content
stringlengths 240
2.34M
|
|---|
<issue_start><issue_comment>Title: Compiler flags for error/warning checks.
username_0: When using `--gc:stack`, `system.nim` imports/exports `MemRegion` type. (not exported with other gc's)
When using it, nimlime marks it's usage as not defined error, which I'm guessing it checks the file with standard gc.
Is there any way to make error checking aware of compiler flags, or some other way to make nimlime ignore this errors?
<issue_comment>username_1: Hm. There are two ways this could be dealt with, neither implemented at this time. Either a setting could be added that allows passing additional arguments to the `nim check` command (which is executed as a subprocess) or an ignore_regex setting could be added to allow filtering the `Nim check` output. |
<issue_start><issue_comment>Title: Support CamelCase CSS names
username_0: Following on from a discussion on the mailling list, https://groups.google.com/a/continuum.io/forum/#!topic/bokeh/BVAPlqMyI1Y.
@bryevdv mentioned that "One thing to mention is that "Black" (capital B) is not recognized as a color, all color names are lower case."
But:
* The link we point to in the docs: http://www.w3schools.com/cssref/css_colornames.asp
* The link we go to in our OpenURL example: http://www.colors.commutercreative.com/
* And, the OpenURL example: https://github.com/bokeh/bokeh/blob/master/examples/glyphs/colors.py
all use CamelCase colors.
I have been trying to use them and just assumed it was broken.
Turns out we need to lowercase them - I can't see this in the docs.
Either we need a warning for this, or, and I think better, we should support the CamelCase which seems to be the convention.
<issue_comment>username_0: I'm changing this to bug, because we appear to support them sometimes and not others. (code snippet coming)
<issue_comment>username_0: ````python
from bokeh.io import output_notebook, show
from bokeh.plotting import figure
from bokeh.properties import value
output_notebook()
f = figure(x_range=(0,3), y_range=(0,4))
f.circle([1], [1], radius=0.2, fill_color=["LemonChiffon"]) # Works
f.circle([2], [1], radius=0.2, fill_color=["lemonchiffon"]) # Works
f.circle([1], [2], radius=0.2, fill_color="LemonChiffon") # Does not work
f.circle([2], [2], radius=0.2, fill_color="lemonchiffon") # Works
f.circle([1], [3], radius=0.2, fill_color=value("LemonChiffon")) # Does not work
f.circle([2], [3], radius=0.2, fill_color=value("lemonchiffon")) # Does not work
show(f)
````
(note the warning comes from the line `f.circle([1], [2], radius=0.2, fill_color="LemonChiffon")`)<issue_closed> |
<issue_start><issue_comment>Title: Feature Request: Option for button instead of toggle for switches
username_0: I have a pi connected to my garage door opener. I am using a command line switch option to open/close my garage door. I am using the same command to both open and close the door, because all it does is turn on a relay for a few seconds and then back off. Would it be possible to have just a push button with one command option. Then maybe in history just show an event and not lable them as on/off?
<issue_comment>username_1: It doesn't sound like you want a switch at all. Why not use a script, they show up as a button on the top.
<issue_comment>username_0: Hmm okay I'll have to try that out, but what would happen if it was grouped with a sensor? I have a sensor as well that tells me the sate of the the door, open or closed.
<issue_comment>username_1: Scripts have a switch to turn them on or cancel when they are running.
<issue_comment>username_0: Great. I'll give it a shot. Thanks.
<issue_comment>username_1: I have opened PR #513 that will allow you to expose shell commands as services so you can call them from a script.
<issue_comment>username_2: @username_0 Does this work for you?
<issue_comment>username_3: I'm having the exact same use case. I don't have a sensor yet to detect if the door is open, but executing the script will make the door close if it was open. I like to have my controls in the group (Living room, Garage, ...), so a script at the top of the screen isn't really the best way. I currently have a command_switch that executes the same command at oncmd and offcmd, but the state can be out of sync offcourse.
<issue_comment>username_1: You can add a script to a group :+1:
<issue_comment>username_3: Hmm, the script is still showing up as a toggle switch, so it is exactly the same as if I would define a command_switch.
<issue_comment>username_1: Scripts will show a toggle because execution can be cancelled. However, they will behave as a button if they have no delay
<issue_comment>username_3: I have created the following bash-script
````bash
#!/bin/bash
sleep 1
````
and my `configuration.yaml` contains this:
````yaml
shell_command:
signal_garage_door: /opt/test.sh
script:
garage_door:
sequence:
- service: shell_command.signal_garage_door
````
But the toggle stays in the on-position.
<issue_comment>username_1: @username_3 that's a bug in the frontend that I just fixed a couple of minutes ago. Should be ok in the next release.<issue_closed> |
<issue_start><issue_comment>Title: Use undo transaction so razor docs don't have redundant undos
username_0: Fixes devdiv 766411 (partly)
The problem is that using uncomment feature on vb code in vbhtml files leaves the undo stack with four individual undos, instead of one or maybe two. (Usually there are two, one for the text change and one for the auto-formatting change.)
Part of the problem is that the razor docs have special logic when applying changes back to the buffer and actually do two edits, one to apply the changes and then another that does indentation adjustment after the fact. This normally causes two undo items per call to workspace TryApplyChanges for a razor document. However, most command handler features wrap calls to TryApplyChanges with undo transactions so they end up with only one. The comment/un-comment handler was not doing this. That has been changed with this PR.
Unfortunately that still leaves 3 undos instead of the desired 2, but its an improvement.
@username_2 @username_1 please review.
<issue_comment>username_1: :+1: other than string nitpick.
<issue_comment>username_2: :+1: |
<issue_start><issue_comment>Title: Parse DATA error?
username_0: How to send DATA end?
https://github.com/username_1/smtpd/blob/master/smtpd.go#L252
```
$ nc localhost 2525
220 localhost MyServerApp SMTP Service ready
EHLO hello server
250 localhost greets hello server
MAIL FROM:<admin@google.com>
250 Ok
RCPT TO:<admin@nasa.gov>
250 Ok
DATA
354 Start mail input; end with <CR><LF>.<CR><LF>
From:admin@google.com
To:admin@nasa.gov
Subject:Test
<h1>Test</h1>
\r\n.\r\n
.\r\n
```
but nothing.
My locale
```
LANG=ru_RU.UTF-8
LC_CTYPE=ru_RU.UTF-8
LC_NUMERIC="ru_RU.UTF-8"
LC_TIME="ru_RU.UTF-8"
LC_COLLATE="ru_RU.UTF-8"
LC_MONETARY="ru_RU.UTF-8"
LC_MESSAGES="ru_RU.UTF-8"
LC_PAPER="ru_RU.UTF-8"
LC_NAME="ru_RU.UTF-8"
LC_ADDRESS="ru_RU.UTF-8"
LC_TELEPHONE="ru_RU.UTF-8"
LC_MEASUREMENT="ru_RU.UTF-8"
LC_IDENTIFICATION="ru_RU.UTF-8"
LC_ALL=
```
<issue_comment>username_1: It seems like you're typing or pasting the \r and \n directly into your terminal, which means they are being interpreted literally instead of being interpreted as control characters. If you want to use Netcat, try putting the entire exchange in quotes on a single line, then pipe it to nc like this:
$ msg="EHLO hello server\nMAIL FROM:<admin@google.com>\nRCPT TO:<admin@nasa.gov>\nDATA\nFrom:admin@google.com\r\nTo:admin@nasa.gov\r\nSubject:Test\r\n\r\n<h1>Test</h1>\r\n.\r\n";
$ echo -ne $msg | nc localhost 2525
<issue_comment>username_0: @username_1 thanks.<issue_closed> |
<issue_start><issue_comment>Title: Steps to run the simulation
username_0: Hi! i successfully build the project. However, i have no idea what to do next. Can you make a brief instruction in order to run the simulation? Eg. what launch file to launch, anything else to set? Thank you so much for helping.
<issue_comment>username_1: Anyone found an answer to this? |
<issue_start><issue_comment>Title: Button bar pills
username_0: <ul class="button-bar">
<li><a href="" class="pill">Test</a></li>
<li><a href="">Another</a></li>
<li><a href="" class="pill">Test</a></li>
</ul>
It would be nice if this would round the appropriate edge, |
<issue_start><issue_comment>Title: infer_android_step 1.0.0
username_0: This is a Android step for [Infer](http://fbinfer.com)
<issue_comment>username_1: Thank you very much!
We will review it ASAP. Probably on tomorrow.
<issue_comment>username_1: Hi @username_0 !
Currently android sdk and tools are only available on the `linux` and `xamarin` stack, but you set `project_type_tags: ios`, is `infer` working whithout this environment?
<issue_comment>username_2: Thanks @username_1,
@username_0 does not work today he will do more test soon and try it on other environment and fix your requests.
<issue_comment>username_1: Hi @username_2 and @username_0 ,
i'm just curious if you want to finish this step share, or should we close this PR?
<issue_comment>username_2: Infer is in pause actually because of issue on the tool. So we will close it. |
<issue_start><issue_comment>Title: [Feature Request] Allowing to edit templates' default values
username_0: Hello,
I've been using Ajenti-v for only couple of days, but already loving so far!
I've a request, I hope this is considered.
I'd love to have a section to edit default values in templates.
E.g: I wanted to edit `worker_connections` in `nxinx.conf`, and there was no place to edit.
`nginx.conf` suggests me to add custom values for environment in `/etc/nginx.custom.events.d/*.conf;`. I've added a worker_connections value, but then, because it's already in the template, nginx config test failed, and I couldn't restart the server. I had to edit manually from `/var/lib/ajenti/plugins/vh-nginx/nginx_templates.py`.
I'd be awesome if there was at least a textarea for users holding the template to edit these variables.
I'd appreciate if this is considered.
Thanks,
Kind Regards,
<issue_comment>username_1: +1 !
<issue_comment>username_2: +1
<issue_comment>username_3: +1
<issue_comment>username_4: +1
<issue_comment>username_5: +1
<issue_comment>username_6: +1
<issue_comment>username_7: +1
<issue_comment>username_8: four years, later, still +1
<issue_comment>username_9: +1 |
<issue_start><issue_comment>Title: Suppress 4-digit Research Field code
username_0: As a student, since I no longer have to use the Proquest research field code on a paper form, I don't need to see it when I select a research field during the online submission process.
- [ ] When I select a research field from the Proquest list, I do not see the 4-digit number code associated with that research field.
<issue_comment>username_0: Related to #188 and #242<issue_closed>
<issue_comment>username_1: Implemented in #277 |
<issue_start><issue_comment>Title: How to put multiple locations in version 0.3.6
username_0: I'm sorry if this is too simple, but this version is possible to place multiple accounts, but can't find a way to put multiple locations. If someone help, thank you very much.
<issue_comment>username_1: I think the only way is to have multiple instances of the app open, scanning different places at the same time. As every instance will be working with the same database, you should see all the scan information.
<issue_comment>username_0: I just do not understand the cause of having the option of multiple accounts. I thought I would have a way to put multiple locations at the same time. But thanks!
<issue_comment>username_2: you can add multiple accounts to increase the scanning radius and scanning speed in one location. Running a 10 radius in less than 3 mins here ;)
<issue_comment>username_0: Wow! Really! I had not realized it. Thank you very much!<issue_closed> |
<issue_start><issue_comment>Title: Use com.twitter.finagle.Service as base type for services
username_0:
<issue_comment>username_0: @username_1 @username_2
<issue_comment>username_1: What is the effect of this? Looks like possibly something that would be useful as an upstream patch?
<issue_comment>username_2: The code was providing the generated Thrift interface to the `proxy` call, which means that the generated class has a parent of `java.lang.Object`. This resulted in the `ClassCastException` that @username_0 was seeing:
```
Caused by java.lang.ClassCastException
amperity.source.service.proxy$java.lang.Object$SourceService$ServiceIface$d5009d79
cannot be cast to com.twitter.finagle.Service
```
Using the base class of `Service` makes the proxied class operate the way that the Finagle code expects services to actually be constructed, extending the base class and implementing the generated interface. |
<issue_start><issue_comment>Title: security.hideProcessInformation prevents systemctl reboot issued by a regular user from working
username_0: ## Issue description
When `security.hideProcessInformation = true`, a regular user logged in locally (as the only `seat`) cannot run `systemctl reboot`, `poweroff`, `suspend`, `halt`, `hibernate` etc.
This ability is a convenience for desktop systemd users. **E.g. this issue prevents GNOME’s power buttons (restart, power off, hibernate) from working.**
### Steps to reproduce
1. Set `security.hideProcessInformation = true`.
1. As a regular user logged in locally (i.e. not over SSH), run `systemctl reboot` (or just `reboot`, it’s the same).
1. You’ll get:
```
Failed to set wall message, ignoring: Interactive authentication required.
Failed to reboot system via logind: Interactive authentication required.
Failed to start reboot.target: Interactive authentication required.
See system logs and 'systemctl status reboot.target' for details.
```
1. Without `hideProcessInformation`, the machine would reboot correctly.
## Technical details
* System: 16.09.1149.e360c72 (Flounder)
* Nix version: nix-env (Nix) 1.11.4
* Nixpkgs version: "16.09.1149.e360c72"
<issue_comment>username_0: `logind` runs with `root` privileges. It *has* the information about all processes. It should not matter that a regular user that wants to reboot doesn’t have it all.
<issue_comment>username_0: Also related: with `hideProcessInformation` set, one cannot change screen brightness in GNOME:
```
Dec 06 17:35:32 nixos pkexec[2315]: mikolaj: Error executing command as another user: Not authorized [USER=root] [TTY=unknown] [CWD=/home/mikolaj] [COMMAND=/nix/store/llny1hmxjbjk3j24ws7bnkh4ghrw8qch-gnome-settings-daemon-3.20.1/libexec/gsd-backlight-helper --set-brightness 891]
Dec 06 17:35:32 nixos gnome-settings-daemon.desktop[1452]: Error executing command as another user: Not authorized
Dec 06 17:35:32 nixos gnome-settings-daemon.desktop[1452]: This incident has been reported.
```
*Side note: with this option enabled, changing brightness and rebooting works fine in Xfce.*
<issue_comment>username_1: I have a pending patch for some issues I had. I'll do some testing and push it if it turns out to resolve this problem as well.
<issue_comment>username_1: I looked at the implementation of hidepid in the kernel. The logic does not take superuser into account; the task must be in the proc gid or have the `sys_ptrace` capability to access process information owned by other users. systemd-logind lacks that capability. (It would have worked out if we didn't use a designated proc group, it'd default to 0). The other problem you noted has to do with polkit arbitration, which is also broken.<issue_closed>
<issue_comment>username_0: Awesome, thank you! :heart_eyes_cat: |
<issue_start><issue_comment>Title: Improper support of reused vocabulary
username_0: /CC @username_1: I just noticed for any vocabulary terms that are reused from other vocabularies (e.g., ADL's verbs) you shouldn't put any blank RDF attributes in the HTML.
For example,
```
<td property="skos:prefLabel" lang="en" xml:lang="en" content="initialized">initialized</td>
<td property="skos:definition" lang="en" xml:lang="en" content="">
<td rel="skos:closeMatch" resource="">
```
Should really just be :
```
<td>initialized</td>
<td>
<td>
```
<issue_comment>username_1: Here is an example of the HTML used by the acrossx profile where they are reusing the ADL "answered" verb:
<!-- An existing Verb borrowed/referenced from another controlled vocabulary-->
<tbody resource="http://adlnet.gov/expapi/verbs/answered" id="answered">
<tr class="warning">
<td>answered</td>
<td></td>
<td property="skos:scopeNote" lang="en" xml:lang="en" content="Used to show the actor's answer in a quiz or practice test.">Used to show the actor's answer in a quiz or practice test.</td>
<td><a href="http://adlnet.gov/expapi/verbs/answered">http://adlnet.gov/expapi/verbs/answered</a></td>
<td></td>
<td></td>
<td rel="xapi:referencedBy" resource="https://w3id.org/xapi/acrossx"><a href="https://w3id.org/xapi/adl" target="_blank">https://w3id.org/xapi/adl</a></td>
</tr>
</tbody><issue_closed> |
<issue_start><issue_comment>Title: Proposal: interdependence testing, leveraging reduxsauce (ideal for generated templates))
username_0: To start of with, thank you for react-boilerplate -- it is a great springboard.
In my code I use reduxsauce. This allows me the following testing by default whenever I create a new container:
* testing that for every constant defined
* there is a defined action
* there is a defined reducer
Using templates, we can create the test whenever we generate a new container. Is there any appetite for this?
Here are the examples of my tests:
```
// actions.test.js
import expect from 'expect';
import * as actions from '../actions';
import constants from '../constants';
describe('Container actions', () => {
describe('Do actions map to constants', () => {
const actionsResults = Object.values(actions)
.map(action => action({}));
Object.keys(constants)
.forEach(key => {
it(`has an action with result type ${key}`, () => {
const func = actionsResults.find(act => act.type === constants[key]);
expect(func).toExist();
});
});
});
});
```
```
// reducer.test.js
import expect from 'expect';
import reducer, {
INITIAL_STATE,
HANDLERS,
} from '../reducer';
import constants from '../constants';
describe('SignUp reducer', () => {
it('returns the initial state', () => {
expect(reducer(undefined, {})).toEqual(INITIAL_STATE);
});
Object.keys(constants)
.forEach(key => {
it(`has reducer function mapped to ${key}`, () => {
expect(typeof HANDLERS[key]).toEqual('function');
});
});
});
```<issue_closed>
<issue_comment>username_1: I think we have enough complexity and abstractions as-is, adding another one for redux isn't necessary. People can add that quite easily if they want to!
