r\n\r\nThe downside is that this logic is also being applied to explicitly-given aliases. Let me think about it.", "The fix involved basically ensuring that user-provided aliases would be preserved, but at the same time I've tried to improve the logic when guessing at the intended destination column when it is not provided." ]
"2021-08-05T20:13:40"
"2021-08-08T13:29:16"
"2021-08-08T13:28:36"
NONE
null
I had a problem trying to query data and assign an alias ended by a parentheses to each row, the data indeed is correct but the last parentheses of the alias gets cut off in the result data. Here's an example: ``` SUM(IF((((`t2`.`class` IN (%s, %s, %s)) AND (`t2`.`concept` != %s)) AND (`t3`.`cost_id` = %s)), IFNULL((`t2`.`value` * (`t3`.`value` / %s)), %s), %s)) AS `cost center (CCA)`, ``` The result is: ``` 'cost center (CCA': Decimal('139.704440000000') ``` The value is correct but the the alias is wrong due to the missing parentheses, is it possible to fix this in any way?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2453/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2453/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2452
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2452/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2452/comments
https://api.github.com/repos/coleifer/peewee/issues/2452/events
https://github.com/coleifer/peewee/issues/2452
959,535,883
MDU6SXNzdWU5NTk1MzU4ODM=
2,452
Accessing views via models not possible
{ "login": "SilvanVerhoeven", "id": 44174681, "node_id": "MDQ6VXNlcjQ0MTc0Njgx", "avatar_url": "https://avatars.githubusercontent.com/u/44174681?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SilvanVerhoeven", "html_url": "https://github.com/SilvanVerhoeven", "followers_url": "https://api.github.com/users/SilvanVerhoeven/followers", "following_url": "https://api.github.com/users/SilvanVerhoeven/following{/other_user}", "gists_url": "https://api.github.com/users/SilvanVerhoeven/gists{/gist_id}", "starred_url": "https://api.github.com/users/SilvanVerhoeven/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SilvanVerhoeven/subscriptions", "organizations_url": "https://api.github.com/users/SilvanVerhoeven/orgs", "repos_url": "https://api.github.com/users/SilvanVerhoeven/repos", "events_url": "https://api.github.com/users/SilvanVerhoeven/events{/privacy}", "received_events_url": "https://api.github.com/users/SilvanVerhoeven/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "What version of sqlite are you using?", "The requirements.txt requires peewee 3.9.6", "No, what version of sqlite.", "I'm sorry. SQLite 3.31.1 (retrieved via `select sqlite_version()`)", "This example code is working fine for me, I suppose you have some issue in your code or you can try updating your peewee if the code below does not work for you:\r\n\r\n```python\r\nfrom peewee import *\r\n\r\ndb = SqliteDatabase(':memory:')\r\n\r\nclass Foo(Model):\r\n name = TextField()\r\n class Meta: database = db\r\n\r\ndb.create_tables([Foo])\r\nfor name in ('huey', 'mickey', 'zaizee'):\r\n Foo.create(name=name)\r\n\r\ndb.execute_sql('create view foo_view as select upper(name) as name from foo')\r\nclass FooView(Foo):\r\n class Meta:\r\n primary_key = False\r\n table_name = 'foo_view'\r\n\r\nprint([row.name for row in FooView.select()])\r\n```\r\n\r\nPrints: `['HUEY', 'MICKEY', 'ZAIZEE']`" ]
"2021-08-03T21:38:10"
"2021-08-04T17:24:16"
"2021-08-04T17:23:18"
NONE
null
I try to access a view in my SQLite database following the instructions from https://stackoverflow.com/questions/38707331/does-peewee-support-interaction-with-mysql-views. I figured that the article's `db_table` needs to be replaced with `table_name` in the current version of peewee. I used the following SQL statement to create the view: `db.execute_sql('CREATE VIEW IF NOT EXISTS foo_view AS SELECT UPPER(name) AS name FROM foo')`. Unfortunately, even after this update my `FooView` model only returns `None` for the name. Printing the vars via `print([vars(fv) for fv in FooView.select(FooView.name)])` shows why: `[{'__data__': {}, '_dirty': set(), '__rel__': {}, 't1': 'HUEY'}, {'__data__': {}, '_dirty': set(), '__rel__': {}, 't1': 'MICKEY'}, {'__data__': {}, '_dirty': set(), '__rel__': {}, 't1': 'ZAIZEE'}]` Compare this to the output of `print([vars(fv) for fv in Foo.select(Foo.name)])`: `[{'__data__': {'name': 'huey'}, '_dirty': set(), '__rel__': {}}, {'__data__': {'name': 'mickey'}, '_dirty': set(), '__rel__': {}}, {'__data__': {'name': 'zaizee'}, '_dirty': set(), '__rel__': {}}}]` Interestingly enough, setting `Foo.Meta.table_name` to `foo_view` yields the same result as simply using `FooView`. I wonder if this is a bug in peewee or whether I am doing something wrong. Help much appreciated!
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2452/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2452/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2451
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2451/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2451/comments
https://api.github.com/repos/coleifer/peewee/issues/2451/events
https://github.com/coleifer/peewee/issues/2451
957,674,183
MDU6SXNzdWU5NTc2NzQxODM=
2,451
Peewee does not automatically reconnect to MySQL Error: peewee.InterfaceError: (0, '')
{ "login": "Baloneo", "id": 16448666, "node_id": "MDQ6VXNlcjE2NDQ4NjY2", "avatar_url": "https://avatars.githubusercontent.com/u/16448666?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Baloneo", "html_url": "https://github.com/Baloneo", "followers_url": "https://api.github.com/users/Baloneo/followers", "following_url": "https://api.github.com/users/Baloneo/following{/other_user}", "gists_url": "https://api.github.com/users/Baloneo/gists{/gist_id}", "starred_url": "https://api.github.com/users/Baloneo/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Baloneo/subscriptions", "organizations_url": "https://api.github.com/users/Baloneo/orgs", "repos_url": "https://api.github.com/users/Baloneo/repos", "events_url": "https://api.github.com/users/Baloneo/events{/privacy}", "received_events_url": "https://api.github.com/users/Baloneo/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "db.close()\r\n```\r\n'Attempting to close database while '\r\npeewee.OperationalError: Attempting to close database while transaction is open\r\n```", "```\r\n\r\nclass ReconnectMixin(object):\r\n \"\"\"\r\n Mixin class that attempts to automatically reconnect to the database under\r\n certain error conditions.\r\n\r\n For example, MySQL servers will typically close connections that are idle\r\n for 28800 seconds (\"wait_timeout\" setting). If your application makes use\r\n of long-lived connections, you may find your connections are closed after\r\n a period of no activity. This mixin will attempt to reconnect automatically\r\n when these errors occur.\r\n\r\n This mixin class probably should not be used with Postgres (unless you\r\n REALLY know what you are doing) and definitely has no business being used\r\n with Sqlite. If you wish to use with Postgres, you will need to adapt the\r\n `reconnect_errors` attribute to something appropriate for Postgres.\r\n \"\"\"\r\n reconnect_errors = (\r\n # Error class, error message fragment (or empty string for all).\r\n (OperationalError, '2006'), # MySQL server has gone away.\r\n (OperationalError, '2013'), # Lost connection to MySQL server.\r\n (OperationalError, '2014'), # Commands out of sync.\r\n (InterfaceError, \"(0, '')\"), # add error \r\n\r\n # mysql-connector raises a slightly different error when an idle\r\n # connection is terminated by the server. This is equivalent to 2013.\r\n (OperationalError, 'MySQL Connection not available.'),\r\n )\r\n\r\n def __init__(self, *args, **kwargs):\r\n super(ReconnectMixin, self).__init__(*args, **kwargs)\r\n\r\n # Normalize the reconnect errors to a more efficient data-structure.\r\n self._reconnect_errors = {}\r\n for exc_class, err_fragment in self.reconnect_errors:\r\n self._reconnect_errors.setdefault(exc_class, [])\r\n self._reconnect_errors[exc_class].append(err_fragment.lower())\r\n\r\n def execute_sql(self, sql, params=None, commit=SENTINEL):\r\n try:\r\n return super(ReconnectMixin, self).execute_sql(sql, params, commit)\r\n except Exception as exc:\r\n exc_class = type(exc)\r\n if exc_class not in self._reconnect_errors:\r\n raise exc\r\n\r\n exc_repr = str(exc).lower()\r\n print('exc_repr', exc_repr, 'ccc', self._reconnect_errors[exc_class])\r\n for err_fragment in self._reconnect_errors[exc_class]:\r\n if err_fragment in exc_repr:\r\n break\r\n else:\r\n raise exc\r\n\r\n if not self.is_closed():\r\n print('in_transaction --->', db.in_transaction())\r\n self.close()\r\n self.connect()\r\n\r\n return super(ReconnectMixin, self).execute_sql(sql, params, commit)\r\n```\r\n`self.is_closed()` is `False`, but `db.in_transaction()` is `True`, in close function:\r\n```\r\n def close(self):\r\n with self._lock:\r\n if self.deferred:\r\n raise InterfaceError('Error, database must be initialized '\r\n 'before opening a connection.')\r\n if self.in_transaction():\r\n raise OperationalError('Attempting to close database while '\r\n 'transaction is open.')\r\n is_open = not self._state.closed\r\n try:\r\n if is_open:\r\n with __exception_wrapper__:\r\n self._close(self._state.conn)\r\n finally:\r\n self._state.reset()\r\n return is_open\r\n```\r\n\r\nalways raise OperationalError, caused to never reconnect successfully, Is there any good solution?\r\n\r\n\r\n", "The reconnecting mixin does not play well with transactions, but this should make sense to you if you think about it. What is peewee supposed to do upon reconnect? Re-execute any previous queries that were executed in the transaction?\r\n\r\nIn cases like these you probably need to handle it at the application level. Attempting to steamroll over the problems by reconnecting is not a good solution in any event." ]
"2021-08-02T02:10:27"
"2021-08-02T13:02:56"
"2021-08-02T13:02:56"
NONE
null
``` class ReconnectMySQLDatabase(ReconnectMixin, MySQLDatabase): pass db = ReconnectMySQLDatabase(current_config.MYSQL_DB_NAME, host=current_config.MYSQL_DB_IP, port=current_config.MYSQL_DB_PORT, user=current_config.MYSQL_DB_USER, password=current_config.MYSQL_DB_PASSWORD) try: with db.atomic(): _insert_into_table(data, ENERGY_TABLE_NAME_DEFAULT()) _insert_into_table(data, ENERGY_TABLE_NAME_MONTH()) _insert_into_table(data, ENERGY_TABLE_NAME_DAY()) except Exception as e: logger.exception('Error') ``` 1 stop mysql server ``` docker-compose down -f mysql-docker-compose.yml ``` 2 log outout ``` Traceback (most recent call last): File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 3297, in commit return self._state.conn.commit() File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/pymysql/connections.py", line 469, in commit self._execute_command(COMMAND.COM_QUERY, "COMMIT") File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/pymysql/connections.py", line 793, in _execute_command raise err.InterfaceError(0, "") pymysql.err.InterfaceError: (0, '') During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 4209, in __exit__ self.commit(False) File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 4188, in commit self.db.commit() File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 3297, in commit return self._state.conn.commit() File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 2917, in __exit__ reraise(new_type, new_type(exc_value, *exc_args), traceback) File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 190, in reraise raise value.with_traceback(tb) File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 3297, in commit return self._state.conn.commit() File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/pymysql/connections.py", line 469, in commit self._execute_command(COMMAND.COM_QUERY, "COMMIT") File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/pymysql/connections.py", line 793, in _execute_command raise err.InterfaceError(0, "") peewee.InterfaceError: (0, '') During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 3301, in rollback return self._state.conn.rollback() File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/pymysql/connections.py", line 479, in rollback self._execute_command(COMMAND.COM_QUERY, "ROLLBACK") File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/pymysql/connections.py", line 793, in _execute_command raise err.InterfaceError(0, "") pymysql.err.InterfaceError: (0, '') ``` 3 start MySql service ``` docker-compose -f mysql-docker-compose.yml up -d ``` still insert error. But sometimes stop mysql service, the error output is: ``` Traceback (most recent call last): File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 3080, in connect self._state.set_connection(self._connect()) File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/peewee.py", line 3982, in _connect conn = mysql.connect(db=self.database, **self.connect_params) File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/pymysql/connections.py", line 353, in __init__ self.connect() File "/home/baloneo/kexin-code/data-collection-service/.venv/lib/python3.8/site-packages/pymysql/connections.py", line 664, in connect raise exc pymysql.err.OperationalError: (2003, "Can't connect to MySQL server on '192.168.1.80' ([Errno 111] Connection refused)") ``` It can automatically reconnect.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2451/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2451/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2450
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2450/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2450/comments