Thanks for the suggestion! |
<issue_start><issue_comment>Title: Time out Issues when Using --upload
username_0: some sort of time out occurring between the python code usb dongle or ant stack, need to dig in to this. But by enabling --debug I got files to start uploading to the garmin during a sync (progress bar on the garmin :D ) but the upload fails towards the end. so It looks like it nearly works :dancer:
<issue_comment>username_1: I'm running into the same timeout issue as username_0. Has there been any progress when running:
antfs-cli --upload
After you've already synced some files? I'm trying this again after leaving it alone for a couple weeks, and now I'm getting timeout errors. There are a few new runs on the watch I'm trying to pull down.
<issue_comment>username_1: I was able to finish the download after deleting the most recent FIT file in my /activities directory, then running the command again.
<issue_comment>username_2: I am also having upload problems. I have created a .tcx file with a course in it (is this the correct file format?) I placed it in the "courses" folder, and when I run antfs-cli --upload I get an error "KeyError: 'response' Interrupted: 'response'
<issue_comment>username_3: TCX won't work, It need to be in in the FIT format. I've never create a FIT course file so I'm not sure what programs are available for creating one. @username_2 Could you post the full error and the log you get?
<issue_comment>username_2: I can't figure out how to generate a .FIT file from a TCX course (I'm trying to take a route I did on the watch and reduce the number of points so maybe when I have the watch follow the course in future it won't crash). Is it worth attempting to upload a non-FIT file, or will that always fail? In the mean time I've encountered a problem with simple downloading, so I'm going to open a new issue to deal with that first, then worry about uploads. |
<issue_start><issue_comment>Title: Support setting progress
username_0: * angular2-ladda can now be used with Rollup module bundler (with commonjs plugin)
* Added support for setting progress by binding to a number
* Enabled validation and code completion when setting a default style
* Simplified and optimized code for reading attributes and applying default styles
<issue_comment>username_0: After merging this, be sure to delete any previously generated index.js, index.d.ts, etc. files from the project root before publishing to npm.
<issue_comment>username_0: @username_1 Can you review this when you get a chance? It might be easiest to review the commits individually. Let me know if you have any questions.
<issue_comment>username_0: @username_1 Do you have time to review this? It adds support for setting progress percentages, and also makes it possible to use the package with the Rollup module bundler and Angular 4. I can answer any questions you have about the commits.
If you aren't able to maintain this package anymore, would you be open to adding me as a contributor? The package has a lot of users, so I feel it's important to keep it up-to-date to support current tooling.
Thanks!
<issue_comment>username_1: Hi @username_0 !
I've been traveling for about a month and it will take about a month more. So I don't have time to maintain this package now, but I'll be back after some time.
Thank you for great job you're doing and sure I'll add you as a collaborator! |
<issue_start><issue_comment>Title: Depth calibration failed with OpenCV 3
username_0: I'm getting an OpenCV error when executing the depth calibration with `rosrun kinect2_calibration kinect2_calibration chess9x6x0.024 calibrate depth`.
I'm using OpenCV 3.1. The other calibration functions work fine. I also tested everything with OpenCV 2.4 with the same files and I'm able to complete the calibration. This error only appears with OpenCV 3.
```
[ INFO] [DepthCalibration::readFiles] restoring file: 0098
[ INFO] [main] starting calibration...
[ INFO] [DepthCalibration::calibrate] frame: ./0000_depth.png
OpenCV Error: Assertion failed (confidence > 0 && confidence < 1) in run, file /home/me/Downloads/opencv-3.1.0/modules/calib3d/src/ptsetreg.cpp, line 178
terminate called after throwing an instance of 'cv::Exception'
what(): /home/me/Downloads/opencv-3.1.0/modules/calib3d/src/ptsetreg.cpp:178: error: (-215) confidence > 0 && confidence < 1 in function run
Aborted (core dumped)
```
I tracked the error to the file _kinect2_calibration.cpp_, lines 1054-1058:
```cpp
#if CV_MAJOR_VERSION == 2
cv::solvePnPRansac(board, points[index], cameraMatrix, distortion, rvec, translation, false, 300, 0.05, board.size(), cv::noArray(), cv::ITERATIVE);
#elif CV_MAJOR_VERSION == 3
cv::solvePnPRansac(board, points[index], cameraMatrix, distortion, rvec, translation, false, 300, 0.05, board.size(), cv::noArray(), cv::SOLVEPNP_ITERATIVE);
#endif
```
And I realized the the function `cv::solvePnPRansac` has different parameters in [OpenCV 2](http://docs.opencv.org/2.4/modules/calib3d/doc/camera_calibration_and_3d_reconstruction.html#solvepnpransac) and [OpenCV 3](http://docs.opencv.org/3.0-beta/modules/calib3d/doc/camera_calibration_and_3d_reconstruction.html#solvepnpransac).
More concretely, the 10th parameter is different. In OpenCV 2.4 it is `int minInliersCount=100`, which is a "big" number. However, in OpenCV 3 it is `double confidence=0.99`, hence the assertion that checks if the confidence is between 0 and 1 fails.
I was able to solve the issue by modifying line 1057 of _kinect2_calibration.cpp_:
```cpp
cv::solvePnPRansac(board, points[index], cameraMatrix, distortion, rvec, translation, false, 300, 0.05, board.size(), cv::noArray(), cv::SOLVEPNP_ITERATIVE);
```
to
```cpp
cv::solvePnPRansac(board, points[index], cameraMatrix, distortion, rvec, translation, false, 300, 0.05, 0.99, cv::noArray(), cv::SOLVEPNP_ITERATIVE);
```
I'm passing 0.99 as confidence, instead of board.size(). However, I still don't know which confidence value should be chosen.
Is anyone else getting this error? I would appreciate if someone can replicate it.
<issue_comment>username_1: Hi,
thanks for the issue. I will fix the `cv::solvePnPRansac` call. But OpenCV 3 is not supported at the moment, because there are issues when using OpenCV 2 and 3 together. Is the kinect2_bridge running fine for you with OpenCV 3?
<issue_comment>username_0: I got errors (related to nodelet, I think) using the launch file provided, but I can execute it directly with `rosrun`:
```
rosrun kinect2_bridge kinect2_bridge _depth_method:=opengl _reg_method:=cpu
```
and it works fine.
I know OpenCV 3 is a nightmare, and I understand why it is not supported (I didn't know before). Thank you anyway.
For me, you can close the issue if there is no more to add.<issue_closed>
<issue_comment>username_1: fixed with 8315130933c60389956821baa091be6791e4a92c. |
<issue_start><issue_comment>Title: UPDATE UPSERT information after operation
username_0: Is there a way to know whether update upsert created the record or updated it?
We have a use case that needs separate operations on create and match - something like Neo4j's Merge - ON CREATE & ON MATCH.
How can we do such a thing in orientdb?
<issue_comment>username_1: What about Version of the record? Can you use that?
<issue_comment>username_0: Right that is some information, but then again it would require me to do 2 queries. Wouldn't it defeat the purpose of UPSERT all together?
If I want to set a particular field only on 'modify' or 'create' and not both then I would require 2 queries for this, it would be the same case as selecting the count of existing records and if it is zero then do something for create else update.
Am I right? Any other alternative than this?
On a side-note..while googling about this I came across multiple requests by other users for this same feature. Can we expect any work on this in the near future? It is quite an important distinction between Orientdb and it's competitors that do offer this query support.
<issue_comment>username_0: @username_1 Thanks for the quick response, I just ran a few tests, and version no. is not very reliant for this purpose. Consider the following scenario:
select from Email where email='test100.hello.com'
Returned 0 records
create vertex Email set email='test100.hello.com'
Returned 1 record @version: 2
delete vertex Email where email='test100.hello.com'
returned 1 row @version: 0 ( successfully deleted )
create vertex Email set email='test100.hello.com'
Returned 1 record @version: 3
Even though the record was newly created, it incremented the version as if it was updated.
<issue_comment>username_1: As OrientDB user for your purposes I would reconsider
UPDATE SET ..... UPSERT RETURN BEFORE
You won't get previous record content for record creations, based on that you can change logic.
Additionaly to that I would use
UPDATE SET X = IFNULL(X,1) ....
which updates only non-present field (for new record)
Does it solve partially your needs?
<issue_comment>username_0: It does, that's a great help.
Thanks :)
Do also consider a simpler syntax for a combination of these commands as I guess this is a nice and important feature.<issue_closed> |
<issue_start><issue_comment>Title: Made a mistake, angled brackets include weren't working
username_0: Angled brackets include were not working in my project but quotes were.
I found an error in `Findecst`.
<issue_comment>username_0: I haven't fundamentally changed how this works from `vrm_pp`, but it does a really annoying thing where it finds the `ecst` repo in my `~/git/ecst` before it finds `./extlibs/ecst` ! I had a brief look and the search order looked OK to me so I just deleted `~/git/ecst` and put it down to shenanigans. |
<issue_start><issue_comment>Title: Include raw as an option
username_0: Raw packet hex data can only be printed as text, PostScript, JSON or EK JSON
<issue_comment>username_0: ping
<issue_comment>username_1: I don't think that if someone picks raw=True it should immediately enable JSON mode. That is not the behaviour a user might expect. Both would have to be true.
<issue_comment>username_0: @username_1 fixed
<issue_comment>username_1: Sorry to bug about this, but I really think there should be some sort of exception if it's enabled and json_mode isn't. Otherwise, a user might do raw=True and it won't work and he won't know why (if he hasn't read the docks). An assertion or exception should be good |
<issue_start><issue_comment>Title: one is not x plus one
username_0: import numpy poly1d
poly1d([1,1]) == poly1d([1])
but it shouldn't, because one does not equal x plus one.
The definition of equality for poly1d is
def __eq__(self, other):
if self.coeffs.shape != other.coeffs.shape:
return False
return (self.coeffs == other.coeffs).all()
And this seems correct. Indeed,
poly1d([1]).coeffs.shape != poly1d([1,1]).coeffs.shape
is True.
So I can't see how this bug happens.
I am using version 1.8.2 of numpy
and version 2.7.9 of python<issue_closed>
<issue_comment>username_1: ```
n [1]: poly1d([1,1]) == poly1d([1])
Out[1]: False
```
Looks fixed in recent versions. 1.8 is pretty old at this point...
<issue_comment>username_0: Thanks |
<issue_start><issue_comment>Title: CouchDB indexing blocking atlas loading
username_0: Could we experiment with the stale=update_after option when querying CouchDB to see if we can avoid waiting for it to reindex? This is affecting active atlases where data is being imported and changed in regular use. |
<issue_start><issue_comment>Title: port labeling, symbol sorting, and reparsing overscaled tiles
username_0: this ports a bunch of stuff from -js:
- sort overlapping symbols in y-direction https://github.com/mapbox/mapbox-gl-js/pull/1184
- better and faster labeling https://github.com/mapbox/mapbox-gl-js/pull/1079
- reparse overscaled tiles https://github.com/mapbox/mapbox-gl-js/pull/1005
This is finally close to merging. Sorry about the amount of changes in one pull request.
TODO:
- [x] fix resource loader test failure https://travis-ci.org/mapbox/mapbox-gl-native/jobs/64636907#L1660
- [ ] testing on an actual ios device. anyone have time to do this?
I haven't noticed any stability problems but a lot has changed so this will probably introduce new bugs.
<issue_comment>username_1: @username_0 This doesn't work very well on device at the moment. It seems to stop loading tiles after a few gestures.
<issue_comment>username_0: @username_1 I just pushed 1f5fb1d which I think should fix that
<issue_comment>username_1: Yep, feels good now!
<issue_comment>username_2: Same here. :+1: @username_0
<issue_comment>username_0: A couple things should be reviewed:
- [ ] redoPlacement uses workers safely
- [Source::update calls VectorTileData::redoPlacement](https://github.com/mapbox/mapbox-gl-native/blob/1f5fb1db4afae8a8bae02c42870f6f9e5e5f7905/src/mbgl/map/source.cpp#L516-L518) in the map thread.
- [VectorTileData::redoPlacement](https://github.com/mapbox/mapbox-gl-native/blob/1f5fb1db4afae8a8bae02c42870f6f9e5e5f7905/src/mbgl/map/vector_tile_data.cpp#L121) checks if placement needs to be redone and that there is no worker working on the tile currently. If there isn't, it calls [VectorTileData::workerRedoPlacement](https://github.com/mapbox/mapbox-gl-native/blob/1f5fb1db4afae8a8bae02c42870f6f9e5e5f7905/src/mbgl/map/vector_tile_data.cpp#L138) in a worker which calls [VectorTileData::endRedoPlacement](https://github.com/mapbox/mapbox-gl-native/blob/1f5fb1db4afae8a8bae02c42870f6f9e5e5f7905/src/mbgl/map/vector_tile_data.cpp#L149) in the main thread when it is done.
- [Source::tileLoadingCompleteCallback calls VectorTileData::redoPlacement] (https://github.com/mapbox/mapbox-gl-native/blob/1f5fb1db4afae8a8bae02c42870f6f9e5e5f7905/src/mbgl/map/source.cpp#L570) in the map thread because the angle may have changed while the tile was loading.
- To avoid unsafely overwriting existing buffers while they are being used, SymbolBucket stores the new buffers separately in [`renderDataInProgress`](https://github.com/mapbox/mapbox-gl-native/blob/1f5fb1db4afae8a8bae02c42870f6f9e5e5f7905/src/mbgl/renderer/symbol_bucket.cpp#L343). `renderDataInProgress` gets moved to `renderData` later [from the main thread](https://github.com/mapbox/mapbox-gl-native/blob/1f5fb1db4afae8a8bae02c42870f6f9e5e5f7905/src/mbgl/map/vector_tile_data.cpp#L153).
- [ ] https://github.com/mapbox/mapbox-gl-native/commit/1e044a175cbb714182709b9929fe4a74bfa42322 which implements reparsing tiles for high zoom levels changes `TileID(z, x, y)` to `TileID(z, x, y, sourceZ)`. The first z is used to differentiate overparsed tiles that share the same actual vectortile. The second is used to download the data. For example if tile 15/200/300 needs to be overscaled for z17, the TileID would be (17, 200, 300, 15). Does this seem ok?
- [ ] @username_0 should double check that out-of-bounds accesses aren't possible in a couple spots where vertices are accessed by index
<issue_comment>username_3: 1.) Is the following scenario handled?
a. worker starts doing replacement.
b. Tile becomes obsolete before worker thread doing replacement finishes.
c. worker should not call endRedoPlacement or somehow VectorTileData pointer should not be invalid when endRedoPlacement is called.
This will work in most cases because TileCache will hold a shared_ptr to VectorTileData. However this needs to work even when TileCache size is zero.
2.) Also does it make sense to do some testing with TileCache size as zero? Tile Cache may hide some bugs because it holds a shared_ptr to TileData.
<issue_comment>username_0: 1) Yes, the TileData destructor calls TileData::cancel, which [resets](https://github.com/mapbox/mapbox-gl-native/blob/8c2dbcec5aa1b7499ddcbb01da82b00a2db2876b/src/mbgl/map/tile_data.cpp#L73) workRequest, which [calls](https://github.com/mapbox/mapbox-gl-native/blob/8c2dbcec5aa1b7499ddcbb01da82b00a2db2876b/src/mbgl/util/work_request.cpp#L13) WorkTask::cancel. WorkTask::cancel [clears](https://github.com/mapbox/mapbox-gl-native/blob/8c2dbcec5aa1b7499ddcbb01da82b00a2db2876b/src/mbgl/util/work_task.cpp#L34) the callback (endRedoPlacement) so that it doesn't get called when WorkTask::runAfter gets called later (it's definitely later, because it gets called on the same thread).
2) Good idea. I tested it for a bit with cache size 0 and will keep that in mind for the future
thanks for the suggestions @username_3
<issue_comment>username_4: So far so good with visual inspection.
<issue_comment>username_1: I don't see any glaring issues, but it's impossible to say much about thread safety conclusively, given the current structure. The TileData class hierarchy and worker code needs an overhaul to cleanly separate data on the worker thread from data on the map thread.
<issue_comment>username_3: @username_0 Does redoPlacement need to be called at higher zoom levels? Specifically when (map's current zoom > SourceInfo.maxzoom + 1). Or is this handled in some other way to ensure good symbol density at higher zoom levels? (CollisionTileData.maxScale is const 2.0)
<issue_comment>username_0: @username_3 this is handled in some other way. mostly by https://github.com/mapbox/mapbox-gl-native/commit/1e044a175cbb714182709b9929fe4a74bfa42322. For zoom levels > maxzoom, tiles from the maximum zoom level are reparsed completely. Layout properties can be functions and in order for these functions to work it needs to reparse those layers at every zoom level. It reparses entire tiles instead of only layers that need it because that was easier to implement. |
<issue_start><issue_comment>Title: JSDOC: Add documentation generation to package.info
username_0: We can add generation of JSDOC derived documentation via docdash: https://github.com/clenemt/docdash
"script": {
"generate-docs": "node_modules/.bin/jsdoc -c jsdoc.json"
}
This way we can have standalone documentation showing how to use the library. |
<issue_start><issue_comment>Title: Feat/search suggestions
username_0: Please test before merging. Things may be broken.
<issue_comment>username_1: My testing so far has show it to be working "as expected"; not to say there aren't any hiccups but I'm pretty happy with the work done. The move to templates for the search system leaves me a little out of my depth for now but let's see how this goes.