https://api.github.com/repos/coleifer/peewee/issues/2450/events
https://github.com/coleifer/peewee/issues/2450
955,993,414
MDU6SXNzdWU5NTU5OTM0MTQ=
2,450
SQL from migrator
{ "login": "konradkur", "id": 22791650, "node_id": "MDQ6VXNlcjIyNzkxNjUw", "avatar_url": "https://avatars.githubusercontent.com/u/22791650?v=4", "gravatar_id": "", "url": "https://api.github.com/users/konradkur", "html_url": "https://github.com/konradkur", "followers_url": "https://api.github.com/users/konradkur/followers", "following_url": "https://api.github.com/users/konradkur/following{/other_user}", "gists_url": "https://api.github.com/users/konradkur/gists{/gist_id}", "starred_url": "https://api.github.com/users/konradkur/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/konradkur/subscriptions", "organizations_url": "https://api.github.com/users/konradkur/orgs", "repos_url": "https://api.github.com/users/konradkur/repos", "events_url": "https://api.github.com/users/konradkur/events{/privacy}", "received_events_url": "https://api.github.com/users/konradkur/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "This is not really exposed. Operations may be composed of multiple steps, so the migrations module uses an abstraction (`Operation`) to run potentially multiple queries. Additionally, steps may include validation that requires a previous step having been run (before the sql is generated). How do you get the sql for an operation in general, though (which may be composed of sub-operations)?\r\n\r\n```python\r\n# i think this should, generally, work\r\nop = migrator.add_column('my_sqlite_table', 'new_col', TextField(default=''))\r\nq = [op]\r\nwhile q:\r\n item = q.pop()\r\n if isinstance(item, Operation):\r\n kw = item.kwargs.copy()\r\n kw['with_context'] = True\r\n q.append(getattr(migrator, item.method)(*item.args, **kw))\r\n elif isinstance(item, list):\r\n q.extend(item[::-1])\r\n elif isinstance(item, Node):\r\n print(migrator.make_context().sql(item).query())\r\n else:\r\n print(item.query())\r\n```\r\n\r\nAlternatively, you can attach a sql log handler:\r\n\r\n```python\r\nimport logging\r\nlogger = logging.getLogger('peewee')\r\nlogger.addHandler(logging.StreamHandler())\r\nlogger.setLevel(logging.DEBUG)\r\n```" ]
"2021-07-29T16:06:18"
"2021-07-29T21:27:43"
"2021-07-29T21:27:43"
NONE
null
Hello my friends, `migrate( migrator.add_column('some_table', 'title', title_field), migrator.add_column('some_table', 'status', status_field), migrator.drop_column('some_table', 'old_column'), )` how to get SQL code? thanks, Konrad
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2450/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2450/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2449
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2449/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2449/comments
https://api.github.com/repos/coleifer/peewee/issues/2449/events
https://github.com/coleifer/peewee/issues/2449
950,542,342
MDU6SXNzdWU5NTA1NDIzNDI=
2,449
Can't import models before creating a Database object
{ "login": "AidenEllis", "id": 64432103, "node_id": "MDQ6VXNlcjY0NDMyMTAz", "avatar_url": "https://avatars.githubusercontent.com/u/64432103?v=4", "gravatar_id": "", "url": "https://api.github.com/users/AidenEllis", "html_url": "https://github.com/AidenEllis", "followers_url": "https://api.github.com/users/AidenEllis/followers", "following_url": "https://api.github.com/users/AidenEllis/following{/other_user}", "gists_url": "https://api.github.com/users/AidenEllis/gists{/gist_id}", "starred_url": "https://api.github.com/users/AidenEllis/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/AidenEllis/subscriptions", "organizations_url": "https://api.github.com/users/AidenEllis/orgs", "repos_url": "https://api.github.com/users/AidenEllis/repos", "events_url": "https://api.github.com/users/AidenEllis/events{/privacy}", "received_events_url": "https://api.github.com/users/AidenEllis/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[]
"2021-07-22T11:05:04"
"2021-07-22T12:28:55"
"2021-07-22T12:28:55"
NONE
null
.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2449/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 1, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2449/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2448
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2448/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2448/comments
https://api.github.com/repos/coleifer/peewee/issues/2448/events
https://github.com/coleifer/peewee/issues/2448
946,093,107
MDU6SXNzdWU5NDYwOTMxMDc=
2,448
Dynamic table names refer to global variables
{ "login": "Baloneo", "id": 16448666, "node_id": "MDQ6VXNlcjE2NDQ4NjY2", "avatar_url": "https://avatars.githubusercontent.com/u/16448666?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Baloneo", "html_url": "https://github.com/Baloneo", "followers_url": "https://api.github.com/users/Baloneo/followers", "following_url": "https://api.github.com/users/Baloneo/following{/other_user}", "gists_url": "https://api.github.com/users/Baloneo/gists{/gist_id}", "starred_url": "https://api.github.com/users/Baloneo/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Baloneo/subscriptions", "organizations_url": "https://api.github.com/users/Baloneo/orgs", "repos_url": "https://api.github.com/users/Baloneo/repos", "events_url": "https://api.github.com/users/Baloneo/events{/privacy}", "received_events_url": "https://api.github.com/users/Baloneo/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "table name is not a pointer, it is set and stored in the model class' `table_name` attribute. If you want to change it at run-time, then you need to call `ModelClass._meta.set_table_name()`." ]
"2021-07-16T09:06:40"
"2021-07-16T13:05:42"
"2021-07-16T13:05:41"
NONE
null
``` GLOBAL_VAR_TABLE_NAME = "foo_table" def make_table_name(model_class): model_name = model_class.__name__ return GLOBAL_VAR_TABLE_NAME class BaseModel(Model): class Meta: table_function = make_table_name class User(BaseModel): # table_name will be "user_tbl". ``` update global variable ``` GLOBAL_VAR_TABLE_NAME = "bar_table" ``` but ``` User().save() ``` still is foo_table, why?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2448/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2448/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2447
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2447/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2447/comments
https://api.github.com/repos/coleifer/peewee/issues/2447/events
https://github.com/coleifer/peewee/pull/2447
945,804,361
MDExOlB1bGxSZXF1ZXN0NjkxMDc5ODQy
2,447
Delete TODO.rst
{ "login": "alexpirine", "id": 1610035, "node_id": "MDQ6VXNlcjE2MTAwMzU=", "avatar_url": "https://avatars.githubusercontent.com/u/1610035?v=4", "gravatar_id": "", "url": "https://api.github.com/users/alexpirine", "html_url": "https://github.com/alexpirine", "followers_url": "https://api.github.com/users/alexpirine/followers", "following_url": "https://api.github.com/users/alexpirine/following{/other_user}", "gists_url": "https://api.github.com/users/alexpirine/gists{/gist_id}", "starred_url": "https://api.github.com/users/alexpirine/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/alexpirine/subscriptions", "organizations_url": "https://api.github.com/users/alexpirine/orgs", "repos_url": "https://api.github.com/users/alexpirine/repos", "events_url": "https://api.github.com/users/alexpirine/events{/privacy}", "received_events_url": "https://api.github.com/users/alexpirine/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I like to keep it as a placeholder in case I want to fill it in." ]
"2021-07-15T22:46:55"
"2021-07-16T00:35:08"
"2021-07-16T00:35:07"
NONE
null
Seems like it hasn't been used for two years.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2447/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2447/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2447", "html_url": "https://github.com/coleifer/peewee/pull/2447", "diff_url": "https://github.com/coleifer/peewee/pull/2447.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2447.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2446
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2446/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2446/comments
https://api.github.com/repos/coleifer/peewee/issues/2446/events
https://github.com/coleifer/peewee/issues/2446
945,654,400
MDU6SXNzdWU5NDU2NTQ0MDA=
2,446
This ORM doesn't seem to work anymore.
{ "login": "Evert-Arends", "id": 16082595, "node_id": "MDQ6VXNlcjE2MDgyNTk1", "avatar_url": "https://avatars.githubusercontent.com/u/16082595?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Evert-Arends", "html_url": "https://github.com/Evert-Arends", "followers_url": "https://api.github.com/users/Evert-Arends/followers", "following_url": "https://api.github.com/users/Evert-Arends/following{/other_user}", "gists_url": "https://api.github.com/users/Evert-Arends/gists{/gist_id}", "starred_url": "https://api.github.com/users/Evert-Arends/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Evert-Arends/subscriptions", "organizations_url": "https://api.github.com/users/Evert-Arends/orgs", "repos_url": "https://api.github.com/users/Evert-Arends/repos", "events_url": "https://api.github.com/users/Evert-Arends/events{/privacy}", "received_events_url": "https://api.github.com/users/Evert-Arends/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I want to apologize. This worked like a charm. My database tool however is completely fucked. Sorry for the inconvenience." ]
"2021-07-15T18:38:26"
"2021-07-15T19:04:51"
"2021-07-15T19:04:51"
NONE
null
I'm using Python 3.9.6, and the example provided in the readme is broken: ``` import datetime import peewee db = peewee.SqliteDatabase('my_database.db') class BaseModel(peewee.Model): class Meta: database = db class User(BaseModel): username = peewee.CharField(unique=True) class Tweet(BaseModel): user = peewee.ForeignKeyField(User, backref='tweets') message = peewee.TextField() created_date = peewee.DateTimeField(default=datetime.datetime.now) is_published = peewee.BooleanField(default=True) db.connect() db.create_tables([User, Tweet]) charlie = User.create(username='charlie') huey = User(username='huey') huey.save() # No need to set `is_published` or `created_date` since they # will just use the default values we specified. Tweet.create(user=charlie, message='My first tweet') ``` Accourding to the readme this code should be working? I would expect a database with 2 tables and some data in it. The db is created but empty.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2446/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2446/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2445
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2445/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2445/comments
https://api.github.com/repos/coleifer/peewee/issues/2445/events
https://github.com/coleifer/peewee/pull/2445
937,125,654
MDExOlB1bGxSZXF1ZXN0NjgzNzMxNDc3
2,445
Fixes bug in JSON IN query for iterables
{ "login": "dustinrb", "id": 3378450, "node_id": "MDQ6VXNlcjMzNzg0NTA=", "avatar_url": "https://avatars.githubusercontent.com/u/3378450?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dustinrb", "html_url": "https://github.com/dustinrb", "followers_url": "https://api.github.com/users/dustinrb/followers", "following_url": "https://api.github.com/users/dustinrb/following{/other_user}", "gists_url": "https://api.github.com/users/dustinrb/gists{/gist_id}", "starred_url": "https://api.github.com/users/dustinrb/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/dustinrb/subscriptions", "organizations_url": "https://api.github.com/users/dustinrb/orgs", "repos_url": "https://api.github.com/users/dustinrb/repos", "events_url": "https://api.github.com/users/dustinrb/events{/privacy}", "received_events_url": "https://api.github.com/users/dustinrb/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I don't like this patch for a couple of reasons, nor am I sure I understand the goal behind supporting such a query. Namely because it mutates the expression state when sql() is called, which is not done elsewhere in the codebase. Additionally, searching for an exact array within a json field seems problematic because there's no possibility of doing so efficiently from the database perspective.\r\n\r\nI'm going to pass on this patch, but will look at the original issue and give it some thought." ]
"2021-07-05T14:04:18"
"2021-07-05T14:57:19"
"2021-07-05T14:57:19"
NONE
null
Resolves issue #2444
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2445/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2445/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2445", "html_url": "https://github.com/coleifer/peewee/pull/2445", "diff_url": "https://github.com/coleifer/peewee/pull/2445.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2445.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2444
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2444/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2444/comments
https://api.github.com/repos/coleifer/peewee/issues/2444/events
https://github.com/coleifer/peewee/issues/2444
937,124,555
MDU6SXNzdWU5MzcxMjQ1NTU=
2,444
IN queries on fields with iterable Python representations