Thanks for the improved search suggestions system :+1: |
<issue_start><issue_comment>Title: Fix dependency chains in posix.mak
username_0: 1) The idea of a build system is that only the files that need to be updated, should be. This PR is an attempt to fix this as since the move to `generated` this isn't the case anymore (at the moment every `build` is a _full_ rebuild
2) The generated `dmd.conf` needs to be put into the `generated` folder as well
3) There was an issue with the dependency chain of the C files. An update to the source files _didn't_ trigger the dependency chain. The reason for this problem was that the generated dependency files need to be included _before_ their rule)
4) The optagen output target was completely broken
- the targets didn't exist
- if run in parallel, it would run the tool was well in parallel
For more infos on (4) see [this guide](https://www.gnu.org/software/automake/manual/html_node/Multiple-Outputs.html#Multiple-Outputs) or [this SO question](http://stackoverflow.com/questions/2973445/gnu-makefile-rule-generating-a-few-targets-from-a-single-source-file).
CC @RazvanN7
<issue_comment>username_0: @andralex while I really appreciate the fast review and intend to merge I think we shouldn't overrun the usual one-day "cry loud if you can" waiting period, especially as this is non-trivial and there were some issues with CircleCi.
(though they are caused by CircleCi expecting `dmd.conf` to be in `src` ...
<issue_comment>username_0: I think I have won the fight against CircleCi & Travis :)
<issue_comment>username_1: How do you generate the `[SQUASH]` commit messages? Tried git-rebase with `--autosquash` but it doesn't understand those, nor the `squash! [SQUASH] message` ones create with `git commit --squash HEAD~1`.
<issue_comment>username_1: Except for complex PR reviews the value of adding new changes as tiny commits and later squashing seems very small compare to all the complications it adds.
<issue_comment>username_0: Well again, the idea is that one doesn't have to look at the entire diff if only one line changed and again with `auto-merge-squash` there wouldn't be extra costs involved. |
<issue_start><issue_comment>Title: Add JsonFullValidator which validates the entire record
username_0: This PR adds a `JsonFullValidator` class which validates an entire record.
I also added some tests although their style is a little different because it was harder to write the tests with everything mocked out than it was to write them with just some real objects.
<issue_comment>username_1: Hi @username_0,
This is an interesting idea but I don’t think it fits with the goal of the gem. The use case is to validate **JSON** data with a JSON schema. A whole ActiveRecord record is not JSON data. Unfortunately, I don’t think we want to support this.
I’m going to close this PR but feel free to comment on it. Thank you! |
<issue_start><issue_comment>Title: Expose CloseButton() in imgui.h
username_0: Hey! Usually live on Bitbucket, but registered here just for this. So we talked on Twitter and you already know half the details, but I guess I'll describe it from the start for anyone else reading this as well.
My suggestion is to **put a public version of `CloseButton()` in _imgui.h_**, and I'll explain why I think this would be nice below. Just giving you the context at the very beginning of this issue before I babble on!
Thought I'd need slightly more than Twitter's 140 characters, but didn't expect it to guess this lengthy, so I hope it's palatable and least misses out on no information.
---
### Context
I'm working on a docking system with tabs, and today I added **buttons to close the tabs**. I wanted them to be **consistent with those used to close windows**, and to have their style and colours controlled by the same style variables, so I dug into ImGui to find out how those are added, and found the function `CloseButton()` in _imgui_internal.h_.
Here's a picture of my implementation using that function to add a close button to the tabs:
The tabs are meant to be consistent with the title bars of windows, and use the same colour.
---
### Issue
Obviously _imgui_internal.h_ is perfectly accessible, but as you said on Twitter it's not "designed for public consumption", and this particular function reflects this, as I had to specify the button's position globally when using it, rather than having the button show up in a relative position like when I call one of the exposed functions, like the regular `Button()`.
I also had to calculate the size (based on radius and padding and stuff) manually, the same way the internal window code for ImGui does it, which would be nice to have abstracted away somehow as well in case of a public version.
Perhaps with a default size if the argument is omitted, so that the size is generally consistent, and only customised if really necessary... I believe my tabs have smaller buttons than the windows ATM, which also makes me think it'd be nice to access title bar size and padding to use for the tabs as well, but I may be getting off topic now, unless this should be generalised into some title bar API rather than just a close button one instead, heh...
All in all, the relevant code for adding the button to the tab looks something like this, which is quite verbose even excluding the additional call to generate the ID for the button:
```
constexpr float padding{4.0f};
const float radius{heightBarTabs * 0.5f - padding};
auto pos(ImGui::GetItemRectMin());
pos.x = ImGui::GetItemRectMax().x;
pos.x -= radius + padding;
pos.y += radius + padding;
if (ImGui::CloseButton(IDButton, pos, radius))
```
---
### Suggestion
So I just wanted to suggest **putting a public version of `CloseButton()` in _imgui.h_** for "public consumption" so that anybody who wants to add something that needs a close button can easily make it consistent with the rest of the UI, as exemplified by my tabs.
(or perhaps indeed something bigger for title bar stuff in general, which my ramblings started getting into halfway instead...)
I think it can probably be useful for immediate one-time inline stuff too, and not necessarily just full-blown custom widgets.
---
This may be simple enough that I could add it to _imgui.h_ myself and submit a pull request through this new account, but yeah. Would like to hear your thoughts first, of course! c:
So, indeed, how do you feel about this suggestion?
<issue_comment>username_1: Hello @username_0,
Sorry I didn't get to answer this earlier. `CloseButton()` is exposed in imgui_internal.h for now, so you may include that for now? Currently it would be a rather odd api to add to imgui.h since it doesn't lay itself out, takes an ImGuiID etc. so it feels it would only be used by custom widgets?
(Also adding a link to #261 since this relates to Tabs)
<issue_comment>username_1: Was scanning old issues I think I'll be closing this for now.
Mostly because your initial intent was to use in a custom docking system, and now we have both TabBar (in master) and Docking (in a branch), reducing pressure for that.
I _do_ agree this would be useful eventually but I would rather wait until we have a concrete use case that warrant creating a public api. There's a various subtleties (overlap, navlayer) involved and even until recently I've been changing things about the `CloseButton()` function, which makes me not eager to make it public unless absolutely necessary.
Thanks again for the suggestion!<issue_closed> |
<issue_start><issue_comment>Title: Bandwidth after resize
username_0: We do not consider resizes in bandwidth accounting. That means a user could resize at the end of the month or something to get more bandwidth, but only pay for the new VM plan for a small period of time. Solution is to maybe reset the VM time created to the time of the resize and then adjust user's additional bandwidth column appropriately (same column that's incremented when user deletes VM).<issue_closed> |
<issue_start><issue_comment>Title: VS Code debugger doesn't exit after program run
username_0: When debugging the following simple project (only a `gulpfile.js`), the debugger doesn't exit after the last line of code has been executed:
[Gulp4ParallelTest.zip](https://github.com/Microsoft/vscode/files/142778/Gulp4ParallelTest.zip)
----
```json
{ "version": "0.2.0"
, "configurations":
[ { "name": "Gulp"
, "type": "node"
, "request": "launch"
, "program": "${workspaceRoot}/node_modules/gulp/bin/gulp.js"
, "stopOnEntry": false
, "args": []
, "cwd": "${workspaceRoot}"
, "runtimeExecutable": null
, "runtimeArgs":
[ "--nolazy"
]
, "env":
{ "NODE_ENV": "development"
}
, "externalConsole": false
, "sourceMaps": false
, "outDir": null
}
]
}
```
<issue_comment>username_1: @username_0 the debugger only exits if node exits. Node programs have the tendency not to exit under debugging. You will find the same behaviour with node-inspector so this is not a VS Code specific problem.
<issue_comment>username_0: Thank's for enlightening me. Would you suggest this is more or less a Node bug? Then I would tend to create an issue there.
I feel it's quite confusing not to safely know whether a program has finished or not. Don't you, too?
<issue_comment>username_1: @username_0 I'm not sure whether this is a node bug. It might be difficult for the node runtime to decide whether there are any events expected and the runtime needs to continue running or not. Yes, please file an issue with node. I'm interested in their response.<issue_closed> |
<issue_start><issue_comment>Title: There is no sweet visualization
username_0: There should be.
https://developers.google.com/chart/interactive/docs/gallery/sankey
looks like a very quick-and-dirty way to get going maybe.
Solution would be an index.html, some assets in assets/, and JS that loads the yml and makes that diagram with companies where people worked together clustered together. |
<issue_start><issue_comment>Title: alias default writes a 1 into the default alias
username_0: curtis:~/ $ nvm alias default v0.12.7 [11:42:36]
default -> v0.12.7
curtis:~/ $ nvm ls [11:42:39]
-> v0.12.7
system
default -> 1 v0.12.7 (-> N/A)
starting a new shell yields:
N/A: version " 1 stable" is not yet installed
^C%
<issue_comment>username_0: the issue is the 'cat' command is not being escaped properly and my shell alias is being used instead.
this ensures that cat is being used predictably:
command cat
\cat
<issue_comment>username_1: Thanks - it's a horrible idea to alias shell builtins, but we definitely should still be robust against it. Will fix shortly.
<issue_comment>username_0: It's a horrible idea to alias shell builtins?
Since when could one assume that people do anything but alias shell
commands?
<issue_comment>username_1: It's a horrible idea precisely because scripts often assume that builtins aren't aliased :-) all my aliases are new names.
<issue_comment>username_0: That is a good offense on avoiding script breakage :D
My take is to escape scripts internally to avoid conflict and prevent hacks
from alterning behaviour intentionally or unintentionally :D<issue_closed>
<issue_comment>username_1: This is now resolved - it'll be an ongoing issue that code review will have to catch. |
<issue_start><issue_comment>Title: Dashboard/Project info: add link to Direct project, visible for Manager and Copilot roles
username_0: @username_3 could you please create a quick design for this and attach here? We can't put this link in the "Links" section b/c it will only be visible to "internal" roles. Thank you.
cc @ajefts
<issue_comment>username_1: Quick crazy idea: we could have pinned links, colored or icon distinguished. To keep the same semantics.
<issue_comment>username_0: @username_1 it's a good idea... but in the "nice to have" territory, so we won't implement this month
<issue_comment>username_2: We can add these links the QuickLinks section (icon next to the the nav links) in topbar.
<issue_comment>username_0: @username_2 we need a project-level link, which will be unique per project, so we'll quickly have a lot of links... to many to include in the nav.
<issue_comment>username_3: 
Speck and Sketch files here: https://drive.google.com/drive/folders/0B1LbmKnex41fZnJ4Rkozcl94VUk?usp=sharing
<issue_comment>username_1: Easy cake.<issue_closed> |
<issue_start><issue_comment>Title: MoveToWorld is unstable
username_0: Client version: 1.11.2
Server OS: Linux
Commit id: (Found at the top of the server's console output)
### Expected behavior
MoveToWorld should not crash the server.
### Actual behavior
If MoveToWorld is used many times, it crashes the server.
```
[14:43:44] D: | Cuberite has encountered an error and needs to close
[14:43:44] Details | SIGSEGV: Segmentation fault
[14:43:44] Cuberite local build build id: Unknown
[14:43:44] from commit id: approx: 9fa8477b73ce878c631ba875ec10746b4143ba75 built at: approx: 2017-03-10 01:45:11 -0600
./Cuberite(_Z15PrintStackTracev+0x1c)[0xa120ac]
./Cuberite[0x9efa89]
/lib/x86_64-linux-gnu/libc.so.6(+0x350e0)[0x7fd7c72400e0]
./Cuberite(_ZNSt4listIP7cEntitySaIS1_EE5beginEv+0x14)[0x8db874]
./Cuberite(_ZNSt4listIP7cEntitySaIS1_EE6removeERKS1_+0x20)[0x8dd130]
./Cuberite(_ZN6cChunk12RemoveEntityEP7cEntity+0x44)[0x8d7ac4]
./Cuberite[0xd87f95]
./Cuberite[14:43:44] Kicking player a for "Server is restarting. Please wait..."
[0xd87b82]
./Cuberite(_ZNKSt8functionIFvR6cWorldEEclES1_+0x58)[0x9dbb58]
./Cuberite(_ZN6cWorld15TickQueuedTasksEv+0x2ac)[0x9cfc7c] [130/1961]
./Cuberite(_ZN6cWorld4TickENSt6chrono8durationIlSt5ratioILl1ELl1000EEEES4_+0x5c6)[0x9c0fb6]
./Cuberite(_ZN6cWorld11cTickThread7ExecuteEv+0xa5)[0x9c0945]
./Cuberite(_ZN9cIsThread9DoExecuteEv+0x39)[0xa04f59]
./Cuberite(_ZNKSt7_Mem_fnIM9cIsThreadFvvEEclIJEvEEvPS0_DpOT_+0x72)[0xa06642]
./Cuberite(_ZNSt12_Bind_simpleIFSt7_Mem_fnIM9cIsThreadFvvEEPS1_EE9_M_invokeIJLm0EEEEvSt12_Index_tupleIJXspT_EEE+0x42)[0xa065a2]
./Cuberite(_ZNSt12_Bind_simpleIFSt7_Mem_fnIM9cIsThreadFvvEEPS1_EEclEv+0x15)[0xa06555]
./Cuberite(_ZNSt6thread5_ImplISt12_Bind_simpleIFSt7_Mem_fnIM9cIsThreadFvvEEPS3_EEE6_M_runEv+0x1c)[0xa0652c]
/usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0xb6970)[0x7fd7c7a9f970]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x8064)[0x7fd7c75be064]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7fd7c72f362d]
[14:43:44] D: | Cuberite has encountered an error and needs to close
[14:43:44] Details | SIGABRT: Server self-terminated due to an internal fault
[14:43:44] Cuberite local build build id: Unknown
[14:43:44] from commit id: approx: 9fa8477b73ce878c631ba875ec10746b4143ba75 built at: approx: 2017-03-10 01:45:11 -0600
./Cuberite(_Z15PrintStackTracev+0x1c)[0xa120ac]
./Cuberite[0x9efaed]
/lib/x86_64-linux-gnu/libc.so.6(+0x350e0)[0x7fd7c72400e0]
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0x37)[0x7fd7c7240067]
/lib/x86_64-linux-gnu/libc.so.6(abort+0x148)[0x7fd7c7241448]
./Cuberite[0x9efa8e]
/lib/x86_64-linux-gnu/libc.so.6(+0x350e0)[0x7fd7c72400e0]
./Cuberite(_ZNSt4listIP7cEntitySaIS1_EE5beginEv+0x14)[0x8db874]
./Cuberite(_ZNSt4listIP7cEntitySaIS1_EE6removeERKS1_+0x20)[0x8dd130]
./Cuberite(_ZN6cChunk12RemoveEntityEP7cEntity+0x44)[0x8d7ac4]
./Cuberite[0xd87f95]
./Cuberite[0xd87b82]
./Cuberite(_ZNKSt8functionIFvR6cWorldEEclES1_+0x58)[0x9dbb58]
./Cuberite(_ZN6cWorld15TickQueuedTasksEv+0x2ac)[0x9cfc7c]
./Cuberite(_ZN6cWorld4TickENSt6chrono8durationIlSt5ratioILl1ELl1000EEEES4_+0x5c6)[0x9c0fb6]
./Cuberite(_ZN6cWorld11cTickThread7ExecuteEv+0xa5)[0x9c0945]
./Cuberite(_ZN9cIsThread9DoExecuteEv+0x39)[0xa04f59]
./Cuberite(_ZNKSt7_Mem_fnIM9cIsThreadFvvEEclIJEvEEvPS0_DpOT_+0x72)[0xa06642]
./Cuberite(_ZNSt12_Bind_simpleIFSt7_Mem_fnIM9cIsThreadFvvEEPS1_EE9_M_invokeIJLm0EEEEvSt12_Index_tupleIJXspT_EEE+0x42)[0xa065a2]
./Cuberite(_ZNSt12_Bind_simpleIFSt7_Mem_fnIM9cIsThreadFvvEEPS1_EEclEv+0x15)[0xa06555]
./Cuberite(_ZNSt6thread5_ImplISt12_Bind_simpleIFSt7_Mem_fnIM9cIsThreadFvvEEPS3_EEE6_M_runEv+0x1c)[0xa0652c]
/usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0xb6970)[0x7fd7c7a9f970]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x8064)[0x7fd7c75be064]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7fd7c72f362d]
Aborted
```
### Steps to reproduce the behavior
Include this code is a plugin:
```
for i=1,10 do
Player:MoveToWorld("world_nether")
end
```
<issue_comment>username_1: In the api doc `MoveToWorld` is obsolete, please try [ScheduleMoveToWorld](https://api.cuberite.org/cPlayer.html#ScheduleMoveToWorld_1).
<issue_comment>username_0: It doesn't happen with ScheduleMoveToWorld, but ScheduleMoveToWorld doesn't send the respawn packet. ScheduleMoveToWorld was supposed to be removed in https://github.com/cuberite/cuberite/pull/3401.
<issue_comment>username_2: The api doc is questionable. See this: https://forum.cuberite.org/thread-2550.html
I thought I stabilized this long ago...
<issue_comment>username_2: Does this require very rapid moves to crash? Or do slow moves (e.g. 1 per sec) also crash?
<issue_comment>username_0: I haven't been able to find a consistent way to reproduce this when using MoveToWorld normally. Sometimes I've walked into a portal with normal walking speed, and the server has crashed.
I'll test this thoroughly tomorrow.
<issue_comment>username_2: Doesn't stepping through portals use `ScheduleMoveToWorld`?