{ "login": "dustinrb", "id": 3378450, "node_id": "MDQ6VXNlcjMzNzg0NTA=", "avatar_url": "https://avatars.githubusercontent.com/u/3378450?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dustinrb", "html_url": "https://github.com/dustinrb", "followers_url": "https://api.github.com/users/dustinrb/followers", "following_url": "https://api.github.com/users/dustinrb/following{/other_user}", "gists_url": "https://api.github.com/users/dustinrb/gists{/gist_id}", "starred_url": "https://api.github.com/users/dustinrb/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/dustinrb/subscriptions", "organizations_url": "https://api.github.com/users/dustinrb/orgs", "repos_url": "https://api.github.com/users/dustinrb/repos", "events_url": "https://api.github.com/users/dustinrb/events{/privacy}", "received_events_url": "https://api.github.com/users/dustinrb/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I see why you closed #2445; it didn't make sense to single out the IN clause in that part of the code.\r\n\r\nI think the broader issue is that you cannot use any field for an IN query where it's Python representation is a sequence. You might be able to get around this by calling\r\n\r\n```python\r\n...where(KeyData.data.in_(KeyData.data.db_value(i) in [[1], [2], [1,2]]))...\r\n```\r\nbut that seems unintuitive from the user perspective. \r\n\r\nPerhaps the better solution would be to handle this in the `Field._in` with the understanding that if the user provides an iterable that each item has `Field.db_value` called on it before it's formed into an Expression.\r\n\r\nI'm happy to work on a patch. I'm just too unfamiliar with codebase to architect this.", "I've implemented a [fix](https://github.com/dustinrb/peewee/tree/json_in_query) that converts `multi_types` to Values in the Expression generated by the `.in_` function instead of mutating one SQL generation.\r\n\r\nThis process occurs outside of a Context block. Will that pose any issues the node tree when generating SQL? ", "I looked at the patch and it has the same problem.\r\n\r\nFor now, you can always try:\r\n\r\n```python\r\nvls = [Value(l, unpack=False) for l in list_of_lists]\r\n.where(KeyData.data.in_(vls))\r\n```", "Thanks for looking into this.\r\n\r\nMaybe I'm not sure what the exact issue is. Your interim fix is essentially what my second patch does in the Field class which happens before SQL generation. Are you saying that query input should be untouched before it's handed over to the SQL generating code? In that case it feels like a better solution would be to add a flag to allow shallow unpacking for a single Expression and/or NodeList.\r\n\r\nThanks for letting me satisfy my curiosity. Let me know if you'd like me to keep working on a fix.", "This is a bit of a strange one, so I'm going to close for now, but will revisit should it occur in another context." ]
"2021-07-05T14:03:13"
"2021-07-30T14:48:41"
"2021-07-30T14:48:41"
NONE
null
When using arrays in a JSON field, the IN operator unpacks all the values into a single list. For example: ```python itr_select_query = (KeyData .select() .where(KeyData.data.in_( [[1], [2], [1,2]] ))) ``` will generate the SQL query: ```python ('SELECT "t1"."id", "t1"."key", "t1"."data" FROM "key_data" AS "t1" WHERE ("t1"."data" IN ((json(?)), (json(?)), (json(?), json(?))))', ['1', '2', '1', '2']) ``` instead of: ```python ('SELECT "t1"."id", "t1"."key", "t1"."data" FROM "key_data" AS "t1" WHERE ("t1"."data" IN ((json(?)), (json(?)), (json(?), json(?))))', ['[1]', '[2]', '[1, 2]']) ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2444/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2444/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2443
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2443/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2443/comments
https://api.github.com/repos/coleifer/peewee/issues/2443/events
https://github.com/coleifer/peewee/pull/2443
936,477,112
MDExOlB1bGxSZXF1ZXN0NjgzMTg5MDU0
2,443
Fix decimal.InvalidOperation error for empty string
{ "login": "soyoung97", "id": 29880214, "node_id": "MDQ6VXNlcjI5ODgwMjE0", "avatar_url": "https://avatars.githubusercontent.com/u/29880214?v=4", "gravatar_id": "", "url": "https://api.github.com/users/soyoung97", "html_url": "https://github.com/soyoung97", "followers_url": "https://api.github.com/users/soyoung97/followers", "following_url": "https://api.github.com/users/soyoung97/following{/other_user}", "gists_url": "https://api.github.com/users/soyoung97/gists{/gist_id}", "starred_url": "https://api.github.com/users/soyoung97/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/soyoung97/subscriptions", "organizations_url": "https://api.github.com/users/soyoung97/orgs", "repos_url": "https://api.github.com/users/soyoung97/repos", "events_url": "https://api.github.com/users/soyoung97/events{/privacy}", "received_events_url": "https://api.github.com/users/soyoung97/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "FYI, this was the full stack trace for my application.\r\n```Traceback (most recent call last):\r\n File \"/usr/local/lib/python3.6/site-packages/flask/app.py\", line 2309, in __call__\r\n return self.wsgi_app(environ, start_response)\r\n File \"/usr/local/lib/python3.6/site-packages/flask/app.py\", line 2295, in wsgi_app\r\n response = self.handle_exception(e)\r\n File \"/usr/local/lib/python3.6/site-packages/flask/app.py\", line 1741, in handle_exception\r\n reraise(exc_type, exc_value, tb)\r\n File \"/usr/local/lib/python3.6/site-packages/flask/_compat.py\", line 35, in reraise\r\n raise value\r\n File \"/usr/local/lib/python3.6/site-packages/flask/app.py\", line 2292, in wsgi_app\r\n response = self.full_dispatch_request()\r\n File \"/usr/local/lib/python3.6/site-packages/flask/app.py\", line 1815, in full_dispatch_request\r\n rv = self.handle_user_exception(e)\r\n File \"/usr/local/lib/python3.6/site-packages/flask/app.py\", line 1718, in handle_user_exception\r\n reraise(exc_type, exc_value, tb)\r\n File \"/usr/local/lib/python3.6/site-packages/flask/_compat.py\", line 35, in reraise\r\n raise value\r\n File \"/usr/local/lib/python3.6/site-packages/flask/app.py\", line 1813, in full_dispatch_request\r\n rv = self.dispatch_request()\r\n File \"/usr/local/lib/python3.6/site-packages/flask/app.py\", line 1799, in dispatch_request\r\n return self.view_functions[rule.endpoint](**req.view_args)\r\n File \"/Users/soyoung/Desktop/coding/research/sql/sql-instruction-web/views.py\", line 121, in inner\r\n return fn(cur_db, table, *args, **kwargs)\r\n File \"/Users/soyoung/Desktop/coding/research/sql/sql-instruction-web/views.py\", line 179, in table_content\r\n total_pages=total_pages)\r\n File \"/usr/local/lib/python3.6/site-packages/flask/templating.py\", line 135, in render_template\r\n context, ctx.app)\r\n File \"/usr/local/lib/python3.6/site-packages/flask/templating.py\", line 117, in _render\r\n rv = template.render(context)\r\n File \"/usr/local/lib/python3.6/site-packages/jinja2/environment.py\", line 1090, in render\r\n self.environment.handle_exception()\r\n File \"/usr/local/lib/python3.6/site-packages/jinja2/environment.py\", line 832, in handle_exception\r\n reraise(*rewrite_traceback_stack(source=source))\r\n File \"/usr/local/lib/python3.6/site-packages/jinja2/_compat.py\", line 28, in reraise\r\n raise value.with_traceback(tb)\r\n File \"/Users/soyoung/Desktop/coding/research/sql/sql-instruction-web/templates/table_content.html\", line 1, in top-level template code\r\n {% extends \"base_table.html\" %}\r\n File \"/Users/soyoung/Desktop/coding/research/sql/sql-instruction-web/templates/base_table.html\", line 1, in top-level template code\r\n {% extends \"base_tables.html\" %}\r\n File \"/Users/soyoung/Desktop/coding/research/sql/sql-instruction-web/templates/base_tables.html\", line 1, in top-level template code\r\n {% extends \"base.html\" %}\r\n File \"/Users/soyoung/Desktop/coding/research/sql/sql-instruction-web/templates/base.html\", line 42, in top-level template code\r\n {% block content %}{% endblock %}\r\n File \"/Users/soyoung/Desktop/coding/research/sql/sql-instruction-web/templates/base_table.html\", line 29, in block \"content\"\r\n {% block inner_content %}\r\n File \"/Users/soyoung/Desktop/coding/research/sql/sql-instruction-web/templates/table_content.html\", line 36, in block \"inner_content\"\r\n {% for row in query %}\r\n File \"/usr/local/lib/python3.6/site-packages/peewee.py\", line 4378, in next\r\n self.cursor_wrapper.iterate()\r\n File \"/usr/local/lib/python3.6/site-packages/peewee.py\", line 4297, in iterate\r\n result = self.process_row(row)\r\n File \"/usr/local/lib/python3.6/site-packages/peewee.py\", line 7450, in process_row\r\n result[attr] = converters[i](row[i])\r\n File \"/usr/local/lib/python3.6/site-packages/peewee.py\", line 4681, in python_value\r\n return decimal.Decimal(text_type(value))\r\ndecimal.InvalidOperation: [<class 'decimal.ConversionSyntax'>]```", "So sqlite does not have an actual dedicated Decimal field. It interprets the value as numeric, I believe, but it looks like in your table you have empty string (`''`) instead of NULL or 0. Python cannot convert empty string into a decimal, so you get the error.\r\n\r\nTo fix this either:\r\n\r\n1. update all decimal values stored as '' to be 0\r\n2. update all decimal values stored as '' to be NULL\r\n3. use a different field type or different column storage.", "For what it's worth, it looks like @nitinprakash96 copy/pasted a bit of code directly from my project `sqlite-web`. I don't know what the deal is, but it looks like he's created a shittier version of my project. You might have better luck using it:\r\n\r\nhttps://github.com/coleifer/sqlite-web", "Oh I didn't know that there was original code! thanks for the pointer! I'll change my codebase to https://github.com/coleifer/sqlite-web.\r\nBy the way, (I have to test whether this problem also happens or not after changing the codebase but if so,) I just thought that it'll be great if we can treat this exception at the peewee library level, since currently I cannot update the original database (It is given to me as official training data). However thanks for the solution and fast response!", "I would like to be able to fix at the library level, but it seems like a fairly particular edge-case arising from a couple different problems:\r\n\r\n1. the use of DECIMAL type in sqlite (sqlite is weakly-typed and inconsistent data values can be a problem. See: https://www.sqlite.org/datatype3.html#type_affinity )\r\n2. storing empty string in a numeric column instead of NULL or 0\r\n3. peewee conversion logic for non-null decimal values\r\n\r\nLike I said, I think there's a couple issues in play, but my suggestion would be to fix-up the data if you can.", "Thanks so much for additional pointer. I'm not an expert at sqlite, so I didn't know about this issue. Now I fully understand that this is an edge case, and it should be better treated at the database level. Thanks again for your consideration! I really appreciate it :)" ]
"2021-07-04T15:04:28"
"2021-07-06T03:20:34"
"2021-07-04T22:36:20"
NONE
null
Hello, thank you for making such a good opensource code! I was working with [sqlite db viewer](https://github.com/nitinprakash96/sqlite-db-viewer) that uses peewee. However, when I was trying to iterate through None values at integer field (by {% for row in query %} in html code), there was a peewee error saying decimal.InvalidOperation. I think this issue could also be related with [this pull request](https://github.com/coleifer/peewee/commit/cd07f4e01a1262166d8944736e11d0681fb76214), since the error appeared at returning value of the function python_value `decimal.Decimal(text_type(value))`. After inspection, I found out that sometimes an empty string('') comes into this function(which it shouldn't since empty string is not a number) and by adding a filter (adding `and row[i] != ''` at ModelDictCursorWrapper), I managed to solve this problem. *Please note that I am new to this code and I may be not understanding this perfectly, so please correct me if I'm wrong in solving this problem. Attatched below is my original code, error traceback, and pdb operations. Thank you! ### Error Traceback ![image](https://user-images.githubusercontent.com/29880214/124389717-5d123400-dd23-11eb-8370-e1d7925abe36.png) ![image](https://user-images.githubusercontent.com/29880214/124389742-8206a700-dd23-11eb-9c39-d4fdae1dd057.png) ### Setting try-catch pdb and results at python_value at peewee.py ![image](https://user-images.githubusercontent.com/29880214/124389767-a3679300-dd23-11eb-88b8-59d465eeaac7.png) ![image](https://user-images.githubusercontent.com/29880214/124389796-c09c6180-dd23-11eb-9b3c-1e28543263c0.png) ### Filtering String values ![image](https://user-images.githubusercontent.com/29880214/124389841-f2152d00-dd23-11eb-9130-c81ee1a011d2.png) ### Correctly displayed application (mine) ![image](https://user-images.githubusercontent.com/29880214/124389888-2557bc00-dd24-11eb-9451-eaa909f3e86f.png) ![image](https://user-images.githubusercontent.com/29880214/124389909-45877b00-dd24-11eb-9075-8963167d886a.png) Here, I think what caused the problem is the ibb, hbp, sh, sf, g_idp column, which has Nonetype values as well as numbers(0). Adding this code to handle exceptions for this case would be great!
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2443/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2443/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2443", "html_url": "https://github.com/coleifer/peewee/pull/2443", "diff_url": "https://github.com/coleifer/peewee/pull/2443.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2443.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2442
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2442/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2442/comments
https://api.github.com/repos/coleifer/peewee/issues/2442/events
https://github.com/coleifer/peewee/issues/2442
935,515,520
MDU6SXNzdWU5MzU1MTU1MjA=
2,442
how should i use ModelSelect.select()?