<issue_comment>username_3: Maybe not, at least for end portals: https://github.com/cuberite/cuberite/blob/master/src/Entities/Entity.cpp#L1519
Have you found that any particular kind of portal is causing occasional crashes?
<issue_comment>username_2: There's a `ScheduleMoveToWorld` in `cNetherPortalScanner`.
<issue_comment>username_0: Sorry, I should have explained better. The portals I'm talking about are cCuboid areas that move you to another world once you step inside them, not the built-in nether and end portals. I'm going to double check today that my implementation of them isn't flawed.
<issue_comment>username_0: I've confirmed that the random crashes are not caused by MoveToWorld itself. The part of the code in my plugin that moves the player to another world is on rare occasions called twice. Returning true prevents this.
MoveToWorld is only a problem when used twice, which shouldn't happen in other plugins that use it.
<issue_comment>username_0: I tested older builds of Cuberite to see when the problem started occurring, and commit 5625598afac8db3b35d790b8d1228997b228f228 seems to be the problem.
<issue_comment>username_3: Predictably, that's when MoveToWorld went from being synchronous to being a queued lambda task.
<issue_comment>username_3: Is it possible for a player to be teleported (MoveToWorld), and then destroyed in the same tick (before the MoveToWorld lambda runs)?
<issue_comment>username_2: I no longer remember the details. Sorry. The code was/is messy so I suspected I probably missed some edge cases.
One argument against `ScheduleMoveToWorld` and in favor of the `MoveToWorld` style was that every functionality had been introducing its own cEntity/cMonster/cWorld/cChunk member variables, and the classes gradually became bloated, so moving as many things to lambdas as possible would remove lots of member variable bloat.
<issue_comment>username_3: That sounds good to me, ScheduleMoveToWorld requires five different member variables. All that would have to happen is MoveToWorld would have to work if it's called multiple times, I don't know how to do that though.
If I recall correctly the problem is that the player isn't yet added to the new world, so maybe if the lambda checks whether the player is actually in the new world? Then if the player isn't in the new world yet, it would have to remove itself from the PlayersToAdd queue of the old new world (the world that it was going to go to until MoveToWorld was called again).
<issue_comment>username_4: One alternative would be a pod type `sMoveToWorld` that only exists if a move is scheduled and then the bloat becomes a single pointer.
<issue_comment>username_2: Possible solution:
- Remove ScheduleMoveToWorld and its member variables
- Add an atomic m_WorldTravelCnt.
```
func MoveToWorld(...)
{
int lambaCnt = ++m_WorldTravelCnt;
Schedule[lambaCnt, etc,](){
if (lambdaCnt != player.m_WorldTravelCnt) return; // not the latest call
...etc...
}
}
<issue_comment>username_3: There may be a possible race condition, if MoveToWorld is called after the previous MoveToWorld call lambda finished but before the player is added to the new world and is still sitting in the PlayersToAdd queue (incidentally, I think this race condition existed with ScheduleMoveToWorld as well...).
<issue_comment>username_4: To expand on my idea, this is how it could be used:
``` cpp
void MoveToWorld(cWorld * a_World, bool a_ShouldSendRespawn, Vector3d a_NewPosition)
{
if (m_MoveToWorld != nullptr)
{
// A world change is already scheduled but not executed, just update the details
m_MoveToWorld->World = a_World;
m_MoveToWorld->ShouldSendRespawn = a_ShouldSendRespawn;
m_MoveToWorld->NewPosition = a_NewPosition;
return;
}
// No world change scheduled, have to spawn a task as well.
m_MoveToWorld = cpp14::make_unique<sMoveToWorld>(a_World, a_ShouldSendRespawn, a_NewPosition);
GetWorld()->QueueTask(
[this](cWorld &)
{
DoMoveToWorld(*m_MoveToWorld);
m_MoveToWorld.reset();
}
);
}
```
Note that Here `DoMoveToWorld` would do the move immediately and not schedule another task. Also, only one world move task ever exists at a time and the only bloat added to `cEntity` is a single pointer which is similar to adding a counter.
<issue_comment>username_3: This may not actually happen in practice, I don't know, but if one world's tick is running slow, then another world could run multiple ticks before the slow world actually gets around to adding the new player. If MoveToWorld is called once a tick on the fast world (to transfer someone to the slow world), then DoMoveToWorld could be called on a player that hasn't yet been added to the slow world.
<issue_comment>username_4: Perhaps the task could check `IsTicking()` and if not just re-queue itself for next tick? Whether or not this issue could come up would depend on whether the players are added before or after the tasks are executed in `cWorld::Tick`. Different worlds ticking at different rates shouldn't be an issue though as the task is always queued on the world that the player is either in or queued to be added to.
<issue_comment>username_2: I have a feeling obscure problems will still lurk. Heh. This is complex. Like I said before, I think one needs to take a step back and carefully inspect the entire system and find a more elegant approach, potentially overhauling the whole thing, including `IsTicking` itself, which is kind of a hack that solved some specific issues and was not thought out entirely.<issue_closed> |
<issue_start><issue_comment>Title: Add delay to software reset.
username_0: Should fix issue 2.
<issue_comment>username_1: Longer-term, I'm on the fence about either:
a) exposing the reset delay as a defaulted argument so that if 2ms turns out to be insufficient in some cases, this can be over-ridden without changing the library
or
b) polling (with a timeout or maximum number of 1ms delays, for instance)
For the moment, anyway, this fix should be fine.
Allowing a selectable I2C interface is a useful enhancement. |
<issue_start><issue_comment>Title: Add Non-.gov URL API
username_0: Add http://search.digitalgov.gov/developer/govt-urls.html to http://open.gsa.gov/developer.
A potential slug is, "This API allows you to find government URLs that don't end in .gov or .mil. URLs are searchable by keyword, state, or scope."
Copied from open.gsa.gov
https://github.com/GSA/open.gsa.gov/issues/62<issue_closed> |
<issue_start><issue_comment>Title: lua@5.3: import from homebrew/versions.
username_0: - [x] Have you followed the [guidelines for contributing](https://github.com/Homebrew/homebrew-core/blob/master/CONTRIBUTING.md)?
- [x] Have you checked that there aren't other open [pull requests](https://github.com/Homebrew/homebrew-core/pulls) for the same formula update/change?
- [x] Have you built your formula locally with `brew install --build-from-source <formula>`, where `<formula>` is the name of the formula you're submitting?
- [x] Does your build pass `brew audit --strict <formula>` (after doing `brew install <formula>`)?
-----
<issue_comment>username_1: @username_0 Nice work! Would also be great to rename `lua51` accordingly, too, and make a PR in homebrew/versions to migrate this formula there too.
<issue_comment>username_0: @username_1
- `--with-default-names` has been removed from `lua@5.3`.
- `lua51` has been renamed to `lua@5.1`, `--universal` has been removed and some minor cleanups have been applied in order to match `lua@5.3`.
- A pull request Homebrew/homebrew-versions#1515 has been submitted to migrate `lua53`.
<issue_comment>username_0: - `--universal` has been removed from `lua`.
- Formulae that depend on `lua51` have been updated to depend on `lua@5.1` with revision bumps.
<issue_comment>username_2: FWIW, I'd done most of this locally already, but the new `luarocks` update which needs to be applied seems to have "issues" with accepting suffixed `lua` executables. But at some point that will need doing. Not a blocker here, just an FYI.
<issue_comment>username_0: The misplaced `revision` in `radare2.rb` has been fixed. 🙈
<issue_comment>username_1: Thanks again for your contribution to Homebrew! Without people like you submitting PRs we couldn't run this project. You rock! |
<issue_start><issue_comment>Title: An error in "ubuntu-setup/README.md"
username_0: Line 18 `(cd tools/ubuntu-setup && source all.sh)` fails in Ubuntu 14.04.
```
$ source all.sh
*** installing basics
[1]+ Stopped "$1/$2"
2139 finished with status 150
*** installing python dependences
[2]+ Stopped "$1/$2"
2142 finished with status 150
*** installing java
[3]+ Stopped "$1/$2"
2144 finished with status 150
```
The reason is that these scripts running in the background needs the input of sudo password.
So I suggest to change "source all.sh" to "sudo ./all.sh".
It works fine for me.<issue_closed> |
<issue_start><issue_comment>Title: Review and update CI testing strategy and runners
username_0: We currently have 3 separate test files that each run independently, and until recently all used macos. Macos is considerably more expensive than ubuntu, and running 3 separate processes with no build cache means a lot of time downloading and rebuilding the same code.
We've done quite a bit of optimization with client code to reduce cost. I think we should apply that thinking here.
The iOS build is probably unnecessary because it just runs xcode to construct the final app, but does no testing. We could keep that step but put it in a "native test" file, that otherwise just runs the kotlin tests for iOS. That would significantly reduce build times. The other file would run Android tests. |
<issue_start><issue_comment>Title: NullPointerException in maven plugin 3.2.1
username_0: Here's the stacktrace -
org.apache.maven.lifecycle.LifecycleExecutionException: Failed to execute goal org.flywaydb:flyway-maven-plugin:3.2.1:migrate (flyway-migrate) on project cts-integration-test: java.lang.NullPointerException
at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:216)
at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:153)
at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:145)
at org.apache.maven.lifecycle.internal.LifecycleModuleBuilder.buildProject(LifecycleModuleBuilder.java:116)
at org.apache.maven.lifecycle.internal.LifecycleModuleBuilder.buildProject(LifecycleModuleBuilder.java:80)
at org.apache.maven.lifecycle.internal.builder.singlethreaded.SingleThreadedBuilder.build(SingleThreadedBuilder.java:51)
at org.apache.maven.lifecycle.internal.LifecycleStarter.execute(LifecycleStarter.java:120)
at org.apache.maven.DefaultMaven.doExecute(DefaultMaven.java:355)
at org.apache.maven.DefaultMaven.execute(DefaultMaven.java:155)
at org.apache.maven.cli.MavenCli.execute(MavenCli.java:584)
at org.apache.maven.cli.MavenCli.doMain(MavenCli.java:216)
at org.apache.maven.cli.MavenCli.main(MavenCli.java:160)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:606)
at org.codehaus.plexus.classworlds.launcher.Launcher.launchEnhanced(Launcher.java:289)
at org.codehaus.plexus.classworlds.launcher.Launcher.launch(Launcher.java:229)
at org.codehaus.plexus.classworlds.launcher.Launcher.mainWithExitCode(Launcher.java:415)
at org.codehaus.plexus.classworlds.launcher.Launcher.main(Launcher.java:356)
Caused by: org.apache.maven.plugin.MojoExecutionException: java.lang.NullPointerException
at org.flywaydb.maven.AbstractFlywayMojo.execute(AbstractFlywayMojo.java:526)
at org.apache.maven.plugin.DefaultBuildPluginManager.executeMojo(DefaultBuildPluginManager.java:132)
at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:208)
... 19 more
Caused by: java.lang.NullPointerException
at java.util.Hashtable.put(Hashtable.java:514)
at java.util.Properties.setProperty(Properties.java:161)
at org.flywaydb.maven.MigrateMojo.doExecute(MigrateMojo.java:38)
at org.flywaydb.maven.AbstractFlywayMojo.execute(AbstractFlywayMojo.java:524)
Looked at that line of code and error is happening because of null current version that's being added to Hashtable.
<issue_comment>username_0: Above issue got resolved by not using<issue_closed> |
<issue_start><issue_comment>Title: release naming
username_0: I just noticed that last two releases were names without `v` as a prefix, while every other release had it. Check it out:
https://github.com/laravel/homestead/releases
<issue_comment>username_1: Thanks. Will make sure next release goes back to the v prefix.
--
- Joe Ferguson
JoeFerguson.me
MemphisPHP.org<issue_closed> |
<issue_start><issue_comment>Title: Number of lines in the description label
username_0: how can I change the word wrapping and the numberOfLines?
<issue_comment>username_0: This is in reference to the NSString "purchaseDescription" in STPCheckoutOptions
<issue_comment>username_0: there is no reference to the label, only the string<issue_closed>
<issue_comment>username_1: Yeah, I'm afraid we don't currently expose that API (this isn't actually even backed by a `UILabel`, it's loaded in a webview). |
<issue_start><issue_comment>Title: Fix `scrollTo` function in `<ScrollView/>`
username_0: Fix `scrollTo` function in `<ScrollView/>`, also change scroll view example in UIExplorer because the 'ScrollViewExample' demonstrates `scrollTo`.
<issue_comment>username_1: Thanks! After recent huge changes in ScrollView, it certainly needs refactoring and probably to finally get rid of two parallels implementations too.
<issue_comment>username_0: So the parallels implementations came from RN? Didn't know that.
<issue_comment>username_1: `RCTScrollView` came directly from RN, and RCTNativeScrollView I made a while ago because of some issues in UIKit-based implementation for a desktop. |
<issue_start><issue_comment>Title: How to translate?
username_0: I'd like to submit a pull request translating this to Portuguese and I wasn't able to find out how.
How should I do it?
Thanks
<issue_comment>username_1: Photon supports all kind of languages, have a look here https://github.com/komoot/photon#customized-search-data<issue_closed> |
<issue_start><issue_comment>Title: Allow for inclusion via CommonJS (browserify)
username_0: Would probably have to update all templates to use something like https://github.com/umdjs/umd/blob/master/commonjsStrict.js. If it's worth it, would be willing to make the changes.
<issue_comment>username_1: +1
<issue_comment>username_2: Anyone who works on this should review how Bootstrap did this:
https://github.com/twbs/bootstrap/blob/master/Gruntfile.js#L479
<issue_comment>username_2: Will be delayed until #1168 works itself out.
<issue_comment>username_3: Note: [there are issues using this with webpack](https://github.com/ExactTarget/fuelux/issues/1498).<issue_closed> |
<issue_start><issue_comment>Title: mtools/mlaunch/mlaunch.py: change default of '--dir' to '/data/'
username_0: Change the default value to '/data/', which is consistent with 'mongod'.
Currently, the default value of '--dir' is './data', which IMHO is
inconvenient for two reasons:
1. mlaunch is so convenient, but I hate the fact that I have to cd to
certain directory first each time I want to run mlaunch
start/stop/list etc.
2. There might be 'data' directories scattered all around the file
system since people may run mlaunch in a number of arbitrary
directories. At least this is the case for me ;)
<issue_comment>username_0: Hi Thomas,
I am sending this pull request just to let you know my two cents. Please feel free to reject it if you don't agree.
Btw. thank you for this fancy tool.
Regards,
Lungang
<issue_comment>username_1: Discussed with Lungang and mentioned three issues when using a global `/data` directory:
1. Permission issues -- using `/data` will require root permission, which is not desired.
2. Cleanup issues -- running mlaunch with a different setting will require manual cleanup of the `/data` directory, which could be error-prone.
3. Surprising behaviour -- especially if mlaunch is used for standing up a testing environment where an existing `mongod` service is already running in the system.
We can use an alias if using a global dbpath is required for certain use cases.
<issue_comment>username_2: I think the 3rd thing mentioned by @username_1 is the main thing. I often have several mlaunch environments going at the same time (on different ports), and I find it convenient to organise them in different directories. If --datadir defaulted to an absolute path, then I would have to specify `--datadir ...` for every `mlaunch` command, which is much less convenient than changing into a dir and then just doing normal `mlaunch` commands.
I do wish we could agree on `mlaunch nuke` (#362) though clearly I can live with just typing `mlaunch stop && rm -rf data` when I'm done with an environment. :) |
<issue_start><issue_comment>Title: 'no-vars-unused' false flag when variable only used in template string
username_0: <!--
This template is for bug reports. If you are reporting a bug, please continue on. If you are here for another reason, please see below:
1. To propose a new rule: http://eslint.org/docs/developer-guide/contributing/new-rules
2. To request a change: http://eslint.org/docs/developer-guide/contributing/changes
3. If you have any questions, please stop by our chatroom: https://gitter.im/eslint/eslint
Note that leaving sections blank will make it difficult for us to troubleshoot and we may have to close the issue.
-->
**What version of ESLint are you using?** 2.2.0
**What configuration and parser (Espree, Babel-ESLint, etc.) are you using?** Defaults
**What did you do? Please include the actual source code causing the issue.**
The code in question:
function sendSteamRepReport(chatId, userId) {
let options = {
'url' : 'http://steamrep.com/api/beta/reputation/${userId}?json=1',
'json' : true
};
request.get(options, (err, response, body) => {
if (err) {
console.error(err.toString());
return;
}
if (body && body.steamrep) {
let steamrep = body.steamrep;
let reputation = steamrep.reputation;
if (steamrep.flags.status === 'exists' && reputation) {
let name = getDisplayName(userId);
let msg = 'SteamRep report for ${name}: ${reputation}';
steamFriends.sendMessage(chatId, msg);
}
}
});
}
`name` is declared the line prior to and used only in the line `let msg = 'SteamRep report for ${name}: ${reputation}';`.
**What did you expect to happen?**
No errors.
**What actually happened? Please include the actual, raw output from ESLint.**
`error 'name' is defined but never used no-unused-vars`
<issue_comment>username_1: Well, looking closely `let msg = 'SteamRep report for ${name}: ${reputation}';` is not template literals :)
<issue_comment>username_1: Also I verified that the following works on master:
```js
let who = "Paul";
let text = `Hello ${who}!`;
console.log(text);
```<issue_closed>
<issue_comment>username_2: @username_0 template literals will only work if you use \` quotes around them. Otherwise those are just strings. |
<issue_start><issue_comment>Title: bad removed comment
username_0: Bad removed all comments in file:
[FieldQuery.zip](https://github.com/username_1/CommentRemover/files/461516/FieldQuery.zip)
I used version 1.3
<issue_comment>username_1: Repro steps in C#:
```c#
/**
*
* <param name="fieldName"></param>
* <param name="term"></param>
* <returns>QueryPhraseMap</returns>
*/
public QueryPhraseMap GetFieldTermMap(String fieldName, String term)
{
}
``` |
<issue_start><issue_comment>Title: llvm-general: Adds Darwin compatibilty
username_0: Fix for #14035
<issue_comment>username_1: This looks good to me superficially. I can give it a try tomorrow. Thanks for this, @username_0!