{ "login": "CooperWanng", "id": 45349258, "node_id": "MDQ6VXNlcjQ1MzQ5MjU4", "avatar_url": "https://avatars.githubusercontent.com/u/45349258?v=4", "gravatar_id": "", "url": "https://api.github.com/users/CooperWanng", "html_url": "https://github.com/CooperWanng", "followers_url": "https://api.github.com/users/CooperWanng/followers", "following_url": "https://api.github.com/users/CooperWanng/following{/other_user}", "gists_url": "https://api.github.com/users/CooperWanng/gists{/gist_id}", "starred_url": "https://api.github.com/users/CooperWanng/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/CooperWanng/subscriptions", "organizations_url": "https://api.github.com/users/CooperWanng/orgs", "repos_url": "https://api.github.com/users/CooperWanng/repos", "events_url": "https://api.github.com/users/CooperWanng/events{/privacy}", "received_events_url": "https://api.github.com/users/CooperWanng/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "If you are explicitly re-selecting columns, you need to specify which columns you want.", "> If you are explicitly re-selecting columns, you need to specify which columns you want.\r\n\r\nthanks,the usage of peewee differs between the two versions(3.6->3.14)" ]
"2021-07-02T07:52:58"
"2021-07-11T13:58:45"
"2021-07-02T12:59:02"
NONE
null
I ran into a problem when I used select function of ModelSelect >query = Auth.select(Auth.id) # Model.select() select t1.id FROM auth AS t1; query = Auth.select(Auth.id).select() # ModelSelect.select() > select FROM auth AS t1; how should i do?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2442/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2442/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2441
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2441/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2441/comments
https://api.github.com/repos/coleifer/peewee/issues/2441/events
https://github.com/coleifer/peewee/pull/2441
935,050,691
MDExOlB1bGxSZXF1ZXN0NjgyMDMxOTUw
2,441
Add reuse_if_open flag to playhouse.dataset.DataSet
{ "login": "Xezed", "id": 22517386, "node_id": "MDQ6VXNlcjIyNTE3Mzg2", "avatar_url": "https://avatars.githubusercontent.com/u/22517386?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Xezed", "html_url": "https://github.com/Xezed", "followers_url": "https://api.github.com/users/Xezed/followers", "following_url": "https://api.github.com/users/Xezed/following{/other_user}", "gists_url": "https://api.github.com/users/Xezed/gists{/gist_id}", "starred_url": "https://api.github.com/users/Xezed/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Xezed/subscriptions", "organizations_url": "https://api.github.com/users/Xezed/orgs", "repos_url": "https://api.github.com/users/Xezed/repos", "events_url": "https://api.github.com/users/Xezed/events{/privacy}", "received_events_url": "https://api.github.com/users/Xezed/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[]
"2021-07-01T17:09:50"
"2021-07-01T20:35:45"
"2021-07-01T20:35:45"
NONE
null
Hey :) I need to use the existing connection with `DataSet`. This pull request will allow me to do so. Thank you for your work on the project. :rays_of_love:
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2441/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2441/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2441", "html_url": "https://github.com/coleifer/peewee/pull/2441", "diff_url": "https://github.com/coleifer/peewee/pull/2441.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2441.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2440
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2440/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2440/comments
https://api.github.com/repos/coleifer/peewee/issues/2440/events
https://github.com/coleifer/peewee/issues/2440
934,606,810
MDU6SXNzdWU5MzQ2MDY4MTA=
2,440
[QueryBuilder] Missing alias when filtering on a nested subquery
{ "login": "kozlek", "id": 3019422, "node_id": "MDQ6VXNlcjMwMTk0MjI=", "avatar_url": "https://avatars.githubusercontent.com/u/3019422?v=4", "gravatar_id": "", "url": "https://api.github.com/users/kozlek", "html_url": "https://github.com/kozlek", "followers_url": "https://api.github.com/users/kozlek/followers", "following_url": "https://api.github.com/users/kozlek/following{/other_user}", "gists_url": "https://api.github.com/users/kozlek/gists{/gist_id}", "starred_url": "https://api.github.com/users/kozlek/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/kozlek/subscriptions", "organizations_url": "https://api.github.com/users/kozlek/orgs", "repos_url": "https://api.github.com/users/kozlek/repos", "events_url": "https://api.github.com/users/kozlek/events{/privacy}", "received_events_url": "https://api.github.com/users/kozlek/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Have you tried attaching an alias to your subquery? Here's my code and the output, which looks OK to me:\r\n\r\n```python\r\n# Get product IDs of best-selling products.\r\nbp = (Sales\r\n .select(Sales.c.product_id, fn.SUM(Sales.c.price).alias('price__sum'))\r\n .group_by(Sales.c.product_id)\r\n .alias('bp')) # NB\r\n\r\n# Select 10 best-selling products.\r\ntop10 = (Select([bp], [bp.c.product_id, bp.c.price__sum])\r\n .order_by(bp.c.price__sum.desc())\r\n .limit(10))\r\n\r\nprod = (Product\r\n .select(Product.c.product_id, Product.c.title)\r\n .where(Product.c.product_id.in_(top10)))\r\n\r\nprint(prod.sql())\r\n```\r\n\r\nOutput:\r\n\r\n```sql\r\nSELECT \"t1\".\"product_id\", \"t1\".\"title\" \r\nFROM \"product\" AS \"t1\"\r\nWHERE (\"t1\".\"product_id\" IN (\r\n SELECT \"bp\".\"product_id\", \"bp\".\"price__sum\" \r\n FROM (\r\n SELECT \"t2\".\"product_id\", SUM(\"t2\".\"price\") AS \"price__sum\" \r\n FROM \"sales\" AS \"t2\" \r\n GROUP BY \"t2\".\"product_id\"\r\n ) AS \"bp\" \r\n ORDER BY \"bp\".\"price__sum\" DESC \r\n LIMIT ?)\r\n)\r\n```", "Another thing you might try, which seems a bit more idiomatic, would be to use a common table expression for the top 10 products:\r\n\r\n```python\r\ntop10 = (Sales\r\n .select(Sales.c.product_id, fn.SUM(Sales.c.price))\r\n .group_by(Sales.c.product_id)\r\n .order_by(fn.SUM(Sales.c.price).desc())\r\n .limit(10)\r\n .cte('top10', columns=('product_id', 'price__sum')))\r\n\r\nquery = (Product\r\n .select(Product.c.id, Product.c.title)\r\n .join(top10, on=(Product.c.id == top10.c.product_id))\r\n .with_cte(top10))\r\n```\r\n\r\nSQL:\r\n\r\n```sql\r\n\r\nWITH \"top10\" (\"product_id\", \"price__sum\") AS (\r\n SELECT \"t1\".\"product_id\", SUM(\"t1\".\"price\") \r\n FROM \"sales\" AS \"t1\" \r\n GROUP BY \"t1\".\"product_id\" \r\n ORDER BY SUM(\"t1\".\"price\") DESC \r\n LIMIT ?) \r\nSELECT \"t2\".\"id\", \"t2\".\"title\" FROM \"products\" AS \"t2\" \r\nINNER JOIN \"top10\" ON (\"t2\".\"id\" = \"top10\".\"product_id\")\r\n```", "Thanks for your answers 🙂\r\n\r\nAttaching aliases seems to work on my simple examples, I'll try to implement that on the complex ones now and I will stick \r\nwith that if it works well 👌\r\nIf I manage to find patch, I will still try to fix auto alias attachement as it is very practical in some cases 🙂\r\n\r\nI would love to use CTE like I'm used to with Postgres, but here I'm working with a Clickhouse OLAP data warehouse which has a limited support for CTE...\r\nAt the time I'm writing these lines, there is no suitable query builder on the python market that supports Clickhouse specificity out of the box. Starting with Peewee's query builder and code some extensions to support Clickhouse flavoured operations (like JOIN USING) was the easiest solution ; way better than restarting a whole query builder from scratch.\r\n\r\nThanks again for your work on Peewee 👏" ]
"2021-07-01T09:08:27"
"2021-07-02T13:37:33"
"2021-07-01T13:23:25"
NONE
null
Hi ! I'm using peewee's QueryBuilder API these days and I'm enjoying its powerful capabilities 😃 However, I have a complex use case where I need to produce a `WHERE IN <subquery>` condition ; where `subquery` is itself a subquery (nested subquery). ## Issue I've tried to recreate a minimal example. Let's say we have two tables: - product(product_id, title) - sales(product_id, price) Now let's request the title of the 10 best-seller products. We expect to have the following SQL: ```sql SELECT "t1"."product_id" AS "product_id", "t1"."title" AS "title" FROM "product" AS "t1" WHERE ("t1"."product_id" IN ( SELECT "t2"."product_id" AS "product_id", "t2"."price__sum" AS "price__sum" FROM ( SELECT "t3"."product_id" AS "product_id", sum("t3"."price") AS "price__sum" FROM "sales" AS "t3" GROUP BY "t3"."product_id" ) AS "t2" ORDER BY "t2"."price__sum" DESC LIMIT 10 ) ) ``` I'm aware we can write the following query in a totally different way, but I'm working with a complex system that needs to have the query written this way (that's why I'm using the QueryBuilder rather than the ORM). Using peewee's QueryBuilder API, we'll do something like this: ```python import peewee # let's define 2 tables t_product = peewee.Table("product") t_sales = peewee.Table("sales") # create a first subquery that computes the sum of the sales subquery_best_products = t_sales.select( t_sales.c.product_id.alias("product_id"), peewee.fn.sum(t_sales.c.price).alias("price__sum") ).group_by(t_sales.c.product_id) # subquery from this previous subquery to order_by and limit 10 nested_subquery_best_products = ( subquery_best_products.select_from( subquery_best_products.c.product_id.alias("product_id"), subquery_best_products.c.price__sum.alias("price__sum"), ) .order_by(subquery_best_products.c.price__sum.desc()) .limit(10) ) # select the title of the 10 best products query = t_product.select( t_product.c.product_id.alias("product_id"), t_product.c.title.alias("title") ).where(t_product.c.product_id.in_(nested_subquery_best_products)) ``` This issue is that `query` will compile into the following: ```sql SELECT "t1"."product_id" AS "product_id", "t1"."title" AS "title" FROM "product" AS "t1" WHERE ("t1"."product_id" IN ( SELECT "t2"."product_id" AS "product_id", "t2"."price__sum" AS "price__sum" FROM ( SELECT "t3"."product_id" AS "product_id", sum("t3"."price") AS "price__sum" FROM "sales" AS "t3" GROUP BY "t3"."product_id" ) ORDER BY "t2"."price__sum" DESC LIMIT 10 ) ) ``` We miss the `t2` alias definition, while it is used in the returning section of the generated SQL ! To ensure that our `nested_subquery_best_products` is correct, we can compile it standalone: ```sql SELECT "t1"."product_id" AS "product_id", "t1"."price__sum" AS "price__sum" FROM ( SELECT "t2"."product_id" AS "product_id", sum("t2"."price") AS "price__sum" FROM "sales" AS "t2" GROUP BY "t2"."product_id" ) AS "t1" ORDER BY "t1"."price__sum" DESC LIMIT 10 ``` The aliases are a bit different (that's normal), but they are correctly defined. So the issue might be scoped to the usage of a nested subquery into a where expression. ## Solution (WIP) I'm digging in peewee's internals and I figured out aliases are applied conditionally depending of the scope of the current context. I guess that when the nested subquery is run in standalone mode, the scope is correctly set to SCOPE_SOURCE, while it is set to SCOPE_NORMAL when compiling where expression. I'll be happy to provide a MR if I manage to find a working patch 👌 Thanks in advance for your help 🙏
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2440/reactions", "total_count": 6, "+1": 5, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 1 }
https://api.github.com/repos/coleifer/peewee/issues/2440/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2439
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2439/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2439/comments
https://api.github.com/repos/coleifer/peewee/issues/2439/events
https://github.com/coleifer/peewee/issues/2439
933,280,236
MDU6SXNzdWU5MzMyODAyMzY=
2,439
migrate leave foreignkey out of consideration
{ "login": "woshimanong1990", "id": 13585117, "node_id": "MDQ6VXNlcjEzNTg1MTE3", "avatar_url": "https://avatars.githubusercontent.com/u/13585117?v=4", "gravatar_id": "", "url": "https://api.github.com/users/woshimanong1990", "html_url": "https://github.com/woshimanong1990", "followers_url": "https://api.github.com/users/woshimanong1990/followers", "following_url": "https://api.github.com/users/woshimanong1990/following{/other_user}", "gists_url": "https://api.github.com/users/woshimanong1990/gists{/gist_id}", "starred_url": "https://api.github.com/users/woshimanong1990/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/woshimanong1990/subscriptions", "organizations_url": "https://api.github.com/users/woshimanong1990/orgs", "repos_url": "https://api.github.com/users/woshimanong1990/repos", "events_url": "https://api.github.com/users/woshimanong1990/events{/privacy}", "received_events_url": "https://api.github.com/users/woshimanong1990/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "The foreign keys are returned by `pragma foreign_key_list(\"...\")`. I'm not sure I understand the issue you're describing." ]
"2021-06-30T03:55:50"
"2021-06-30T13:17:55"
"2021-06-30T13:17:55"
NONE
null
I drop a field to a tabel ,but not I success. Finally, I found, it's a bug. you create a tmp table and then, drop old tabel. BUT, leave foreignkey out of consideration ` # Find any foreign keys we may need to remove. self.database.get_foreign_keys(table) ` do nothing I used sqlite. maybe I should use `PRAGMA foreign_keys=OFF` first
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2439/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2439/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2438
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2438/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2438/comments
https://api.github.com/repos/coleifer/peewee/issues/2438/events
https://github.com/coleifer/peewee/issues/2438
933,265,316
MDU6SXNzdWU5MzMyNjUzMTY=
2,438
migrator.add_column throws error on column with default value for sqlite
{ "login": "binh-vu", "id": 4346739, "node_id": "MDQ6VXNlcjQzNDY3Mzk=", "avatar_url": "https://avatars.githubusercontent.com/u/4346739?v=4", "gravatar_id": "", "url": "https://api.github.com/users/binh-vu", "html_url": "https://github.com/binh-vu", "followers_url": "https://api.github.com/users/binh-vu/followers", "following_url": "https://api.github.com/users/binh-vu/following{/other_user}", "gists_url": "https://api.github.com/users/binh-vu/gists{/gist_id}", "starred_url": "https://api.github.com/users/binh-vu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/binh-vu/subscriptions", "organizations_url": "https://api.github.com/users/binh-vu/orgs", "repos_url": "https://api.github.com/users/binh-vu/repos", "events_url": "https://api.github.com/users/binh-vu/events{/privacy}", "received_events_url": "https://api.github.com/users/binh-vu/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Closing the issue. I found the root of this problem.\r\n\r\nI added the new field to my model definition with index. When I run my migration script, peewee saw a new field, it doesn't create the column. However, it creates the index. So the database has an index of an unknown column. A query to create column is fine, but an update query to that column throws error in python. Dropping the index and it works as expected." ]
"2021-06-30T03:13:57"
"2021-06-30T04:03:10"
"2021-06-30T04:03:10"
NONE
null
Peewee version: `3.14.4` SQLite version: `3.33.0` When I run, ``` migrator = SqliteMigrator(db) with db.atomic(): migrate( migrator.add_column("table", 'is_deleted', BooleanField(default=True)), ) ``` It throws the following error: ``` peewee.DatabaseError: database disk image is malformed ``` This behavior contradicts to the [documentation](https://docs.peewee-orm.com/en/latest/peewee/playhouse.html#supported-operations) I investigated more and found that: * When the table is empty, it executes this query: `ALTER TABLE "table" ADD COLUMN "is_deleted" INTEGER` * However, then the table is not empty, it executes this query: `UPDATE "table" SET "is_deleted" = ?`. Manually executing the query on my sqlite database, it throws the error: `Result: no such column: is_deleted`. * I dump the content of the database to an sql file and re-import them on new database but still get the same error. * Interestingly, `add_column` (nullable), then `apply_default` also doesn't work and I got the same error. Turns out the query `UPDATE ...` causing the issue. So I need to do it in one query `ALTER TABLE "table" ADD COLUMN "is_deleted" INTEGER NOT NULL DEFAULT 0`.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2438/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2438/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2437
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2437/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2437/comments
https://api.github.com/repos/coleifer/peewee/issues/2437/events
https://github.com/coleifer/peewee/issues/2437
932,902,106
MDU6SXNzdWU5MzI5MDIxMDY=
2,437
How do I union queries where the model contains a JSON field?
{ "login": "dsmurrell", "id": 4035854, "node_id": "MDQ6VXNlcjQwMzU4NTQ=", "avatar_url": "https://avatars.githubusercontent.com/u/4035854?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dsmurrell", "html_url": "https://github.com/dsmurrell", "followers_url": "https://api.github.com/users/dsmurrell/followers", "following_url": "https://api.github.com/users/dsmurrell/following{/other_user}", "gists_url": "https://api.github.com/users/dsmurrell/gists{/gist_id}", "starred_url": "https://api.github.com/users/dsmurrell/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/dsmurrell/subscriptions", "organizations_url": "https://api.github.com/users/dsmurrell/orgs", "repos_url": "https://api.github.com/users/dsmurrell/repos", "events_url": "https://api.github.com/users/dsmurrell/events{/privacy}", "received_events_url": "https://api.github.com/users/dsmurrell/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Doing a bit of searching, it looks like the UNION operation is trying to take out duplicates and needs to compare the JSON content... is there any way in the union from peewee that I can specify that I'd like only apply distinct on say `id`?", "Yes, you need to use UNION ALL if you don't want duplicates removed (which is why you're getting that error). You can use `.union_all()` or the addition operator overload.", "Thanks, I came across that option in my searches.\r\n\r\nSo you're saying that if I am concerned about duplicates, I can `.union_all()`, and then use `.distinct(model.id)`?\r\n\r\nI was typing this out below before you responded but then I got consumed by something else:\r\n\r\n\"Trying out `BinaryJSONField` instead of `JSONField` now and this seems to work (using union and BinaryJSONFields), but I'm not sure what the implications are of changing our whole schema to use `BinaryJSONField` instead of `JSONField` everywhere. Are these supposed to be interchangeable or will changing the schema change the data in some way?\"\r\n\r\nI'd imagine the we don't want to be doing JSON equality operations anyway to filter distinct records so I'll try out using distinct after union_all. Thanks again for the help!", "This is a part of the SQL specification, that UNION must perform some type of equality to remove duplicates, whereas UNION ALL is not constrained in that way." ]
"2021-06-29T16:35:24"
"2021-06-29T18:53:32"
"2021-06-29T17:57:26"
NONE
null
If I have two or more queries to join (say q1, q2 and q3) and I'm using set operations on them like: `query = q1 & q2` or `query = q2 | q3` or `query = q1 - q3` I know that there is a limitation that the selected columns for these should be the same, but I'm finding that I get the errors: `peewee.ProgrammingError: could not identify an equality operator for type json` `psycopg2.errors.UndefinedFunction: could not identify an equality operator for type json` whenever all fields are selected and the model includes a `JSONField` from `playhouse.postgres_ext` Is this any way to resolve this? My use case is that I'm passing a list of queries through to another function where I want to union them and so I don't have the luxury of adding `.where(xyz)` statements onto an exiting query to limit the selection using multiple where filters. Also, the set operations seem more powerful in general as passing a list of where params is quite limiting.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2437/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2437/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2436
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2436/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2436/comments
https://api.github.com/repos/coleifer/peewee/issues/2436/events
https://github.com/coleifer/peewee/issues/2436
931,789,027
MDU6SXNzdWU5MzE3ODkwMjc=
2,436
What you think about rename to "peewo"?
{ "login": "Timtaran", "id": 60805088, "node_id": "MDQ6VXNlcjYwODA1MDg4", "avatar_url": "https://avatars.githubusercontent.com/u/60805088?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Timtaran", "html_url": "https://github.com/Timtaran", "followers_url": "https://api.github.com/users/Timtaran/followers", "following_url": "https://api.github.com/users/Timtaran/following{/other_user}", "gists_url": "https://api.github.com/users/Timtaran/gists{/gist_id}", "starred_url": "https://api.github.com/users/Timtaran/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Timtaran/subscriptions", "organizations_url": "https://api.github.com/users/Timtaran/orgs", "repos_url": "https://api.github.com/users/Timtaran/repos", "events_url": "https://api.github.com/users/Timtaran/events{/privacy}", "received_events_url": "https://api.github.com/users/Timtaran/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[]
"2021-06-28T17:16:07"
"2021-06-28T17:22:28"
"2021-06-28T17:22:28"
NONE
null
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2436/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2436/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2435
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2435/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2435/comments
https://api.github.com/repos/coleifer/peewee/issues/2435/events
https://github.com/coleifer/peewee/issues/2435
931,134,143
MDU6SXNzdWU5MzExMzQxNDM=
2,435
Some obstacles when trying to add a driver for ClickHouse
{ "login": "pm5", "id": 119645, "node_id": "MDQ6VXNlcjExOTY0NQ==", "avatar_url": "https://avatars.githubusercontent.com/u/119645?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pm5", "html_url": "https://github.com/pm5", "followers_url": "https://api.github.com/users/pm5/followers", "following_url": "https://api.github.com/users/pm5/following{/other_user}", "gists_url": "https://api.github.com/users/pm5/gists{/gist_id}", "starred_url": "https://api.github.com/users/pm5/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pm5/subscriptions", "organizations_url": "https://api.github.com/users/pm5/orgs", "repos_url": "https://api.github.com/users/pm5/repos", "events_url": "https://api.github.com/users/pm5/events{/privacy}", "received_events_url": "https://api.github.com/users/pm5/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I've fixed the first item, thank you for bringing that to my attention.\r\n\r\nThe second item will not be so easy, since it implies that you are passing the parameters as a `dict`. Every other driver I know of just uses a tuple to represent the query parameters, so there is probably some work involved in getting this working. It looks like this driver is implemented very idiosyncratically compared to the others, e.g.\r\n\r\n```\r\n>>> cursor.executemany('INSERT INTO test (x) VALUES', [[200]])\r\n```\r\n\r\nWhat the heck is going on here? Is the driver inserting placeholders?\r\n\r\nI'd suggest taking it up with the clickhouse implementer(s) to see about getting this resolved upstream, as it will likely be rather tricky with Peewee." ]
"2021-06-28T03:38:57"
"2021-06-28T12:50:43"
"2021-06-28T12:50:43"
NONE
null
I saw #912 and tried to add a driver for [ClickHouse](https://clickhouse.tech/) using [clickhouse-driver](https://clickhouse-driver.readthedocs.io/en/latest/features.html#python-db-api-2-0) but there seems to be some obstacles: - The [documentation](http://docs.peewee-orm.com/en/latest/peewee/database.html#adding-a-new-database-driver) seems outdated. `Database._connect` is no longer taking arguments and one has to use `foodb.connect(self.database, **self.connect_params)` instead. - The DB API 2.0 implemented by clickhouse-driver specifically uses Python extended format codes for SQL parameters, i.e. `%(name)s`, which is [allowed](https://www.python.org/dev/peps/pep-0249/#paramstyle). I've found `Database.param` in peewee, but it seems a bit hard to get it work with this format. Any suggestions?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2435/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2435/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2434
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2434/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2434/comments
https://api.github.com/repos/coleifer/peewee/issues/2434/events
https://github.com/coleifer/peewee/issues/2434
930,942,716
MDU6SXNzdWU5MzA5NDI3MTY=
2,434
time generation on linux vs windows
{ "login": "naamlev", "id": 66517559, "node_id": "MDQ6VXNlcjY2NTE3NTU5", "avatar_url": "https://avatars.githubusercontent.com/u/66517559?v=4", "gravatar_id": "", "url": "https://api.github.com/users/naamlev", "html_url": "https://github.com/naamlev", "followers_url": "https://api.github.com/users/naamlev/followers", "following_url": "https://api.github.com/users/naamlev/following{/other_user}", "gists_url": "https://api.github.com/users/naamlev/gists{/gist_id}", "starred_url": "https://api.github.com/users/naamlev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/naamlev/subscriptions", "organizations_url": "https://api.github.com/users/naamlev/orgs", "repos_url": "https://api.github.com/users/naamlev/repos", "events_url": "https://api.github.com/users/naamlev/events{/privacy}", "received_events_url": "https://api.github.com/users/naamlev/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I think you are misusing `prefetch()`. In your first example, you shouldn't need prefetch at all if each X has only one (or zero) Y / Z. Just write:\r\n\r\n```python\r\n(X\r\n .select(X, Y, Z)\r\n .join_from(X, Y, JOIN.LEFT_OUTER)\r\n .join_from(X, Z, JOIN.LEFT_OUTER))\r\n```\r\n\r\nFor your new version, since a Z can have many T's (if I'm reading this correctly), then yes you would probably be best off to use `prefetch()`. Again, though, you may be able to reuse the above snippet for the X, Y, and Z portions and just use prefetch for the T's.\r\n\r\nI cannot really advise on how to optimize this, it depends so much on how many rows of X, Y, Z and T you are getting, what database you are using, etc, etc. It might be possible to use an aggregation to roll-up the related T's and skip using prefetch altogether.\r\n\r\nAre you able to express the entire query in a single SQL SELECT (using plain old sql)? If yes, then we can probably optimize this. If no, then it will be hard to say.\r\n\r\nI'll reopen this if there is a bug, but feel free to respond and I will do my best to help you." ]