<issue_comment>username_1: Looks good, builds, and tests pass for me!
<issue_comment>username_2: I would very much prefer to have those fixes applied to the package *upstream*. Would you mind submitting a PR to the libraries' authors instead, so that all users can benefit from these fixes?
<issue_comment>username_0: Upstream is working fine for me outside of Nix on OS X. As far as I can tell / unless I've gotten something horribly confused, these changes are required to make llvm-general work on Nix.
<issue_comment>username_1: An option to use `libc++` in the `cabal` file would make sense, but we’d still need to enable it on Darwin in `nixpkgs`.
<issue_comment>username_0: Is there anything I need to do to get this in?
<issue_comment>username_2: Does this problem still exist?
<issue_comment>username_0: My understanding is that there haven't been any changes around this in either llvm-general or in nixpkgs, so I'm pretty sure that it does.
I've since moved from using Nix on OS X to using NixOS in a VM, so I don't have a way of checking it anymore. |
<issue_start><issue_comment>Title: Deployment options for Azure Database for PostgreSQL
username_0: Hi, I just read the article on the three deployment levels or options for Azure Database for PostgreSQL and I am totally shocked, because for the last 4 months, I studied and believed that there were only two. On the Microsoft-learning webpage for preparing to take the DP-900 exam:
'https://docs.microsoft.com/en-us/learn/modules/explore-provision-deploy-relational-database-offerings-azure/3-azure-database-open-source': It specifically states that "Azure Database for PostgreSQL has two deployment options: Single-server and Hyperscale."
So when I took a mock exam to prepare for the real one next week, that question came up. I answered that there were two deployment levels, but I got it wrong. The exam responsible, named Whizlabs, had a link from Microsoft overview: 'https://docs.microsoft.com/en-us/azure/postgresql/overview', where it states that there were three. So if Microsoft Docs is different to Microsoft learning, what are we supposed to do?, Which one is correct?
Thank you for your speedy answer.
Segb.
---
#### Document Details
⚠ *Do not edit this section. It is required for docs.microsoft.com ➟ GitHub issue linking.*
* ID: 8e30e735-75da-4043-c515-909f97a1bbba
* Version Independent ID: 471da130-86c5-aa29-f212-dd700edec9fb
* Content: [What is Azure Database for PostgreSQL](https://docs.microsoft.com/en-us/azure/postgresql/overview)
* Content Source: [articles/postgresql/overview.md](https://github.com/MicrosoftDocs/azure-docs/blob/main/articles/postgresql/overview.md)
* Service: **postgresql**
* Sub-service: **single-server**
* GitHub Login: @sunilagarwal
* Microsoft Alias: **sunila**
<issue_comment>username_1: @username_0 Thank you for the feedback. We are actively investigating and will get back to you soon.
<issue_comment>username_1: @username_0 Thanks for your patience. You should always consider below link for latest updates:
[What is Azure Database for PostgreSQL?](https://docs.microsoft.com/en-us/azure/postgresql/overview)
It seems Microsoft learn content is not up to date. You can submit the request for the same in below link:
[Report Feedback](https://docs.microsoft.com/en-us/learn/support/troubleshooting#report-feedback)
<issue_comment>username_0: You are receiving this because you were mentioned.Message ID: ***@***.******@***.***>>
<issue_comment>username_1: Apologies for the trouble you faced because of inconsistency in the documents. Learn document is more of detailed step by step instructions build on basis of other documents and sometimes the latest changes take a little while to get updated on them. However I will pass this feedback to the learn team. Again sorry for the double effort you need to spend and all the best for you exams.<issue_closed>
<issue_comment>username_0: You are receiving this because you were mentioned.Message ID: ***@***.******@***.***>> |
<issue_start><issue_comment>Title: [wip] Performance improvements
username_0: | Q | A
| ------------- | ---
| Bug fix? | no
| New feature? | no
| BC breaks? | yes
| Deprecations? | no
| Tests pass? | Not yet (still experimenting)
| Related tickets | #212
| License | MIT
| Doc PR | ø
## Add `getIriFromItem` cache
The collection here has ~8k items, no relations. Dev mode active but caching with apcu is forced.
[Blackfire before](https://blackfire.io/profiles/325404d0-c782-4f05-88ed-4676b6c3368b/graph):
- 2.87s spent in `getIriFromItem`
- 2.35 s for `getIdentifiersFromItem`
[Blackfire after](https://blackfire.io/profiles/7215dbe1-efb1-4835-aa8f-5296761c280a/graph):
- 898 ms spent in `getIriFromItem`
- 371 ms for `getIdentifiersFromItem`
When discussing my performance issue with @username_3 we noticed that `getIdentifiersFromItem` was called a lot (previous method is `getIriFromItem`). This is called for every item in the collection.
### Profiler (not really representative, but with this fix I basically win 1 precious second for 8k elements):
#### before
#### after
In this PR, we cache the **properties** that are **identifiers** for the given class.
<issue_comment>username_1: I like the idea, but the implementation is wrong. We should decorate the IriConverter.
<issue_comment>username_0: 👍 indeed
<issue_comment>username_0: Okay so I did this with a decorator, but because what's slow is `getIdentifiersFromItem` (which is a private method) I'm not sure it's ideal. Indeed, I had to copy the code from `IriConverter` (trying to avoid BC break at the same time), therefore I see no point in decorating :/.
You can check the code to understand what I mean...
A solution would be to have a new service for `getIdentifiersFromItem` that we can cache without changing the IriConverter. Lmk if you've a better idea.
-----
From my benchmarks, I know that we can still win a lot by caching the identifiers getters instead of calling the `PropertyAccessor` (`$identifiers[$propertyName] = $item->{$accessor}();`.
Some links (note that I'm using the normalizer dumper, 8000 items collection):
- [IriConverter without cache](https://blackfire.io/profiles/beab0865-7e50-40ed-823c-bc743e69edb3/graph) 8.36s
- [IriConverter with this PR cache](https://blackfire.io/profiles/51b859bd-956e-4f35-b5c5-79d3968717b7/graph) 6.89s
- [IriConverter with direct getter instead of PropertyAccessor + cache](https://blackfire.io/profiles/e9fda268-d964-4692-85cf-c85ed32391c9/graph) 4.78s
<issue_comment>username_1: `ItemIdentifiersResolverInterface`?
<issue_comment>username_0: `ItemIdentifiersResolverInterface`? Extract is a bit strong no?
Would `get${ucfirst($propertyName)}` be acceptable to retrieve identifiers instead of using the property accessor?
<issue_comment>username_1: No, we cannot make such assumptions. Plus, it'd be a BC break.
<issue_comment>username_2: What we can do however is provide implementation of the `PropertyAccessorInterface` doing that. I usually do this for perf in my project. Such class should be included directly in Symfony.
<issue_comment>username_0: @username_2 You mean having a kind of IdentifierPropertyAccessor witch would end up doing something simpler than the original class (still no assumption could be made regarding `get{$attribute}` ? We also don't need `setValue` in this context.
<issue_comment>username_2: @username_0 yes, that you can register for your project if you stick to a particular convention (like public properties, or getters and setters). It prevents to trigger the introspection system, so it improves performances.
<issue_comment>username_0: Oh I'm doing this already on my project for this particular class :). So, I'll just add a `ItemIdentifiersExtractorInterface` so that identifiers can be cached here but I won't touch the propertyAccessor thing.
<issue_comment>username_0: ping @username_1 @username_2
I tried to avoid breaking changes lmk what you think.
Note that Behat tests are green, just waiting for implementation approval to add tests (and #1014 comes first obviously).
<issue_comment>username_3: 👍 Great work @username_0
<issue_comment>username_0: I did a great job covering this part as it wasn't tested anywhere (except through IriConverter, but not fully covered).
There's still a point to discuss: the location of the `IdentifiersExtractor`.
I'm not sure that it should actually be under `ApiPlatform\Core\Metadata` because it doesn't interfere with metadata, rather uses the metadata to extract identifiers. Therefore, IMHO this class should be in `ApiPlatform\Core\Api` (so is `ResourceClassResolver` for example). Let me know if you disagree we can always move it elsewhere :).
<issue_comment>username_0: All green ping @api-platform/core-team
<issue_comment>username_3: @username_0 could you please update with master ?
<issue_comment>username_0: @username_2 this should be good now! |
<issue_start><issue_comment>Title: matplotlib 2.0.0 dependency
username_0: I currently have matplotlib 1.5.x and ete3 and other software depends on it, thus I cannot have color bars for samples. It throws this error msg: AttributeError: 'LinearSegmentedColormap' object has no attribute 'colors'
cae.plot_sort('smj_genusspecies', gui='jupyter', databases=[], cmap='jet', sample_color_bars=['Q2']) |
<issue_start><issue_comment>Title: Kicked from server when mounting wild horse
username_0: Client version: 1.9.4
Commit id: (Found at the top of the server's console output)
mount wild horse
kicked from server
Join new survivor wold and mount a wild horse
<issue_comment>username_1: Horses are completely broken. I think they're a good candidate for being disabled by default until they are fixed.
<issue_comment>username_2: Since [the issue reporting broken behaviour when riding a horse](https://github.com/cuberite/cuberite/issues/703) is more than 2 years old, I indeed also think they should be disabled by default till they are fixed. This at least removes the hopes of people trying to ride them.
<issue_comment>username_2: I can report that this doesn't happen (anymore), running a 1.10 client on the latest git build. Horses are still really broken though.
<issue_comment>username_3: I can't reproduce this issue with 1.9.4, maybe it has been fixed?
<issue_comment>username_4: Closing as stale, and non-reproducible.<issue_closed> |
<issue_start><issue_comment>Title: Develop a bundle-splitting engineering plan
username_0: Develop a detailed engineering plan for bundle-splitting.
Use the bundle-splitting scenarios from #1282 as a starting point, along with code from internal Office repos. Cover both bundle production and bundle consumption. Include expectations regarding bundle size increase compared to equivalent single-bundle runs.
Output is a plan and a task list.
<issue_comment>username_0: ### Notes on one possible solution
A split bundle exports a collection of packages, each under a unique name. The split bundle author specifies this mapping as kit configuration.
For example, it might look like this:
```json
"rnx-kit": {
"splitBundle": {
"interface": {
"MyKitGlobal.Telemetry": "telemetryjs",
"MyKitGlobal.StringResources": "string-resources",
"MyKitGlobal.ReactNative.Components": "react-native"
}
}
}
```
The bundler will take this configuration, and use it to generate a programmatically-consumable, type-safe API. Something along the lines of a typescript .d.ts file.
Other packages which depend on this kit will naturally gain access to the type-safe API, and benefit from it in IDEs like VSCode and at lint/build time. |
<issue_start><issue_comment>Title: Result promise of RequestInstantiateAll will be fulfilled with the re…
username_0: Here, we assume that the result promise of the RequestInstantiateAll will be fulfilled
with the module record instance. But actually it will be fulfilled with the registry entry.
Since the result of the RequestInstantiate is also fulfilled with the registry entry,
we align the RequestInstantiateAll to the RequestInstantiate.
<issue_comment>username_0: Or is it better fulfilling `[[instantiate]]` and the result of `RequestInstantiateAll` with the module `instance` value?
<issue_comment>username_0: In the meantime, since `RequestLink` also assumes that the result of `RequestInstantiateAll` will be fulfilled with the `entry`, I aligned to that.
<issue_comment>username_1: In general, we have a problem with fulfilling with registry entry, `[[Module]]` or `[[Namespace]]`, we need to revisit all of them to make sure we are doing the right thing. I will be taking care of it on PR #65.
<issue_comment>username_1: @username_0 html version was updated to reflect this fix.
<issue_comment>username_0: Thank you! |
<issue_start><issue_comment>Title: Create script to readily see oustanding project lead requests
username_0: We need a script to readily see outstanding requests to be a project
lead (I lost track) - you can't rely on emails.
_Imported from trac ticket [#404](http://trac.gpolab.bbn.com/proto-ch/ticket/404), created by sedwards on 02-21-2013 at 15:49, last modified: 03-30-2015 at 16:51_
<issue_comment>username_1: Note this depends on storing these requests in a DB table.
_Trac comment by username_1 on 07-30-2013 at 16:48_
<issue_comment>username_1: Charlie's admin page will cover this
<issue_comment>username_1: Charlie's admin page is in.<issue_closed> |
<issue_start><issue_comment>Title: ERROR
username_0: ## Known problems for 1.3.183:
1. `System.NullReferenceException`: https://github.com/Wox-launcher/Wox/releases/tag/v1.3.357
2. `System.UriFormatException`: delete your old theme file
3. `System.Threading.Tasks.TaskCanceledException`: https://ci.appveyor.com/project/username_1/wox/build/1.3.423/artifacts
## English
Try search existing issues before you open new one
1. Wox version / commit number:
2. Steps to **reproduce** the bug:
3. Windows version:
4. Exception:
5. **whole** log file （`%APPDATA%\Wox\Logs\version\date.txt`):
6. Screenshot (optional):
## 中文 / Chinese
提问之前搜索一下已经有的 issue
1. Wox 版本 / commit number：
2. 如何**重现** bug：
3. Windows 版本：
4. Exception：
5. **整个** log 文件 （`%APPDATA%\Wox\Logs\version\date.txt`）：
6. 截图 （可选）：
<issue_comment>username_1: ?????<issue_closed> |
<issue_start><issue_comment>Title: Mschultz4 patch 2
username_0: Fixed several broken links in table of contents
<issue_comment>username_1: LGTM except _Positional parameters_.
<issue_comment>username_0: I changed the section header and contents from argument to parameter, but the entire section uses the term argument. Should I change all argument references to parameter?
<issue_comment>username_1: Yes, please.
<issue_comment>username_2: Confirmed that this PR fixes all broken links in the Table of Contents.
<issue_comment>username_0: All set. Please review. |
<issue_start><issue_comment>Title: [Validator] Dutch translation fixes
username_0: | Q | A
| ------------- | ---
| Branch? | "master"
| Bug fix? | yes
| New feature? | no
| BC breaks? | no
| Deprecations? | no
| Tests pass? | yes
| Fixed tickets | ~
| License | MIT
| Doc PR | ~
Some dutch translation fixes. I probably can rebase this against 2.*, but maybe someone wants to comment first or spots any other spelling errors.
Ignore fabbot.io :)
<issue_comment>username_1: @WouterJ @Iltar Would you mind to have a look at this?
<issue_comment>username_2: Sorry to post this in Dutch, but else it becomes an unreadable mess :)
* `Moet` en `moeten` klinkt erg dwingend, `mag niet` klinkt wat vriendelijker.
* `kon niet gevonden worden` is meer gebruikelijk dan `kon niet worden gevonden.` Maar `Het bestand kon niet worden geüpload.` is dan weer duidelijker dan de vorige versie 👍
Vertalingen hoeven niet letterlijk te worden overgenomen, zolang als de oorspronkelijke bedoeling duidelijk is. Maar dat is slechts mijn mening, dus ik ben benieuwd wat andere vinden 😄
<issue_comment>username_3: I agree with @username_2's suggestions. Other than that, looks good to me!
<issue_comment>username_0: This is about `This value is not valid.` meaning the same as `This value is invalid`. Which both can separately translated in dutch. Sstok's point is it basically doesnt matter to much how you translate this (it means the same, ie `This value is invalid` could be translated as `... is ongeldig` (literal) or `... is not valid` (less literal, but the same thing). Currently i "corrected" this for `This value should be false.`, however i tend to agree with username_2 (mostly pragmaticly) but still i think we should do a "correct" translation when possible (ie not changing syntax/words to much).
<issue_comment>username_0: Ok sorry for the maybe incoherent story above. Rewrote it a bit earlier to clarify.
Anyway, for now translations with `should be`. `should not be` and `could not be` are at least consistent.
- `This value should be false.` was the only exception in dutch using different syntax for `This should be X`. Imho it still reads as well/friendly.
- `could not be` is the problem maybe, with my last commit at least the syntax is consistent. Imho all read as well, however @username_2 got me confused as he states it depends on the past participle to determine the most common syntax. Dutch will understand both that's not the problem..
<issue_comment>username_0: I revered 2 changes for `could not be`. I understand what @username_2 means. In dutch one syntax sounds better/more fluent then the other depending on the verb (said it a few times ;-)). However i have no clear ref on this, but both are valid spelling.
Currently some translations still switch from past to present tense, i guess this is also somewhat context related eventually.
Long story short; this is good to go.
<issue_comment>username_4: Thank you @username_0. |
<issue_start><issue_comment>Title: invalid csrf token error
username_0: I'm getting this `invalid csrf token` error when I am trying to log in the user to the site.
```javascript
{ ForbiddenError: invalid csrf token
at csrf (/Users/Documents/web-new/node_modules/csurf/index.js:113:19)
```
The login with storing session in redis works without the csurf module. With the csurf module, the session ID is getting stored in redis but I am not able to return the proper response to the client to log in the user. I am using Angular2 with node/express. From what I understand, Angular2 by default supports CSRF/XSRF with the CookieXSRFStrategy when using HTTP service, so all I need to do is configure something on the node/express side.