"2021-06-27T14:19:49"
"2021-06-27T14:56:44"
"2021-06-27T14:56:43"
NONE
null
Hey , I am using peewee 3.13.2 Part of the project include using join over multiple tables and retrieving foreign keys using prefetch. I've recently added to the join new table that has many to many relationship and prefetch to retrieve it key. **example:** _old version:_ X.select(X) .join_from(X, Y, peewee.JOIN.LEFT_OUTER) .join_from(X, Z, peewee.JOIN.LEFT_OUTER) .where(query) .prefetch(Y, Z) _new version:_ X.select(X) .join_from(X, Y, peewee.JOIN.LEFT_OUTER) .join_from(X, Z, peewee.JOIN.LEFT_OUTER) .join_from(Z, T, peewee.JOIN.LEFT_OUTER) .switch(X) .where(query) .prefetch(Y, Z,T) **questions** 1. is there a better why (lower time complexity) to model this behavior without limiting the query that the user can give? 2. why do i get better time complexity in windows? * The same branch (and same test) get on windows: 340 sec and in linux:870 sec * Just for comparison before the change i've got windows:200 sec linux: 250 sec
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2434/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2434/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2433
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2433/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2433/comments
https://api.github.com/repos/coleifer/peewee/issues/2433/events
https://github.com/coleifer/peewee/issues/2433
930,723,091
MDU6SXNzdWU5MzA3MjMwOTE=
2,433
Prefetching from multiple models
{ "login": "jonathanmach", "id": 12788052, "node_id": "MDQ6VXNlcjEyNzg4MDUy", "avatar_url": "https://avatars.githubusercontent.com/u/12788052?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jonathanmach", "html_url": "https://github.com/jonathanmach", "followers_url": "https://api.github.com/users/jonathanmach/followers", "following_url": "https://api.github.com/users/jonathanmach/following{/other_user}", "gists_url": "https://api.github.com/users/jonathanmach/gists{/gist_id}", "starred_url": "https://api.github.com/users/jonathanmach/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/jonathanmach/subscriptions", "organizations_url": "https://api.github.com/users/jonathanmach/orgs", "repos_url": "https://api.github.com/users/jonathanmach/repos", "events_url": "https://api.github.com/users/jonathanmach/events{/privacy}", "received_events_url": "https://api.github.com/users/jonathanmach/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "You don't need prefetch. Prefetch is for when the direction of traversal is reversed, e.g. you have a list of users and you want to get all their kits as well.\r\n\r\nJust do joins (you'll need aliases because you reference User multiple times):\r\n\r\n```python\r\n\r\nOrderUser = User.alias()\r\nOrderApprover = User.alias()\r\nKitClaimer = User.alias()\r\nKitCanceller = User.alias()\r\nq = (Kit.select(Kit, Order, OrderUser, OrderApprover, KitClaimer, KitCanceller)\r\n .join_from(Kit, Order)\r\n .join_from(Order, OrderUser, JOIN.INNER, on=Order.user)\r\n .join_from(Order, OrderApprover, JOIN.INNER, on=Order.approved_by_user)\r\n .join_from(Kit, KitClaimer, JOIN.LEFT_OUTER, on=Kit.claimed_by_user)\r\n .join_from(Kit, KitCanceller, JOIN.LEFT_OUTER, on=Kit.cancelled_by_user))\r\n```" ]
"2021-06-26T15:54:47"
"2021-06-28T15:46:49"
"2021-06-26T21:51:42"
CONTRIBUTOR
null
Hi all! Hoping everyone is safe! I've read the documentation about N+1 queries and prefetch, but I'm facing a strange behavior. Let's say I have the following models: ``` Users Orders user (FK) approved_by_user (FK) Kits order (FK) claimed_by_user (FK) cancelled_by_user (FK) ``` ```python # I can query for Kits and prefetch Users: for kit in prefetch(Kits.select(), Users): print(kit.claimed_by_user) print(kit.cancelled_by_user) # 👍 Works fine, no N+1 queries 🎉 --- # or query for Kits and prefetch Orders: for kit in prefetch(Kits.select(), Orders): print(kit.order) # 👍 Works fine, no N+1 queries 🎉 --- # But I can't prefetch both Users and Orders: for kit in prefetch(Kits.select(), Orders, Users): print(kit.claimed_by_user) print(kit.cancelled_by_user) print(kit.order) # It generates N+1 queries here❗️ 😨😨 ``` I imagine I'm missing some small detail here. My main goal is to respond to a REST request with a JSON in the following format: ``` [ Kits claimed_by_user : {...} cancelled_by_user : {...} order : {user: {...}, approved_by_user: {...}} ] ``` Thanks in advance!
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2433/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2433/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2432
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2432/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2432/comments
https://api.github.com/repos/coleifer/peewee/issues/2432/events
https://github.com/coleifer/peewee/issues/2432
928,623,933
MDU6SXNzdWU5Mjg2MjM5MzM=
2,432
Question: Is there a way to use one field to set another field when the create method is used?
{ "login": "dsmurrell", "id": 4035854, "node_id": "MDQ6VXNlcjQwMzU4NTQ=", "avatar_url": "https://avatars.githubusercontent.com/u/4035854?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dsmurrell", "html_url": "https://github.com/dsmurrell", "followers_url": "https://api.github.com/users/dsmurrell/followers", "following_url": "https://api.github.com/users/dsmurrell/following{/other_user}", "gists_url": "https://api.github.com/users/dsmurrell/gists{/gist_id}", "starred_url": "https://api.github.com/users/dsmurrell/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/dsmurrell/subscriptions", "organizations_url": "https://api.github.com/users/dsmurrell/orgs", "repos_url": "https://api.github.com/users/dsmurrell/repos", "events_url": "https://api.github.com/users/dsmurrell/events{/privacy}", "received_events_url": "https://api.github.com/users/dsmurrell/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I think I got there:\r\n\r\n```\r\n @classmethod\r\n def create(cls, **query):\r\n query['email_lower'] = query[\"email\"].lower()\r\n return super(Users, cls).create(**query)\r\n```\r\n\r\n@coleifer, is there any reason to use signals over this way of doing it?", "The proper way to do this would probably be to just create an index on the lower-case version of the email, since I assume you only are using it for uniqueness or querying.\r\n\r\n```sql\r\ncreate unique index \"user_email_lower\" on users(lower(email))\r\n```\r\n\r\nIf you really want to have it in its own field, then I'd override the `save()` method.\r\n\r\n```python\r\ndef save(self, *args, **kwargs):\r\n self.email_lower = self.email.lower()\r\n return super(...).save(*args, **kwargs)\r\n```" ]
"2021-06-23T20:40:16"
"2021-06-24T13:05:28"
"2021-06-24T13:05:28"
NONE
null
There are multiple uses of the `Users.create(email=X)` method in our codebase. Is there any way to modify the Users class or override the create method to make a field `email_lower` get set to another field that is always passed in (`email`)? I've tried: ``` def create(self, **kwargs): super().create(email_lower=kwargs["email"].lower(), **kwargs) ``` but I think that this doesn't work because the object isn't instantiated before the `create` method on the subclass is called. I also tried this: ``` def __init__(self, **kwargs): super(Users, self).__init__(email_lower=kwargs["email"].lower(), **kwargs) ``` but it didn't work either.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2432/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2432/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2431
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2431/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2431/comments
https://api.github.com/repos/coleifer/peewee/issues/2431/events
https://github.com/coleifer/peewee/pull/2431
921,538,451
MDExOlB1bGxSZXF1ZXN0NjcwNTI2Mzgw
2,431
feat(style): let the output follows PEP8
{ "login": "HydrogenDeuterium", "id": 36532302, "node_id": "MDQ6VXNlcjM2NTMyMzAy", "avatar_url": "https://avatars.githubusercontent.com/u/36532302?v=4", "gravatar_id": "", "url": "https://api.github.com/users/HydrogenDeuterium", "html_url": "https://github.com/HydrogenDeuterium", "followers_url": "https://api.github.com/users/HydrogenDeuterium/followers", "following_url": "https://api.github.com/users/HydrogenDeuterium/following{/other_user}", "gists_url": "https://api.github.com/users/HydrogenDeuterium/gists{/gist_id}", "starred_url": "https://api.github.com/users/HydrogenDeuterium/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/HydrogenDeuterium/subscriptions", "organizations_url": "https://api.github.com/users/HydrogenDeuterium/orgs", "repos_url": "https://api.github.com/users/HydrogenDeuterium/repos", "events_url": "https://api.github.com/users/HydrogenDeuterium/events{/privacy}", "received_events_url": "https://api.github.com/users/HydrogenDeuterium/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Eh, I'll pass." ]
"2021-06-15T15:43:38"
"2021-06-15T15:58:08"
"2021-06-15T15:58:07"
NONE
null
when using pwiz generates file to import,IDE always warning 'PEP 8: E302 expected 2 blank lines, found 1'. I try to fix it.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2431/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2431/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2431", "html_url": "https://github.com/coleifer/peewee/pull/2431", "diff_url": "https://github.com/coleifer/peewee/pull/2431.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2431.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2430
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2430/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2430/comments
https://api.github.com/repos/coleifer/peewee/issues/2430/events
https://github.com/coleifer/peewee/issues/2430
919,888,943
MDU6SXNzdWU5MTk4ODg5NDM=
2,430
[Feature request] Add more items to the log output
{ "login": "iHalt10", "id": 24728342, "node_id": "MDQ6VXNlcjI0NzI4MzQy", "avatar_url": "https://avatars.githubusercontent.com/u/24728342?v=4", "gravatar_id": "", "url": "https://api.github.com/users/iHalt10", "html_url": "https://github.com/iHalt10", "followers_url": "https://api.github.com/users/iHalt10/followers", "following_url": "https://api.github.com/users/iHalt10/following{/other_user}", "gists_url": "https://api.github.com/users/iHalt10/gists{/gist_id}", "starred_url": "https://api.github.com/users/iHalt10/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/iHalt10/subscriptions", "organizations_url": "https://api.github.com/users/iHalt10/orgs", "repos_url": "https://api.github.com/users/iHalt10/repos", "events_url": "https://api.github.com/users/iHalt10/events{/privacy}", "received_events_url": "https://api.github.com/users/iHalt10/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I'm going to pass on this for now." ]
"2021-06-13T21:55:20"
"2021-06-14T12:57:04"
"2021-06-14T12:57:04"
NONE
null
Hello, this is a small feature request, but I'd like to add more items to the log output. ```py from peewee import * import logging logger = logging.getLogger('peewee') logger.addHandler(logging.StreamHandler()) logger.setLevel(logging.DEBUG) a_db = SqliteDatabase('a.db') b_db = SqliteDatabase('b.db') class A(Model): name = TextField() class Meta: database = a_db class B(Model): name = TextField() class Meta: database = b_db a_db.create_tables([A]) b_db.create_tables([B]) # Output: # ('CREATE TABLE IF NOT EXISTS "a" ("id" INTEGER NOT NULL PRIMARY KEY, "name" TEXT NOT NULL)', []) # ('CREATE TABLE IF NOT EXISTS "b" ("id" INTEGER NOT NULL PRIMARY KEY, "name" TEXT NOT NULL)', []) ``` In the above, it is difficult to understand "which database was executed", can change it to the following code? ```py ... def execute_sql(self, sql, params=None, commit=SENTINEL): logger.debug((sql, params, self.database)) # Output: # ('...', [], 'a.db') ... ``` > https://github.com/coleifer/peewee/blob/master/peewee.py#L3146 Or, I would like to have a function that can extend the log output more items, in addition to the "(sql, params)" items. ```py ... class Database(_callable_context_manager): ... def get_log_items(self, sql, params, db): return (sql, params) def execute_sql(self, sql, params=None, commit=SENTINEL): logger.debug(self.get_log_items(sql, params, self)) ... ``` myself code ```py ... def my_get_log_items(sql, params, db): return (sql, params, db.database) db = SqliteDatabase('a.db') db.get_log_items = my_get_log_items db.execute_sql(...) # Output: # ('...', [], 'a.db') ``` Umm... If you have time, I'd appreciate your opinion.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2430/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2430/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2429
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2429/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2429/comments
https://api.github.com/repos/coleifer/peewee/issues/2429/events
https://github.com/coleifer/peewee/issues/2429
919,671,266
MDU6SXNzdWU5MTk2NzEyNjY=
2,429
create_table / drop_table implementation for MySQLMigrator and other engines