I'm also attaching the screenshot of XSRF-TOKEN that I am seeing in the application tab of devtools.
Could you kindly recommend how I might fix this error?
My code is here below.
```javascript
import path from 'path';
import session from 'express-session';
import connectRedis from 'connect-redis';
import rp from 'request-promise';
import * as _ from 'lodash';
import cors from 'cors';
import csurf from 'csurf';
const redisStore = connectRedis(session);
const dbStore = new redisStore(db);
/* ~~ configure session ~~ */
app.use(session({
secret: app.getValue('env').SESSION_SECRET,
store: dbStore,
saveUninitialized: false,
resave: false,
rolling: true,
cookie: {
maxAge: 1000 * 60 * 30 // in milliseconds; 30 min
}
}));
/* ~~ login user ~~ */
let csrf = csurf();
app.post('/loginUser', csrf, (req, res, next) => {
let user = {};
let loginOptions = {
method: 'POST',
url: `${baseUrl}/client/login`,
rejectUnauthorized: false,
qs: {
token: app.getValue('token'),
email: req.body.email,
password: req.body.password
},
headers: {
'Content-Type': 'application/json'
},
json: true
};
rp(loginOptions)
.then(response => {
let userToken = response.response.token;
let clientId = response.response.clientId;
req.session.key = req.session.id;
user.userToken = userToken;
user.clientId = clientId;
let clientAttributeOptions = {
url: `${baseUrl}/client/${clientId}/namevalue`,
rejectUnauthorized: false,
qs: {
token: app.getValue('token'),
usertoken: userToken
},
json: true
};
return rp(clientAttributeOptions);
})
.then(response => {
req.session.user = user;
res.send({user:user})
})
.catch(err => {
next(err);
})
});
```<issue_closed>
<issue_comment>username_0: My issue was that I was including the csrf function as a middleware only in the `app.post('/loginUser) ` route.
When I included it for all routes, the module worked fine.
```let csrf = csurf();
app.get('/*', csrf, (req, res) => {
res.sendFile(app.get('indexHTMLPath'));
});``` |
<issue_start><issue_comment>Title: Cannot read property 'concat' of undefined in TResetAllButton.js
username_0: Simulations cannot run :(
TResetAllButton.js?bust=1493732212455:39 Uncaught TypeError: Cannot read property 'concat' of undefined
at TResetAllButton.js?bust=1493732212455:39
at Object.execCb (require-2.1.11.js:1650)
at Module.check (require-2.1.11.js:866)
at Module.<anonymous> (require-2.1.11.js:1113)
at require-2.1.11.js:132
at require-2.1.11.js:1156
at each (require-2.1.11.js:57)
at Module.emit (require-2.1.11.js:1155)
at Module.check (require-2.1.11.js:917)
at Module.enable (require-2.1.11.js:1143)
Assigning to @username_1 since he is the last one who made a commit to TResetAllButton<issue_closed> |
<issue_start><issue_comment>Title: Crash on Acer Chromebook R11. Said to report it.
username_0: WARNING! An unknown error has occurred, starting debug...
Starting debug... (Please report this crash on 'https://github.com/username_1/kickthemout/issues' with your private informations removed if necessary)
Current defaultGatewayMac: 7C:0E:CE:E9:4F:A7
Reloading mac getter function...
Reloaded defaultGatewayMac: 7C:0E:CE:E9:4F:A7
Known gateway IP: 10.6.80.1
Current hostslist array:
[['10.6.82.40', '1C:CB:99:29:42:11'], ['10.6.91.14', 'B8:44:D9:27:F1:5B'], ['10.6.84.6', '48:3B:38:10:50:E4'], ['10.6.91.13', 'E0:C7:67:BC:AF:E7'], ['10.6.82.3', '1C:91:48:78:31:4F'], ['10.6.90.97', '1C:5C:F2:92:BA:C2'], ['10.6.90.96', '48:DB:50:02:C4:F2'], ['10.6.86.78', '1C:1A:C0:70:87:F8'], ['10.6.93.96', 'B4:4B:D2:F1:7D:2A'], ['10.6.90.92', '1C:1A:C0:74:64:0F'], ['10.6.90.164', '78:F8:82:A1:04:C8'], ['10.6.82.105', 'F4:31:C3:7C:83:21'], ['10.6.88.124', '28:A0:2B:CE:DF:F6'], ['10.6.88.127', 'E0:B9:BA:88:4F:FE'], ['10.6.81.18', '54:35:30:DD:F2:DD'], ['10.6.81.110', 'AC:5F:3E:2D:6F:02'], ['10.6.94.132', '28:C2:DD:38:F9:2B'], ['10.6.82.35', 'C8:1E:E7:87:93:70'], ['10.6.92.166', '40:E2:30:61:D9:A1'], ['10.6.80.72', '1C:1A:C0:75:38:61'], ['10.6.92.32', '54:35:30:DD:F8:4F'], ['10.6.80.76', 'E4:9A:79:E1:07:1D'], ['10.6.92.30', 'E0:DB:10:28:97:94'], ['10.6.88.221', 'AC:37:43:A5:92:F2'], ['10.6.94.151', '8C:8E:F2:5C:44:26'], ['10.6.91.144', '28:F0:76:23:88:66'], ['10.6.84.76', '4C:74:BF:34:2D:66'], ['10.6.84.77', '60:D9:C7:8B:67:C1'], ['10.6.84.70', '40:E2:30:61:FB:5D'], ['10.6.89.23', 'C0:1A:DA:D5:13:C4'], ['10.6.89.25', '84:38:35:96:D6:D5'], ['10.6.89.29', 'AC:5F:3E:A3:43:76'], ['10.6.93.255', '28:C2:DD:46:28:5D'], ['10.6.81.188', '48:D7:05:55:82:0E'], ['10.6.81.186', '08:74:02:10:51:FF'], ['10.6.81.181', '1C:1A:C0:75:05:EB'], ['10.6.83.115', 'F4:F5:24:E2:A6:4C'], ['10.6.90.158', 'D0:33:11:0F:78:46'], ['10.6.80.143', '1C:1A:C0:71:0B:20'], ['10.6.83.112', 'B8:1D:AA:AC:6B:C1'], ['10.6.94.230', '54:35:30:DD:85:A9'], ['10.6.92.246', '2C:B4:3A:26:BE:8B'], ['10.6.80.148', '28:C2:DD:38:EC:D3'], ['10.6.85.95', 'FC:15:B4:AB:5A:B7'], ['10.6.89.176', 'E0:B9:BA:08:5C:B0'], ['10.6.82.251', '1C:1A:C0:6E:CE:66'], ['10.6.84.134', '40:E2:30:61:FC:D3'], ['10.6.84.131', '1C:91:48:BE:AE:D0'], ['10.6.94.67', '1C:5C:F2:A9:39:E2'], ['10.6.92.239', '6C:72:E7:9C:0B:40'], ['10.6.93.100', '90:2E:1C:52:9B:BC'], ['10.6.89.182', 'CC:3A:61:98:F6:7C'], ['10.6.89.183', '28:18:78:31:CD:CD'], ['10.6.93.132', '28:C2:DD:38:EC:C3'], ['10.6.95.14', 'F4:0E:22:30:50:18'], ['10.6.92.237', '28:C2:DD:16:DA:4D'], ['10.6.88.89', '74:E1:B6:5B:0D:2A'], ['10.6.88.88', 'E0:B9:BA:08:5A:61'], ['10.6.80.209', '20:7D:74:29:76:C2'], ['10.6.88.81', '74:E1:B6:57:D6:FE'], ['10.6.88.83', 'E0:B9:BA:80:FD:CA'], ['10.6.88.82', 'E0:B9:BA:17:32:C7'], ['10.6.88.87', 'D0:03:4B:05:16:48'], ['10.6.88.86', 'E0:B9:BA:88:1F:E0'], ['10.6.95.187', '30:63:6B:BB:88:B6'], ['10.6.95.186', '28:B2:BD:89:10:6B'], ['10.6.82.155', '74:E1:B6:56:9D:08'], ['10.6.89.160', '78:FD:94:25:6E:42'], ['10.6.95.188', '40:E2:30:61:FB:63'], ['10.6.91.56', '90:72:40:E8:54:D8'], ['10.6.89.167', '7C:01:91:05:1B:F6'], ['10.6.92.7', '28:F0:76:10:0D:CC'], ['10.6.92.155', '40:E2:30:62:00:51'], ['10.6.93.92', 'BC:9F:EF:88:C8:ED'], ['10.6.92.151', '40:E2:30:61:FF:C3'], ['10.6.92.150', '80:A5:89:0E:1A:3F'], ['10.6.92.152', '28:F0:76:24:A4:BC'], ['10.6.93.176', '40:E2:30:61:FF:CD'], ['10.6.84.149', '74:E1:B6:50:C0:58'], ['10.6.92.159', '40:E2:30:61:C2:BB'], ['10.6.92.0', '54:35:30:DD:EC:33'], ['10.6.86.32', '74:E1:B6:5B:1E:06'], ['10.6.93.94', '28:F0:76:0F:AA:02'], ['10.6.90.36', '5C:97:F3:34:06:72'], ['10.6.95.114', '40:E2:30:61:C2:C9'], ['10.6.83.235', '54:35:30:DD:80:FF'], ['10.6.83.233', '1C:CB:99:C0:52:E1'], ['10.6.90.159', 'AC:BC:32:23:CB:D6'], ['10.6.87.90', 'E0:B9:BA:81:0F:7F'], ['10.6.90.153', '1C:1A:C0:6E:DE:87'], ['10.6.93.101', '3C:AB:8E:72:0C:2F'], ['10.6.82.252', '6C:72:E7:89:95:DC'], ['10.6.90.154', '28:F0:76:1C:2F:74'], ['10.6.81.69', 'FC:C2:DE:15:DC:DF'], ['10.6.81.17', '1C:1A:C0:79:9E:8D'], ['10.6.94.253', 'E0:9D:31:0D:53:0C'], ['10.6.83.51', '90:2E:1C:51:B5:85'], ['10.6.83.50', '1C:1A:C0:71:A3:A2'], ['10.6.81.140', 'E8:50:8B:5B:F0:51'], ['10.6.81.144', '1C:91:48:15:6C:E1'], ['10.6.80.183', '9C:D9:17:41:D1:8E'], ['10.6.84.88', '60:D9:C7:89:7D:AE'], ['10.6.84.211', '40:E2:30:61:FA:A7'], ['10.6.84.85', '68:94:23:C9:27:1F'], ['10.6.84.87', '60:D9:C7:8C:AD:C1'], ['10.6.80.188', '40:E2:30:61:D9:9F'], ['10.6.84.80', '60:D9:C7:8B:43:27'], ['10.6.88.162', '74:E1:B6:55:22:67'], ['10.6.88.165', '1C:1A:C0:76:0E:7B'], ['10.6.83.245', '90:2E:1C:51:F0:45'], ['10.6.88.169', '74:E1:B6:58:92:9F'], ['10.6.95.205', '90:2E:1C:52:64:7B'], ['10.6.91.219', '00:56:CD:30:C3:01'], ['10.6.91.218', '90:48:9A:4F:32:87'], ['10.6.91.215', '54:35:30:DD:91:1B'], ['10.6.91.216', '90:48:9A:4F:62:B5'], ['10.6.92.47', '50:9E:A7:99:25:97'], ['10.6.89.79', 'A0:99:9B:1D:3A:9B'], ['10.6.89.78', '4C:74:BF:98:7B:51'], ['10.6.88.187', '8C:70:5A:C7:E7:78'], ['10.6.89.75', 'A0:D3:C1:DF:E8:44'], ['10.6.80.186', '00:34:DA:7E:F0:9A'], ['10.6.93.201', '48:3B:38:18:21:23'], ['10.6.93.205', 'F8:27:93:58:D5:0C'], ['10.6.93.206', '90:48:9A:4F:6C:8B'], ['10.6.81.24', '28:C2:DD:33:5E:D7'], ['10.6.94.242', '10:41:7F:26:30:67'], ['10.6.84.38', '54:35:30:DD:BC:B9'], ['10.6.84.39', '54:35:30:DD:BD:35'], ['10.6.93.104', '28:F0:76:24:A2:E4'], ['10.6.80.110', '10:41:7F:CB:A1:9F'], ['10.6.83.141', '54:35:30:DD:B2:A9'], ['10.6.83.142', '54:35:30:DD:F0:81'], ['10.6.91.105', '78:4B:87:E7:0E:07'], ['10.6.91.104', '28:F0:76:10:0D:B0'], ['10.6.80.116', 'F4:31:C3:7F:40:8A'], ['10.6.89.124', '2C:F0:A2:9F:40:FD'], ['10.6.91.95', '6C:40:08:22:F1:A4'], ['10.6.89.126', '60:D9:C7:8B:25:5F'], ['10.6.91.91', '84:38:35:8F:B7:4C'], ['10.6.87.219', 'E0:B9:BA:81:1F:C6'], ['10.6.92.112', '28:F0:76:24:B8:3A'], ['10.6.90.233', '1C:1A:C0:70:D7:96'], ['10.6.84.104', '4C:57:CA:6C:FE:53'], ['10.6.94.37', '54:35:30:DD:B5:B5'], ['10.6.95.42', '90:48:9A:4F:35:BD'], ['10.6.89.232', '14:BD:61:D4:55:83'], ['10.6.89.83', '4C:57:CA:AA:92:26'], ['10.6.89.80', '60:E3:AC:E1:A8:A3'], ['10.6.89.81', 'B4:4B:D2:8E:81:F9'], ['10.6.92.79', '1C:1A:C0:6E:C9:7D'], ['10.6.89.89', '98:B8:E3:B5:AE:65'], ['10.6.82.168', 'F0:24:75:B0:6E:EE'], ['10.6.93.37', '40:4D:7F:95:4B:04'], ['10.6.93.31', '8C:8E:F2:51:7E:88'], ['10.6.82.165', '90:2E:1C:51:D2:77'], ['10.6.81.249', '24:0D:C2:12:BA:1E'], ['10.6.81.241', '28:A0:2B:43:63:4F'], ['10.6.81.247', 'A8:96:8A:93:A3:73'], ['10.6.80.234', 'E8:B1:FC:C0:9B:CB'], ['10.6.92.186', '40:E2:30:61:FB:61'], ['10.6.92.187', '40:E2:30:61:FA:F1'], ['10.6.92.184', '1C:1A:C0:73:1A:21'], ['10.6.95.196', '90:2E:1C:51:B5:4E'], ['10.6.80.239', '84:A1:34:9A:EF:A0'], ['10.6.80.238', '60:D9:C7:8B:AB:20'], ['10.6.92.181', '40:E2:30:61:A2:5F'], ['10.6.93.81', 'E0:B9:BA:17:1E:A8'], ['10.6.93.80', 'CC:29:F5:4D:2E:7C'], ['10.6.90.115', '18:AF:61:66:7F:BA'], ['10.6.93.82', '2C:F0:A2:47:F2:F6'], ['10.6.90.112', 'F0:24:75:BB:BA:27'], ['10.6.82.118', '28:C2:DD:33:7C:A9'], ['10.6.93.86', '28:C2:DD:38:F8:67'], ['10.6.86.64', '40:E2:30:61:FD:29'], ['10.6.93.149', '28:C2:DD:38:F7:9B'], ['10.6.82.229', 'E0:5F:45:93:9B:C1'], ['10.6.92.53', '68:94:23:C9:40:F2'], ['10.6.83.14', '40:E2:30:61:FA:F7'], ['10.6.81.104', '24:E3:14:6D:D3:FF'], ['10.6.90.49', '28:C2:DD:38:F7:F9'], ['10.6.90.45', 'E0:C7:67:B5:8A:25'], ['10.6.90.46', '8C:8E:F2:62:FE:3D'], ['10.6.90.47', 'AC:7F:3E:56:38:46'], ['10.6.90.41', '9C:F4:8E:0B:99:5B'], ['10.6.95.203', '40:E2:30:61:FB:B7'], ['10.6.81.99', 'D0:22:BE:F8:D7:F4'], ['10.6.83.207', '28:C2:DD:38:EC:ED'], ['10.6.83.206', '28:C2:DD:33:76:8B'], ['10.6.83.201', 'FC:C2:DE:70:F4:54'], ['10.6.83.200', 'A4:4B:15:A2:2C:42'], ['10.6.80.66', '2C:B4:3A:23:77:48'], ['10.6.89.90', '70:70:0D:D7:98:60'], ['10.6.91.152', '20:AB:37:25:C0:77'], ['10.6.83.221', '1C:1A:C0:73:BA:95'], ['10.6.91.150', '90:48:9A:4F:79:CD'], ['10.6.94.103', '90:48:9A:4F:4F:E5'], ['10.6.91.157', '28:F0:76:0F:A6:DE'], ['10.6.88.251', 'D0:25:98:52:FE:9D'], ['10.6.84.69', '48:4B:AA:63:A1:C9'], ['10.6.83.67', '1C:1A:C0:75:D3:A1'], ['10.6.83.65', 'F0:25:B7:AD:94:50'], ['10.6.81.171', 'A4:31:35:86:DE:A8'], ['10.6.81.176', '68:C4:4D:A8:45:88'], ['10.6.80.155', '28:C2:DD:5D:30:13'], ['10.6.83.107', '2C:0E:3D:E8:BF:7C'], ['10.6.80.150', '04:4B:ED:45:9B:85'], ['10.6.81.20', '54:35:30:DD:F0:59'], ['10.6.88.130', 'E0:B9:BA:08:64:27'], ['10.6.88.136', '1C:1A:C0:7A:1C:2A'], ['10.6.81.25', '28:C2:DD:33:9A:51'], ['10.6.81.27', '28:C2:DD:38:EC:87'], ['10.6.89.4', '20:AB:37:3A:9E:32'], ['10.6.89.1', '68:96:7B:55:51:BE'], ['10.6.89.2', 'E0:B9:BA:17:1E:9F'], ['10.6.95.237', '90:2E:1C:51:E9:F1'], ['10.6.95.236', '60:D9:C7:8D:84:03'], ['10.6.81.8', '4C:66:41:6C:8B:91'], ['10.6.94.75', '40:E2:30:61:C0:C1'], ['10.6.90.234', '5C:AF:06:84:A9:82'], ['10.6.89.41', 'A8:96:8A:8B:8D:A9'], ['10.6.89.47', 'A4:B8:05:8B:D1:75'], ['10.6.90.230', '80:58:F8:A3:C1:C1'], ['10.6.89.77', '70:70:0D:D8:0B:5B'], ['10.6.89.49', '1C:1A:C0:6E:CB:32'], ['10.6.90.239', 'D0:25:98:48:EF:5F'], ['10.6.91.250', '28:B2:BD:85:6D:08'], ['10.6.91.255', '54:35:30:DD:BC:C1'], ['10.6.91.48', '28:F0:76:23:87:E6'], ['10.6.91.141', 'EC:9B:F3:D9:82:E0'], ['10.6.89.151', 'CC:25:EF:32:10:CE'], ['10.6.91.41', '28:F0:76:24:A0:D2'], ['10.6.92.148', '04:F7:E4:2A:D1:9C'], ['10.6.92.141', '1C:1A:C0:76:CA:B5'], ['10.6.92.144', '40:E2:30:62:00:39'], ['10.6.84.151', 'E0:B9:BA:08:63:A3'], ['10.6.93.181', '2C:F0:A2:D9:2B:05'], ['10.6.84.153', 'E0:B9:BA:08:64:36'], ['10.6.84.155', '78:02:F8:BE:74:7F'], ['10.6.93.185', '28:C2:DD:38:E6:EB'], ['10.6.84.158', '34:14:5F:C9:FA:01'], ['10.6.84.159', '90:3C:92:88:CA:A8'], ['10.6.93.188', '28:C2:DD:5D:36:89'], ['10.6.91.74', 'EC:1F:72:69:38:1E'], ['10.6.90.247', 'F0:D1:A9:C9:C6:B7'], ['10.6.90.245', '10:A5:D0:E8:65:93'], ['10.6.93.209', '40:E2:30:61:FF:E1'], ['10.6.92.15', '28:F0:76:24:A4:7A'], ['10.6.92.17', '54:35:30:DD:EA:21'], ['10.6.92.16', '28:F0:76:10:0D:D0'], ['10.6.92.11', '5C:F7:E6:DF:A0:E1'], ['10.6.92.12', '54:35:30:DD:8A:79'], ['10.6.82.232', '00:B3:62:99:78:91'], ['10.6.86.227', 'EC:1F:72:34:1F:53'], ['10.6.82.237', '28:C2:DD:38:EC:95'], ['10.6.82.236', '28:C2:DD:5D:31:ED'], ['10.6.91.39', 'F0:24:75:AD:4E:13'], ['10.6.82.63', '00:56:CD:7A:58:8D'], ['10.6.91.31', '28:F0:76:24:A4:98'], ['10.6.91.109', '28:F0:76:0F:A6:CE'], ['10.6.81.210', '48:43:7C:7D:70:58'], ['10.6.81.216', '74:E1:B6:5A:A0:97'], ['10.6.81.217', '1C:1A:C0:74:4C:4E'], ['10.6.81.215', 'E4:9A:79:CA:3C:D3'], ['10.6.83.5', '20:AB:37:48:29:F5'], ['10.6.93.119', '90:2E:1C:51:52:93'], ['10.6.93.110', '28:C2:DD:20:65:BF'], ['10.6.90.142', 'F0:99:BF:28:9D:7F'], ['10.6.90.143', '1C:1A:C0:72:9A:53'], ['10.6.82.242', '28:C2:DD:46:4D:17'], ['10.6.82.243', '28:C2:DD:5D:32:39'], ['10.6.82.241', '28:C2:DD:38:E6:CB'], ['10.6.88.103', '40:E2:30:61:FB:9D'], ['10.6.88.109', '1C:1A:C0:79:FE:09'], ['10.6.95.1', '70:70:0D:E6:31:42'], ['10.6.95.191', '90:2E:1C:51:B5:2B'], ['10.6.83.20', '90:2E:1C:52:64:76'], ['10.6.81.137', 'A0:91:69:A2:ED:05'], ['10.6.81.134', '74:E1:B6:81:97:7D'], ['10.6.81.133', '40:E2:30:61:C3:37'], ['10.6.80.191', '60:D9:C7:8A:41:70'], ['10.6.91.183', '38:71:DE:1C:51:1B'], ['10.6.93.64', '78:FD:94:9D:7C:F7'], ['10.6.84.202', '00:CD:FE:82:89:A9'], ['10.6.80.198', 'A0:99:9B:0D:FE:4F'], ['10.6.84.209', 'A4:B8:05:9A:95:FA'], ['10.6.91.4', '78:4B:87:DD:E7:53'], ['10.6.88.179', '40:E2:30:61:D9:BD'], ['10.6.91.202', 'E0:5F:45:98:10:1B'], ['10.6.84.59', 'AC:CF:85:01:2B:3E'], ['10.6.91.206', '28:F0:76:10:0E:D4'], ['10.6.91.207', '1C:5C:F2:40:8C:9B'], ['10.6.91.205', '90:48:9A:4F:63:DD'], ['10.6.91.208', 'A4:B8:05:7D:7C:E9'], ['10.6.91.160', '90:48:9A:4F:5F:75'], ['10.6.88.207', 'B8:44:D9:5D:0C:23'], ['10.6.92.248', '40:E2:30:61:FB:91'], ['10.6.88.194', '68:DB:CA:11:D0:0B'], ['10.6.92.247', '40:E2:30:61:FD:09'], ['10.6.88.197', '1C:1A:C0:6E:E2:79'], ['10.6.80.125', '18:F6:43:5A:61:15'], ['10.6.80.124', '28:C2:DD:38:EC:5D'], ['10.6.80.142', 'AC:37:43:A0:58:05'], ['10.6.90.199', '88:66:A5:0A:48:78'], ['10.6.90.198', '60:D9:C7:8A:09:1D'], ['10.6.84.115', '60:D9:C7:8A:33:FB'], ['10.6.84.110', '1C:99:4C:B6:04:B2'], ['10.6.91.81', '54:35:30:DD:BC:BB'], ['10.6.82.194', '8C:8E:F2:31:52:80'], ['10.6.88.71', '1C:1A:C0:70:3B:DE'], ['10.6.91.87', '84:38:35:A1:59:D3'], ['10.6.84.99', '60:D9:C7:8B:44:53'], ['10.6.90.201', '60:D9:C7:8A:2E:56'], ['10.6.89.204', '90:3C:92:AC:F0:03'], ['10.6.89.207', '24:F0:94:77:1A:E6'], ['10.6.95.35', 'A4:08:EA:21:47:32'], ['10.6.89.201', '78:31:C1:AC:AD:73'], ['10.6.89.200', 'F8:27:93:A6:2F:98'], ['10.6.94.235', '54:35:30:DD:B3:71'], ['10.6.89.98', '64:9A:BE:A7:FD:43'], ['10.6.82.178', '2C:20:0B:09:99:8A'], ['10.6.91.77', '1C:1A:C0:76:0E:7B'], ['10.6.80.228', '1C:1A:C0:76:7B:08'], ['10.6.91.72', '1C:1A:C0:6E:C9:7D'], ['10.6.80.223', '60:D9:C7:8A:2C:DC'], ['10.6.81.254', 'D0:C5:F3:05:9F:86'], ['10.6.80.225', '60:D9:C7:8B:29:D5'], ['10.6.92.171', '7C:04:D0:7C:76:02'], ['10.6.92.176', 'E0:B9:BA:17:1E:60'], ['10.6.92.178', '40:E2:30:62:01:69'], ['10.6.90.106', '1C:1A:C0:75:01:50'], ['10.6.90.103', 'D8:CF:9C:E4:3E:EB'], ['10.6.86.56', '4C:66:41:32:45:37'], ['10.6.91.80', '1C:1A:C0:6E:C5:61'], ['10.6.89.115', '2C:F0:A2:E1:61:E9'], ['10.6.92.68', '54:35:30:DD:BC:B7'], ['10.6.92.62', '90:48:9A:4F:55:F3'], ['10.6.90.51', 'E8:50:8B:50:6B:23'], ['10.6.90.57', 'D0:03:4B:B0:7F:7C'], ['10.6.90.56', 'F4:31:C3:6B:03:DF'], ['10.6.90.55', 'B4:8B:19:BE:80:63'], ['10.6.82.11', '74:E1:B6:5A:8B:E9'], ['10.6.83.218', '28:BE:03:CB:6C:34'], ['10.6.81.222', '28:C2:DD:33:96:CD'], ['10.6.83.210', '28:C2:DD:38:ED:9F'], ['10.6.83.213', '28:C2:DD:38:E6:D1'], ['10.6.83.97', 'BC:F5:AC:DE:D0:07'], ['10.6.83.95', '34:AA:8B:09:93:E3'], ['10.6.80.96', '1C:1A:C0:6E:CF:3E'], ['10.6.80.95', '54:35:30:DD:BC:CB'], ['10.6.88.245', 'BC:54:36:34:95:E9'], ['10.6.81.42', '1C:1A:C0:7D:73:FC'], ['10.6.81.40', '00:CD:FE:9F:A4:F4'], ['10.6.81.44', '28:C2:DD:5D:38:51'], ['10.6.81.45', '84:A1:34:48:BC:45'], ['10.6.95.87', '28:C2:DD:38:E6:D9'], ['10.6.83.70', '38:71:DE:EB:92:D2'], ['10.6.83.73', '2C:B4:3A:2B:8B:99'], ['10.6.80.169', '20:A9:0E:30:0C:EE'], ['10.6.83.138', '54:35:30:DD:CC:B9'], ['10.6.88.32', '1C:1A:C0:76:5B:EB'], ['10.6.80.163', '28:C2:DD:46:44:E1'], ['10.6.94.254', '5C:F7:E6:9D:D9:DF'], ['10.6.88.143', 'B8:53:AC:E3:77:89'], ['10.6.81.30', 'A4:B8:05:70:C1:92'], ['10.6.91.231', 'B8:53:AC:45:2E:B7'], ['10.6.89.237', 'FC:C2:DE:A1:56:F2'], ['10.6.95.46', '40:E2:30:61:D5:75'], ['10.6.94.44', '40:E2:30:61:FD:27'], ['10.6.89.51', 'E8:3A:12:A3:BF:1B'], ['10.6.89.240', '20:2D:07:08:73:82'], ['10.6.92.211', '40:E2:30:61:FC:8B'], ['10.6.89.245', '2C:20:0B:17:36:CF'], ['10.6.89.54', '1C:1A:C0:6E:DE:63'], ['10.6.89.57', '70:3E:AC:9B:7D:2D'], ['10.6.89.248', 'CC:61:E5:8F:8B:3E'], ['10.6.91.248', '28:F0:76:10:0E:B2'], ['10.6.91.247', 'E4:9A:79:7C:4F:71'], ['10.6.91.240', '28:C6:8E:FB:D6:F4'], ['10.6.82.134', '68:C4:4D:99:50:FF'], ['10.6.82.135', 'A0:3B:E3:82:85:43'], ['10.6.89.146', '1C:1A:C0:6E:CC:96'], ['10.6.93.93', '28:F0:76:24:BB:7E'], ['10.6.89.145', '28:C6:8E:FB:DA:8E'], ['10.6.95.44', '90:48:9A:4F:43:95'], ['10.6.94.50', '40:E2:30:61:FC:E1'], ['10.6.82.201', '1C:1A:C0:70:0E:CF'], ['10.6.92.139', '40:E2:30:61:FA:B5'], ['10.6.92.138', '40:E2:30:61:FD:37'], ['10.6.92.135', '40:E2:30:61:FA:9F'], ['10.6.92.133', '40:E2:30:61:D5:93'], ['10.6.84.165', '80:E6:50:CB:A0:F5'], ['10.6.93.191', '28:C2:DD:5D:36:FF'], ['10.6.84.167', '40:E2:30:62:01:BD'], ['10.6.93.193', '28:C2:DD:46:28:C5'], ['10.6.84.161', 'E0:B9:BA:81:31:9C'], ['10.6.93.196', '28:C2:DD:26:D5:63'], ['10.6.93.197', '90:48:9A:4F:AB:E9'], ['10.6.86.14', '54:35:30:DD:DF:51'], ['10.6.90.211', '38:CA:DA:88:8F:21'], ['10.6.90.19', '68:AE:20:72:CF:AE'], ['10.6.82.220', '20:7D:74:2D:E1:50'], ['10.6.90.16', '64:BC:0C:46:19:E9'], ['10.6.90.11', '9C:20:7B:5F:5A:6B'], ['10.6.91.175', 'AC:0D:1B:FA:76:3E'], ['10.6.95.174', 'D8:BB:2C:45:48:A5'], ['10.6.91.28', '40:4D:7F:C5:44:49'], ['10.6.95.176', '40:E2:30:61:D4:EF'], ['10.6.91.23', '60:D9:C7:8A:03:5E'], ['10.6.95.178', 'CC:20:E8:18:07:6D'], ['10.6.93.216', '40:E2:30:61:FB:2D'], ['10.6.80.211', '34:E2:FD:E3:A1:99'], ['10.6.80.216', 'DC:41:5F:F2:F1:E2'], ['10.6.80.215', '74:4A:A4:79:27:1F'], ['10.6.83.8', '64:BC:0C:43:01:6A'], ['10.6.90.82', '30:63:6B:C2:2B:42'], ['10.6.90.177', 'E4:8B:7F:ED:3B:9D'], ['10.6.90.176', '6C:8D:C1:AE:7E:BD'], ['10.6.93.121', '4C:32:75:93:C7:59'], ['10.6.90.170', '68:C4:4D:46:C0:03'], ['10.6.88.116', '28:C2:DD:46:2B:3B'], ['10.6.83.34', '1C:1A:C0:75:40:7D'], ['10.6.90.61', 'DC:0C:5C:D6:BC:4A'], ['10.6.83.39', '2C:B4:3A:1D:A0:B9'], ['10.6.81.122', 'CC:20:E8:63:92:DB'], ['10.6.81.123', '38:D4:0B:25:77:6A'], ['10.6.91.192', '28:F0:76:10:0D:C0'], ['10.6.92.28', '90:48:9A:4F:6F:9B'], ['10.6.92.21', '6C:72:E7:A0:70:40'], ['10.6.91.178', 'D0:33:11:03:03:C1'], ['10.6.84.40', '54:35:30:DD:98:F5'], ['10.6.91.173', '28:F0:76:10:0D:C8'], ['10.6.85.106', 'E0:5F:45:26:63:61'], ['10.6.88.230', '10:41:7F:14:F4:AD'], ['10.6.93.72', 'C0:EE:FB:43:23:EB'], ['10.6.89.13', '64:BC:0C:47:61:3E'], ['10.6.82.245', '28:C2:DD:38:ED:13'], ['10.6.93.74', 'CC:08:8D:E9:E6:F9'], ['10.6.81.192', 'A4:B8:05:D5:B2:DF'], ['10.6.94.163', '40:E2:30:62:00:3F'], ['10.6.83.166', '1C:1A:C0:7D:74:E0'], ['10.6.94.160', '40:E2:30:62:01:8D'], ['10.6.80.134', '1C:1A:C0:C3:BF:3F'], ['10.6.80.138', '28:C2:DD:46:02:F3'], ['10.6.84.123', '48:43:7C:3F:95:AB'], ['10.6.84.125', '60:D9:C7:8B:CB:D0'], ['10.6.84.129', '60:D9:C7:8A:42:C0'], ['10.6.94.10', '90:2E:1C:52:69:B2'], ['10.6.90.180', 'F4:31:C3:1A:82:86'], ['10.6.92.229', '40:E2:30:61:F9:EB'], ['10.6.89.219', 'F4:31:C3:70:76:8A'], ['10.6.80.158', '1C:1A:C0:70:B6:EA'], ['10.6.90.217', 'AC:BC:32:40:8B:F6'], ['10.6.89.217', '84:A1:34:AA:D4:13'], ['10.6.88.98', 'E0:B9:BA:88:60:5D'], ['10.6.95.21', 'B4:8B:19:B1:56:54'], ['10.6.95.23', '60:D9:C7:8B:42:53'], ['10.6.95.22', '40:E2:30:61:9D:E3'], ['10.6.88.91', 'A4:B8:05:D4:19:FE'], ['10.6.89.179', '40:E2:30:61:FA:09'], ['10.6.82.140', '24:F0:94:20:10:0B'], ['10.6.91.67', '00:61:71:D5:AC:5C'], ['10.6.92.5', '54:35:30:DD:F0:67'], ['10.6.92.4', '54:35:30:DD:B2:91'], ['10.6.80.255', '5C:AD:CF:1A:64:FC'], ['10.6.91.68', 'B4:74:43:C8:D8:E5'], ['10.6.80.251', '7C:01:91:02:5C:63'], ['10.6.93.169', '28:C2:DD:33:78:E1'], ['10.6.90.138', '1C:1A:C0:75:B4:03'], ['10.6.92.165', '90:2E:1C:51:52:F2'], ['10.6.87.37', 'D4:F4:6F:C7:8C:C3'], ['10.6.90.132', '68:C4:4D:E9:87:4D'], ['10.6.90.134', 'BC:9F:EF:69:E6:23'], ['10.6.89.108', '38:71:DE:9B:1E:BB'], ['10.6.89.109', 'C4:B3:01:8D:B8:74'], ['10.6.89.102', '3C:AB:8E:62:18:60'], ['10.6.92.72', '28:F0:76:10:0E:0A'], ['10.6.92.71', '48:A1:95:A2:4B:15'], ['10.6.92.76', 'E8:B1:FC:B0:B0:E4'], ['10.6.86.240', 'D0:C5:F3:2E:E4:F9'], ['10.6.95.100', '40:E2:30:61:FE:3D'], ['10.6.81.238', 'F4:31:C3:8A:D2:65'], ['10.6.83.220', 'DC:A4:CA:0E:8E:EE'], ['10.6.80.83', '28:C2:DD:46:4D:1B'], ['10.6.87.88', '74:E1:B6:52:CC:7A'], ['10.6.80.89', 'C0:1A:DA:B5:1C:8C'], ['10.6.91.135', '28:F0:76:1D:67:D2'], ['10.6.91.132', '28:F0:76:0F:AA:12'], ['10.6.91.180', '90:48:9A:4F:6F:03'], ['10.6.81.57', '1C:1A:C0:70:33:1A'], ['10.6.81.56', 'C0:CB:38:94:70:26'], ['10.6.83.49', '1C:1A:C0:70:FA:58'], ['10.6.83.41', '40:E2:30:62:01:B1'], ['10.6.83.42', '28:C2:DD:38:EC:67'], ['10.6.81.155', '5C:AD:CF:1E:4E:0B'], ['10.6.81.157', '4C:74:BF:C4:BF:DF'], ['10.6.90.9', 'C4:9A:02:64:9C:E8'], ['10.6.90.8', '6C:3E:6D:8D:09:B5'], ['10.6.90.6', '7C:04:D0:7D:78:53'], ['10.6.90.4', 'C4:B3:01:B8:CE:A9'], ['10.6.83.123', 'A0:99:9B:1D:43:19'], ['10.6.90.1', 'C0:1A:DA:9C:F1:ED'], ['10.6.92.226', '28:F0:76:24:BD:7A'], ['10.6.88.159', '1C:1A:C0:6E:E8:56'], ['10.6.88.154', '90:B6:86:8C:4A:C2'], ['10.6.88.157', '48:A1:95:99:B8:7F'], ['10.6.88.9', '1C:1A:C0:6E:E6:DC'], ['10.6.83.149', 'C0:1A:DA:E0:93:FF'], ['10.6.94.51', '54:35:30:DD:BC:93'], ['10.6.83.196', '28:F0:76:24:BB:32'], ['10.6.80.1', '7C:0E:CE:E9:4F:A7'], ['10.6.89.66', '70:70:0D:C1:E0:7A'], ['10.6.89.255', 'F4:F5:24:E7:0D:0F'], ['10.6.89.252', '14:1F:78:52:A4:5A'], ['10.6.89.61', '28:F0:76:10:0E:00'], ['10.6.93.215', '40:E2:30:61:D9:CB'], ['10.6.93.219', '28:C2:DD:33:79:19'], ['10.6.85.36', '40:E2:30:61:FB:3D'], ['10.6.80.103', '88:28:B3:42:AF:07'], ['10.6.83.156', 'EC:1F:72:83:EC:D1'], ['10.6.80.105', '38:CA:DA:8A:32:1A'], ['10.6.89.133', 'F0:CB:A1:7D:E4:93'], ['10.6.89.132', '1C:5C:F2:1A:30:BA'], ['10.6.87.209', '90:B6:86:7C:DE:12'], ['10.6.89.139', '28:A0:2B:A1:2A:42'], ['10.6.89.138', '80:ED:2C:74:7D:CF'], ['10.6.84.178', '24:F0:94:11:D4:A2'], ['10.6.92.128', '40:E2:30:61:FF:EB'], ['10.6.92.129', '40:E2:30:61:FB:6B'], ['10.6.92.125', '40:E2:30:62:01:8F'], ['10.6.84.170', '28:27:BF:0A:A0:93'], ['10.6.92.121', '54:35:30:DD:F7:11'], ['10.6.92.122', '40:E2:30:61:FF:67'], ['10.6.84.175', '90:2E:1C:51:D2:68'], ['10.6.89.223', 'EC:35:86:77:45:5A'], ['10.6.89.226', '04:DB:56:26:69:D6']]
DEBUG FINISHED.