{ "login": "krokwen", "id": 4995998, "node_id": "MDQ6VXNlcjQ5OTU5OTg=", "avatar_url": "https://avatars.githubusercontent.com/u/4995998?v=4", "gravatar_id": "", "url": "https://api.github.com/users/krokwen", "html_url": "https://github.com/krokwen", "followers_url": "https://api.github.com/users/krokwen/followers", "following_url": "https://api.github.com/users/krokwen/following{/other_user}", "gists_url": "https://api.github.com/users/krokwen/gists{/gist_id}", "starred_url": "https://api.github.com/users/krokwen/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/krokwen/subscriptions", "organizations_url": "https://api.github.com/users/krokwen/orgs", "repos_url": "https://api.github.com/users/krokwen/repos", "events_url": "https://api.github.com/users/krokwen/events{/privacy}", "received_events_url": "https://api.github.com/users/krokwen/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Peewee has a schema migrations module: http://docs.peewee-orm.com/en/latest/peewee/playhouse.html#schema-migrations\r\n\r\nI don't quite understand the issue, so I don't know what to tell you, but this doesn't seem to me a peewee-specific issue.", "I'm talking about same module.\nIt's not able to create/drop table by itself.\nFor now you can create or drop table only with existing model class. But how to do it when model is no longer exists?\n\nOr i have to define model class in migration script every time? It's a huge crutch...\n", "I don't really understand the problem. If you have the model class, just import it into the migration and call `create_table()` - without the model definition Peewee has no idea what columns to add or anything. For dropping a table, I also don't see why this shouldn't be possible by using a model class imported from your app -- run the migration(s), then remove the model code.\r\n\r\nYou can also stub out a model class if all you need to do is drop it, or just run `database.execute_sql('DROP TABLE...')`." ]
"2021-06-12T22:09:25"
"2021-06-13T16:06:04"
"2021-06-13T00:45:08"
NONE
null
It's a required functionality for cases when you have global schema changes, Use case: day 1: Model A. Dataset A day 2: Model B. Dataset B day 3: Model B updated. Dataset A patrially moved to model B day 4: Model A is no longer needed and removed from code. And i have to apply these changes on late environment that on 'day 1' So, in this case it's impossible to use model's create and drop methods, because at the moment of migration run the old model will be already removed from code, but i have to apply all the changes in migrations.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2429/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2429/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2428
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2428/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2428/comments
https://api.github.com/repos/coleifer/peewee/issues/2428/events
https://github.com/coleifer/peewee/issues/2428
918,622,711
MDU6SXNzdWU5MTg2MjI3MTE=
2,428
Cannot reset ForeignKeyField to None
{ "login": "orsinium", "id": 9638362, "node_id": "MDQ6VXNlcjk2MzgzNjI=", "avatar_url": "https://avatars.githubusercontent.com/u/9638362?v=4", "gravatar_id": "", "url": "https://api.github.com/users/orsinium", "html_url": "https://github.com/orsinium", "followers_url": "https://api.github.com/users/orsinium/followers", "following_url": "https://api.github.com/users/orsinium/following{/other_user}", "gists_url": "https://api.github.com/users/orsinium/gists{/gist_id}", "starred_url": "https://api.github.com/users/orsinium/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/orsinium/subscriptions", "organizations_url": "https://api.github.com/users/orsinium/orgs", "repos_url": "https://api.github.com/users/orsinium/repos", "events_url": "https://api.github.com/users/orsinium/events{/privacy}", "received_events_url": "https://api.github.com/users/orsinium/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "This was happening because peewee only cleared the related object reference (the A instance on B) if the actual value of the FK changed. Since it was `NULL` to start with (e.g. A.id is None), peewee did not register that there was anything to do. In practice I don't believe this would lead to any inconsistency in the database, but I can see how it might be confusing.\r\n\r\nFixed.", "Thank you for the answer! I think I understand now: setting b.a to None was interpreted as setting b.a.id to None, which is already true. In my application, I distinguish between \"the A record is not created yet\" and \"there is no A record related to this B record\". So you're right, it was confusing. I found a workaround, though, by implementing `A.__bool__`." ]
"2021-06-11T11:24:44"
"2021-06-13T14:59:08"
"2021-06-11T12:58:56"
NONE
null
```python import peewee class A(peewee.Model): pass class B(peewee.Model): a = peewee.ForeignKeyField(A, null=True) b = B(a=A()) # works as expected: b.a # <A: None> # unexpected: b.a = None b.a # <A: None> ``` Expected behavior: when `None` is assigned to the ForeignKeyField, the field value is None. Actual behavior: assigning `None` to the ForeignKeyField has no effect. How to reset a foreign key to None?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2428/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2428/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2427
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2427/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2427/comments
https://api.github.com/repos/coleifer/peewee/issues/2427/events
https://github.com/coleifer/peewee/issues/2427
913,595,164
MDU6SXNzdWU5MTM1OTUxNjQ=
2,427
peewee create and use id field for deferred foreign key with primary_key=true
{ "login": "penja", "id": 2717390, "node_id": "MDQ6VXNlcjI3MTczOTA=", "avatar_url": "https://avatars.githubusercontent.com/u/2717390?v=4", "gravatar_id": "", "url": "https://api.github.com/users/penja", "html_url": "https://github.com/penja", "followers_url": "https://api.github.com/users/penja/followers", "following_url": "https://api.github.com/users/penja/following{/other_user}", "gists_url": "https://api.github.com/users/penja/gists{/gist_id}", "starred_url": "https://api.github.com/users/penja/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/penja/subscriptions", "organizations_url": "https://api.github.com/users/penja/orgs", "repos_url": "https://api.github.com/users/penja/repos", "events_url": "https://api.github.com/users/penja/events{/privacy}", "received_events_url": "https://api.github.com/users/penja/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[]
"2021-06-07T14:12:07"
"2021-06-08T00:05:47"
"2021-06-08T00:05:47"
NONE
null
Hello. We have the deferred foreign key that used like primary key in new table. In this case peewee ignore `primary_key=True` param in column declaration and try to use `id` like identifier in all sql queries. Please see invalid behavior in following code snippet. ``` import peewee from peewee import * import logging import sys log = logging.getLogger() log.setLevel(logging.DEBUG) handler = logging.StreamHandler(sys.stdout) log.addHandler(handler) print(peewee.__version__) db = SqliteDatabase(":memory:") class Base(Model): class Meta: database = db class B(Base): a = DeferredForeignKey("A", primary_key=True) x = CharField(null=True) class A(Base): id = AutoField() db.create_tables([B, A]) ``` Output ``` root@87a4315fc6f4:/vagrant# python dummy_sqlite.py 3.14.4 ('CREATE TABLE IF NOT EXISTS "a" ("id" INTEGER NOT NULL PRIMARY KEY)', []) ('CREATE TABLE IF NOT EXISTS "b" ("id" INTEGER NOT NULL PRIMARY KEY, "a_id" INTEGER NOT NULL PRIMARY KEY, "x" VARCHAR(255))', []) Traceback (most recent call last): File "/usr/local/lib/python3.8/site-packages/peewee.py", line 3144, in execute_sql cursor.execute(sql, params or ()) sqlite3.OperationalError: table "b" has more than one primary key During handling of the above exception, another exception occurred: ``` Thanks in advance for your response and probably for the fix. P.S There is workaround by in my opinion this is not obvious! ``` class B(Base): class Meta: primary_key = False a = DeferredForeignKey("A", primary_key=True) x = CharField(null=True) #or class B(Base): class Meta: primary_key = peewee.CompositeKey("a") a = DeferredForeignKey("A") x = CharField(null=True) ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2427/reactions", "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2427/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2426
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2426/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2426/comments
https://api.github.com/repos/coleifer/peewee/issues/2426/events
https://github.com/coleifer/peewee/issues/2426
912,118,561
MDU6SXNzdWU5MTIxMTg1NjE=
2,426
Syntax error when selecting in VALUES
{ "login": "Hippopotas", "id": 9470058, "node_id": "MDQ6VXNlcjk0NzAwNTg=", "avatar_url": "https://avatars.githubusercontent.com/u/9470058?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Hippopotas", "html_url": "https://github.com/Hippopotas", "followers_url": "https://api.github.com/users/Hippopotas/followers", "following_url": "https://api.github.com/users/Hippopotas/following{/other_user}", "gists_url": "https://api.github.com/users/Hippopotas/gists{/gist_id}", "starred_url": "https://api.github.com/users/Hippopotas/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Hippopotas/subscriptions", "organizations_url": "https://api.github.com/users/Hippopotas/orgs", "repos_url": "https://api.github.com/users/Hippopotas/repos", "events_url": "https://api.github.com/users/Hippopotas/events{/privacy}", "received_events_url": "https://api.github.com/users/Hippopotas/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "This works fine for me:\r\n\r\n```python\r\ndb = SqliteDatabase(':memory:')\r\n\r\nclass KV(Model):\r\n key = TextField()\r\n value = IntegerField()\r\n class Meta:\r\n database = db\r\n\r\ndb.create_tables([KV])\r\n\r\nKV.insert_many([('k1', 1), ('k2', 2), ('k3', 3), ('k4', 4), ('k5', 5)]).execute()\r\n\r\nvl = ValuesList([('k1', 1), ('k3', 3), ('k5', 5)])\r\nvl_e = EnclosedNodeList([vl])\r\n\r\nquery = (KV.select()\r\n .where(Tuple(KV.key, KV.value).in_(vl_e)))\r\nfor row in query:\r\n print(row.key, row.value)\r\n```\r\n\r\nPrints:\r\n\r\n```\r\nk1 1\r\nk3 3\r\nk5 5\r\n```", "Hm. I ran your example code, but got the same error. Could I ask what the sql being generated is?\r\n\r\nEdit: never mind, ignore me. I reinstalled everything, and now it miraculously works. Sorry for the bother." ]
"2021-06-05T06:27:19"
"2021-06-06T05:12:51"
"2021-06-05T13:35:46"
NONE
null
Peewee version: 3.4.14 SQLite version: 3.35 When running code of the form (as recommended in issues [2344](https://github.com/coleifer/peewee/issues/2344), [2411](https://github.com/coleifer/peewee/issues/2411), [2414](https://github.com/coleifer/peewee/issues/2414)): ```python foo = [(1, 2), (3, 4)] bar = ValuesList(foo, columns=('col1', 'col2')) baz = EnclosedNodeList([bar]) t1.select().where(Tuple(t1.col1, t1.col2).in_(baz)) ``` I am getting a different kind of error: `OperationalError: near ",": syntax error` The query it is attempting to execute is apparently (in (query, param) form) ``` ('SELECT "t1"."col1", "t1"."col2" FROM "t1" WHERE (("t1"."col1", "t1"."col2") IN (VALUES (?, ?), (?, ?)))', [1, 2, 3, 4]) ``` The same occurs on ```python t1.raw("""SELECT col1 FROM t1 WHERE ((col1, col2) IN (VALUES (1, 2))""").execute() ``` After some tinkering, it appears that the problematic syntax is the comma between `col1, col2` I am unsure why this is happening. I played around with the parentheses a bit, but could not figure out a fix. This does not occur on non-row-values queries, e.g. ``` t1.select().where(t1.col1 << [1, 2, 3]) ('SELECT "t1"."col1", "t1"."col2" FROM "t1" WHERE ("t1"."col1" IN (?, ?, ?))', [1, 2, 3]) ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2426/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2426/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2425
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2425/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2425/comments
https://api.github.com/repos/coleifer/peewee/issues/2425/events
https://github.com/coleifer/peewee/issues/2425
911,302,399
MDU6SXNzdWU5MTEzMDIzOTk=
2,425
Multi-encoding support in playhouse DataSet
{ "login": "JarneVerhaeghe", "id": 61466015, "node_id": "MDQ6VXNlcjYxNDY2MDE1", "avatar_url": "https://avatars.githubusercontent.com/u/61466015?v=4", "gravatar_id": "", "url": "https://api.github.com/users/JarneVerhaeghe", "html_url": "https://github.com/JarneVerhaeghe", "followers_url": "https://api.github.com/users/JarneVerhaeghe/followers", "following_url": "https://api.github.com/users/JarneVerhaeghe/following{/other_user}", "gists_url": "https://api.github.com/users/JarneVerhaeghe/gists{/gist_id}", "starred_url": "https://api.github.com/users/JarneVerhaeghe/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/JarneVerhaeghe/subscriptions", "organizations_url": "https://api.github.com/users/JarneVerhaeghe/orgs", "repos_url": "https://api.github.com/users/JarneVerhaeghe/repos", "events_url": "https://api.github.com/users/JarneVerhaeghe/events{/privacy}", "received_events_url": "https://api.github.com/users/JarneVerhaeghe/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[]
"2021-06-04T09:16:26"
"2021-06-04T12:50:29"
"2021-06-04T12:50:29"
NONE
null
It appears the DataSet module from playhouse does not support other encodings such as 'latin-1'. I locally solved it by adjusting the open_file, thaw, and freeze function to have an extra argument: Line 21-22 ``` def open_file(f, mode,encoding_='utf8'): return open(f, mode, encoding=encoding_) ``` Line 152-156 ``` def freeze(self, query, format='csv', filename=None, file_obj=None, encoding='utf8', **kwargs): self._check_arguments(filename, file_obj, format, self._export_formats) if filename: file_obj = open_file(filename, 'w',encoding) ``` Line 164-168 ``` def thaw(self, table, format='csv', filename=None, file_obj=None, strict=False, encoding='utf8', **kwargs): self._check_arguments(filename, file_obj, format, self._export_formats) if filename: file_obj = open_file(filename, 'r',encoding) ``` This issue is more of a suggestion for multi-encoding support. I did not want to open a pull request for this.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2425/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2425/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2423
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2423/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2423/comments
https://api.github.com/repos/coleifer/peewee/issues/2423/events
https://github.com/coleifer/peewee/issues/2423
907,659,141
MDU6SXNzdWU5MDc2NTkxNDE=
2,423
Create tables from ModelSelect outside the public schema
{ "login": "albireox", "id": 568775, "node_id": "MDQ6VXNlcjU2ODc3NQ==", "avatar_url": "https://avatars.githubusercontent.com/u/568775?v=4", "gravatar_id": "", "url": "https://api.github.com/users/albireox", "html_url": "https://github.com/albireox", "followers_url": "https://api.github.com/users/albireox/followers", "following_url": "https://api.github.com/users/albireox/following{/other_user}", "gists_url": "https://api.github.com/users/albireox/gists{/gist_id}", "starred_url": "https://api.github.com/users/albireox/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/albireox/subscriptions", "organizations_url": "https://api.github.com/users/albireox/orgs", "repos_url": "https://api.github.com/users/albireox/repos", "events_url": "https://api.github.com/users/albireox/events{/privacy}", "received_events_url": "https://api.github.com/users/albireox/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "I've chosen to support this by allowing a tuple for the `table_name` parameter. So:\r\n\r\n```python\r\n\r\nquery = SomeModel.select()...\r\nquery.create_table('my_new_table')\r\n\r\n# OR\r\nquery.create_table(('schema', 'my_new_table'))\r\n```" ]
"2021-05-31T18:43:49"
"2021-06-16T13:08:58"
"2021-06-16T13:07:57"
NONE
null
I think there is a bug or limitation in the `ModelSelect.create_table()` method. The way it works right now, you can only create tables in the public schema. This is because `Entity` expects multiple arguments, one for each part of the path, but `create_table` can send only one. If you pass the whole path as a string (e.g., `sandbox.test_table`) it gets quoted as `"sandbox.test_table"` which still points to the public schema. I think the solution is either to allow to pass a tuple to `create_table()` (which will break compatibility) or to change the function signature so that it's more greedy getting the arguments so that you can pass `create_table('sandbox', 'test_table')` and all those arguments make it to the `Entity` instance.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2423/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2423/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2422
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2422/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2422/comments
https://api.github.com/repos/coleifer/peewee/issues/2422/events
https://github.com/coleifer/peewee/pull/2422
906,971,690
MDExOlB1bGxSZXF1ZXN0NjU3ODgxODkz
2,422
Update quickstart.rst
{ "login": "universuen", "id": 52519513, "node_id": "MDQ6VXNlcjUyNTE5NTEz", "avatar_url": "https://avatars.githubusercontent.com/u/52519513?v=4", "gravatar_id": "", "url": "https://api.github.com/users/universuen", "html_url": "https://github.com/universuen", "followers_url": "https://api.github.com/users/universuen/followers", "following_url": "https://api.github.com/users/universuen/following{/other_user}", "gists_url": "https://api.github.com/users/universuen/gists{/gist_id}", "starred_url": "https://api.github.com/users/universuen/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/universuen/subscriptions", "organizations_url": "https://api.github.com/users/universuen/orgs", "repos_url": "https://api.github.com/users/universuen/repos", "events_url": "https://api.github.com/users/universuen/events{/privacy}", "received_events_url": "https://api.github.com/users/universuen/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Opener requested I delete this PR. Unfortunately, I don't believe I can delete this - I don't see an option in the github UI." ]
"2021-05-31T02:07:36"
"2021-06-03T15:16:24"
"2021-05-31T22:36:20"
NONE
null
Found a missing word
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2422/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2422/timeline
null
null
false
{ "url": "https://api.github.com/repos/coleifer/peewee/pulls/2422", "html_url": "https://github.com/coleifer/peewee/pull/2422", "diff_url": "https://github.com/coleifer/peewee/pull/2422.diff", "patch_url": "https://github.com/coleifer/peewee/pull/2422.patch", "merged_at": null }
https://api.github.com/repos/coleifer/peewee/issues/2421
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2421/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2421/comments
https://api.github.com/repos/coleifer/peewee/issues/2421/events
https://github.com/coleifer/peewee/issues/2421
905,310,745
MDU6SXNzdWU5MDUzMTA3NDU=
2,421
Preferred way to do `database.execute_sql` ?
{ "login": "andersentobias", "id": 4043059, "node_id": "MDQ6VXNlcjQwNDMwNTk=", "avatar_url": "https://avatars.githubusercontent.com/u/4043059?v=4", "gravatar_id": "", "url": "https://api.github.com/users/andersentobias", "html_url": "https://github.com/andersentobias", "followers_url": "https://api.github.com/users/andersentobias/followers", "following_url": "https://api.github.com/users/andersentobias/following{/other_user}", "gists_url": "https://api.github.com/users/andersentobias/gists{/gist_id}", "starred_url": "https://api.github.com/users/andersentobias/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/andersentobias/subscriptions", "organizations_url": "https://api.github.com/users/andersentobias/orgs", "repos_url": "https://api.github.com/users/andersentobias/repos", "events_url": "https://api.github.com/users/andersentobias/events{/privacy}", "received_events_url": "https://api.github.com/users/andersentobias/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Peewee has been on 3.0 for about 3 years now, so you might want to upgrade as there are many improvements.\r\n\r\nThat said, as long as you're opening the pool connection before the request starts, and closing it (returning it back to the pool) after the request, then your code should work just fine." ]
"2021-05-28T12:26:05"
"2021-05-28T12:36:58"
"2021-05-28T12:36:58"
NONE
null
We're using a DB connection pool with peewee 2.8.0. We needed to optimize a query by writing handwritten SQL to our PostgreSQL db. What is the preferred way of doing such a query? We're using peewee together with a Flask app, if that helps. Right now, the code is: ```python database = PooledPostgresqlExtDatabase(...) # Lots of code... def func(): cursor = database.execute_sql('SELECT ...') # More code obviously. result = cursor.fetchall() # ... ``` Note that we're not closing the db cursor manually.
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2421/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2421/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2420
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2420/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2420/comments
https://api.github.com/repos/coleifer/peewee/issues/2420/events
https://github.com/coleifer/peewee/issues/2420
903,470,809
MDU6SXNzdWU5MDM0NzA4MDk=
2,420
How to get BaseModel Fields?
{ "login": "beucismis", "id": 40023234, "node_id": "MDQ6VXNlcjQwMDIzMjM0", "avatar_url": "https://avatars.githubusercontent.com/u/40023234?v=4", "gravatar_id": "", "url": "https://api.github.com/users/beucismis", "html_url": "https://github.com/beucismis", "followers_url": "https://api.github.com/users/beucismis/followers", "following_url": "https://api.github.com/users/beucismis/following{/other_user}", "gists_url": "https://api.github.com/users/beucismis/gists{/gist_id}", "starred_url": "https://api.github.com/users/beucismis/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/beucismis/subscriptions", "organizations_url": "https://api.github.com/users/beucismis/orgs", "repos_url": "https://api.github.com/users/beucismis/repos", "events_url": "https://api.github.com/users/beucismis/events{/privacy}", "received_events_url": "https://api.github.com/users/beucismis/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Post questions on stackoverflow or IRC. This is for tracking issues.\r\n\r\n```python\r\n\r\nUser._meta.fields # dict of field name -> field instance\r\nUser._meta.sorted_fields # list of field instances in order\r\nUser._meta.sorted_field_names # Just the names\r\n```", "Thanks." ]
"2021-05-27T09:35:26"
"2021-05-27T12:49:13"
"2021-05-27T12:43:22"
NONE
null
I want to get the defined `Fields` as a list. ``` class User(BaseModel): id = BigIntegerField(primary_key=True) nick = CharField(unique=True) age = SmallIntegerField(unique=True) class Meta: database = db ```
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2420/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2420/timeline
null
completed
null
null
https://api.github.com/repos/coleifer/peewee/issues/2419
https://api.github.com/repos/coleifer/peewee
https://api.github.com/repos/coleifer/peewee/issues/2419/labels{/name}
https://api.github.com/repos/coleifer/peewee/issues/2419/comments
https://api.github.com/repos/coleifer/peewee/issues/2419/events
https://github.com/coleifer/peewee/issues/2419
901,992,407
MDU6SXNzdWU5MDE5OTI0MDc=
2,419
The client was disconnected by the server because of inactivity.
{ "login": "GreatBahram", "id": 14103831, "node_id": "MDQ6VXNlcjE0MTAzODMx", "avatar_url": "https://avatars.githubusercontent.com/u/14103831?v=4", "gravatar_id": "", "url": "https://api.github.com/users/GreatBahram", "html_url": "https://github.com/GreatBahram", "followers_url": "https://api.github.com/users/GreatBahram/followers", "following_url": "https://api.github.com/users/GreatBahram/following{/other_user}", "gists_url": "https://api.github.com/users/GreatBahram/gists{/gist_id}", "starred_url": "https://api.github.com/users/GreatBahram/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/GreatBahram/subscriptions", "organizations_url": "https://api.github.com/users/GreatBahram/orgs", "repos_url": "https://api.github.com/users/GreatBahram/repos", "events_url": "https://api.github.com/users/GreatBahram/events{/privacy}", "received_events_url": "https://api.github.com/users/GreatBahram/received_events", "type": "User", "site_admin": false }
[]
closed
false
null
[]
null
[ "Thanks" ]
"2021-05-26T08:28:09"
"2021-05-26T15:36:53"
"2021-05-26T12:38:35"
NONE
null
Hi there, I got this exception when I have a long connection that exceeds both `wait_timeout` and `interactive_timeout` timeouts: ```python File "/usr/lib/python3/dist-packages/peewee.py", line 5513, in get return sq.get() File "/usr/lib/python3/dist-packages/peewee.py", line 5911, in get return clone.execute(database)[0] File "/usr/lib/python3/dist-packages/peewee.py", line 1587, in inner return method(self, database, *args, **kwargs) File "/usr/lib/python3/dist-packages/peewee.py", line 1658, in execute return self._execute(database) File "/usr/lib/python3/dist-packages/peewee.py", line 1809, in _execute cursor = database.execute(self) File "/usr/lib/python3/dist-packages/peewee.py", line 2666, in execute return self.execute_sql(sql, params, commit=commit) File "/usr/lib/python3/dist-packages/peewee.py", line 2660, in execute_sql self.commit() File "/usr/lib/python3/dist-packages/peewee.py", line 2451, in __exit__ reraise(new_type, new_type(*exc_args), traceback) File "/usr/lib/python3/dist-packages/peewee.py", line 178, in reraise raise value.with_traceback(tb) File "/usr/lib/python3/dist-packages/peewee.py", line 2653, in execute_sql cursor.execute(sql, params or ()) File "/usr/lib/python3/dist-packages/MySQLdb/cursors.py", line 209, in execute res = self._query(query) File "/usr/lib/python3/dist-packages/MySQLdb/cursors.py", line 315, in _query db.query(q) File "/usr/lib/python3/dist-packages/MySQLdb/connections.py", line 226, in query _mysql.connection.query(self, query) peewee.OperationalError: (4031, 'The client was disconnected by the server because of inactivity. See wait_timeout and interactive_timeout for configuring this behavior.') ``` I spotted the [shortcuts](https://github.com/coleifer/peewee/blob/master/playhouse/shortcuts.py) module inside the peewee code handles these errors: 2006,2013, 2014. I think this can be added to the `ReconnectMixin` class as well. Do you want me to add this into the `ReconnectMixin` class?
{ "url": "https://api.github.com/repos/coleifer/peewee/issues/2419/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/coleifer/peewee/issues/2419/timeline
null
completed
null
null