Shutting down...
<issue_comment>username_1: Hi there. What OS are you running?
<issue_comment>username_1: Closing this issue due to inactivity. Thanks.<issue_closed> |
<issue_start><issue_comment>Title: Fail compile if there are type errors. Fixes #11
username_0: Turns out the TS transpileModule API wasn't returning all of the errors when you ask it to report diagnostics. So I've created a local version of that function using slightly lower level APIs. This seems to do the trick, surfacing the errors about type mismatches. Hopefully it catches all of the types of errors people care about.
<issue_comment>username_1: LGTM. Great fix, thanks @username_0 for your contribution ! |
<issue_start><issue_comment>Title: Circular imports
username_0: I have a Flask app that has a structure like the following:
```
project
-- users
__init__.py
models.py
views.py
serializers.py
-- genres
__init__.py
models.py
views.py
serializers.py
```
I have a ModelSchema class in one of my folders that uses models.py from the other folder. SQLAlchemy by default solves this problem by using the string name of the class but I get circular import errors in Marshmallow SQLAlchemy. Is there a way to solve this? Or shoud I change my structure?
<issue_comment>username_1: i have the same problem and its really bothering . does anyone know how to solve this ?<issue_closed>
<issue_comment>username_2: See https://github.com/marshmallow-code/flask-marshmallow/issues/143; the solution is to keep the schemas in a separate module from models and ensure that models are all initialized before schemas are initialized.
This needs to be documented better, as suggested on the linked issue. |
<issue_start><issue_comment>Title: update wct-local@^2.0.15
username_0: - [X] Changelog updated
<issue_comment>username_1: 42 sandbox.stub(steps, 'prepare', async() => undefined);
~~~~~~~~~~~~~~~~~~~~
test/unit/cli.ts(42,36): error TS7011: Function expression, which lacks return-type annotation, implicitly has an 'any' return type.
43 sandbox.stub(steps, 'runTests', async() => undefined);
~~~~~~~~~~~~~~~~~~~~
test/unit/cli.ts(43,37): error TS7011: Function expression, which lacks return-type annotation, implicitly has an 'any' return type.
80 sandbox.stub(steps, 'prepare', async(_context: Context) => undefined);
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
test/unit/grunt.ts(80,38): error TS7011: Function expression, which lacks return-type annotation, implicitly has an 'any' return type.
95 sandbox.stub(steps, 'runTests', async() => undefined);
~~~~~~~~~~~~~~~~~~~~
test/unit/grunt.ts(95,41): error TS7011: Function expression, which lacks return-type annotation, implicitly has an 'any' return type.
41 sandbox.stub(steps, 'prepare', async(_context: Context) => undefined);
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
test/unit/gulp.ts(41,36): error TS7011: Function expression, which lacks return-type annotation, implicitly has an 'any' return type.
```
<issue_comment>username_2: Tests pass now with fixes landed on master. Merging! |
<issue_start><issue_comment>Title: MatrixExpr: add MatPow doit(), changes to doit for MatAdd, MatMul, Trace
username_0: I'm been looking at `doit()` support in MatrixExpr. In particular, I'm interested in cases where these have ImmutableMatrices inside and they could be evaluated (e.g., after a substitution).
Here are some fixes:
* MatPow: added `doit()`.
* MatPow: refactor my older `.as_explicit()` code to use the the `doit()`.
* MatAdd: had rather ineffective `doit()`, improved.
* MatMul: change the default from `deep=False` to `deep=True`. I cannot find a reason for it, so I've switched it.
* MatMul: clean up test code.
* Trace: change the default to `deep=True`.
The `deep=True` changes might be controversial: if someone knows why it `deep=False` before, I could always revert that (and add a comment to the code!)
Anyway, here is what it does:
````
In [6]: A = Matrix([[1,2],[3,4]])
In [7]: B = Matrix([[2,3],[4,5]])
In [8]: n = Symbol('n', integer=True)
In [9]: w = MatPow(A, n) + MatMul(A, MatPow(B,2*n))
In [10]: pprint(w)
2⋅n n
⎡1 2⎤⋅⎛⎡2 3⎤⎞ + ⎛⎡1 2⎤⎞
⎢ ⎥ ⎜⎢ ⎥⎟ ⎜⎢ ⎥⎟
⎣3 4⎦ ⎝⎣4 5⎦⎠ ⎝⎣3 4⎦⎠
In [21]: pprint(w.subs(n, 1))
2 1
⎡1 2⎤⋅⎛⎡2 3⎤⎞ + ⎛⎡1 2⎤⎞
⎢ ⎥ ⎜⎢ ⎥⎟ ⎜⎢ ⎥⎟
⎣3 4⎦ ⎝⎣4 5⎦⎠ ⎝⎣3 4⎦⎠
In [22]: pprint(w.subs(n, 1).doit()) % these fixes make this work
⎡73 97 ⎤
⎢ ⎥
⎣163 215⎦
````
<issue_comment>username_1: +1. Looks good to me, thanks for this fix. I'm not certain why we had `deep=false` before. Git blame shows @mrocklin would be the one to know. I suspect it has something to do with the intentional non-automatic evaluation of the matrix expressions module, which his thesis relied on to aid in expression rewriting.
<issue_comment>username_0: Thanks @username_1.
I've got a failing test on python 3, I think my Trace test might be exposing a bug in MatAdd._eval_trace. I'll investigate later.
<issue_comment>username_0: Ok, I've used a different test for my doit deep changes. The trace of an MatAdd was failing before this PR too: I thought I'd fixed it but perhaps just papered over it on python2 but not python3. So reverted that change for now. I then added the failure in as an XFAIL (with a FIXME note).
<issue_comment>username_1: Looks good to me. Are you good with merging this now? If you want, you could probably squash the last couple test-fixing commits together to clean up the history a bit. Otherwise, I'll merge if you feel this is *done*.
<issue_comment>username_0: Ok, please merge this. Only change is a issue number for the XFAIL test, and some rebasing.
(It should pass tests, previous failure looks like a timeout.) |
<issue_start><issue_comment>Title: Implement SvgIcon Component
username_0: See #233 for details.
<issue_comment>username_1: I think it'd be nice to have support for both passing the SVG data in directly and also for using an existing SVG [symbol](https://developer.mozilla.org/en-US/docs/Web/SVG/Element/symbol).
Maybe if a `use` or similar prop is there it'll use that value for a `<use>` element's `xlink:href` attribute.
```jsx
render() {
return (
<svg ...>
<use xlink:href={this.props.use} ... />
</svg>
)
}
```
For passing the data directly you can look into how Material-UI does it:
* http://www.material-ui.com/#/components/svg-icon
* https://github.com/callemall/material-ui/blob/master/src/SvgIcon/SvgIcon.js
<issue_comment>username_0: Yeah, I would personally prefer the usage of an svg sprite map over inline solution. I'll make sure both ways are implemented.<issue_closed> |
<issue_start><issue_comment>Title: adapter.serialize receives a snapshot
username_0: Since the beta 16, the serialize method of adapter base class receives a snapshot, not a record. It should not use record._createSnapshot anymore.
<issue_comment>username_1: Nice catch @username_0! Could you update the docs for the method too? (`@param {DS.Model} record`)
<issue_comment>username_0: Done !
<issue_comment>username_1: Do you mind squashing the commits?
<issue_comment>username_0: I have squashed the commits. It should be ready to merge now, isn't it ?
<issue_comment>username_1: Thank you! |
<issue_start><issue_comment>Title: Add manylinux test
username_0: - Create a manylinux distro of the extension sample and check it in.
- Make a new test (probably that extends/calls the existing one) that only runs on linux and that uses the anylinux wheel rather than building one. |
<issue_start><issue_comment>Title: Crashes on startup.
username_0: This might not be a Mantle probem, but right before it crashes the console says the Mantle and Tinkers' Construct environments are healthy.
Crash log:
---- Minecraft Crash Report ----
// Hey, that tickles! Hehehe!
Time: 12/15/14 10:06 AM
Description: There was a severe problem during mod loading that has caused the game to fail
cpw.mods.fml.common.LoaderException: java.lang.NoClassDefFoundError: net/minecraftforge/common/Configuration
at cpw.mods.fml.common.LoadController.transition(LoadController.java:162)
at cpw.mods.fml.common.Loader.preinitializeMods(Loader.java:515)
at cpw.mods.fml.client.FMLClientHandler.beginMinecraftLoading(FMLClientHandler.java:239)
at net.minecraft.client.Minecraft.func_71384_a(Minecraft.java:480)
at net.minecraft.client.Minecraft.func_99999_d(Minecraft.java:867)
at net.minecraft.client.main.Main.main(SourceFile:148)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown Source)
at java.lang.reflect.Method.invoke(Unknown Source)
at net.minecraft.launchwrapper.Launch.launch(Launch.java:135)
at net.minecraft.launchwrapper.Launch.main(Launch.java:28)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown Source)
at java.lang.reflect.Method.invoke(Unknown Source)
at org.multimc.onesix.OneSixLauncher.launchWithMainClass(OneSixLauncher.java:286)
at org.multimc.onesix.OneSixLauncher.launch(OneSixLauncher.java:376)
at org.multimc.EntryPoint.listen(EntryPoint.java:165)
at org.multimc.EntryPoint.main(EntryPoint.java:54)
Caused by: java.lang.NoClassDefFoundError: net/minecraftforge/common/Configuration
at imlookingatblood.pokefenn.ImLookingAtBlood.fmlPreInitiation(ImLookingAtBlood.java:28)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown Source)
at java.lang.reflect.Method.invoke(Unknown Source)
at cpw.mods.fml.common.FMLModContainer.handleModStateEvent(FMLModContainer.java:513)
at sun.reflect.GeneratedMethodAccessor4.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown Source)
at java.lang.reflect.Method.invoke(Unknown Source)
at com.google.common.eventbus.EventSubscriber.handleEvent(EventSubscriber.java:74)
at com.google.common.eventbus.SynchronizedEventSubscriber.handleEvent(SynchronizedEventSubscriber.java:47)
at com.google.common.eventbus.EventBus.dispatch(EventBus.java:322)
at com.google.common.eventbus.EventBus.dispatchQueuedEvents(EventBus.java:304)
at com.google.common.eventbus.EventBus.post(EventBus.java:275)
at cpw.mods.fml.common.LoadController.sendEventToModContainer(LoadController.java:208)
at cpw.mods.fml.common.LoadController.propogateStateMessage(LoadController.java:187)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown Source)
at java.lang.reflect.Method.invoke(Unknown Source)
at com.google.common.eventbus.EventSubscriber.handleEvent(EventSubscriber.java:74)
at com.google.common.eventbus.SynchronizedEventSubscriber.handleEvent(SynchronizedEventSubscriber.java:47)
at com.google.common.eventbus.EventBus.dispatch(EventBus.java:322)
at com.google.common.eventbus.EventBus.dispatchQueuedEvents(EventBus.java:304)
at com.google.common.eventbus.EventBus.post(EventBus.java:275)
at cpw.mods.fml.common.LoadController.distributeStateMessage(LoadController.java:118)
at cpw.mods.fml.common.Loader.preinitializeMods(Loader.java:513)
... 18 more
Caused by: java.lang.ClassNotFoundException: net.minecraftforge.common.Configuration
at net.minecraft.launchwrapper.LaunchClassLoader.findClass(LaunchClassLoader.java:191)
at java.lang.ClassLoader.loadClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
... 45 more
Caused by: java.lang.NullPointerException
at net.minecraft.launchwrapper.LaunchClassLoader.findClass(LaunchClassLoader.java:182)
... 47 more
A detailed walkthrough of the error, its code path and all known details is as follows:
---------------------------------------------------------------------------------------
-- System Details --
Details:
Minecraft Version: 1.7.10
Operating System: Windows 8.1 (amd64) version 6.3
Java Version: 1.7.0_71, Oracle Corporation
Java VM Version: Java HotSpot(TM) 64-Bit Server VM (mixed mode), Oracle Corporation
Memory: 284172840 bytes (271 MB) / 653262848 bytes (623 MB) up to 954728448 bytes (910 MB)
JVM Flags: 3 total; -XX:HeapDumpPath=MojangTricksIntelDriversForPerformance_javaw.exe_minecraft.exe.heapdump -Xms512m -Xmx1024m
AABB Pool Size: 0 (0 bytes; 0 MB) allocated, 0 (0 bytes; 0 MB) used
[Truncated]
FastCraft{1.9} [FastCraft] (fastcraft-1.9.jar) Unloaded->Constructed->Pre-initialized
McMultipart{1.1.0.314} [Minecraft Multipart Plugin] (ForgeMultipart-1.7.10-1.1.0.314-universal.jar) Unloaded->Constructed->Pre-initialized
iChunUtil{4.1.2} [iChunUtil] (iChunUtil-4.1.2.jar) Unloaded->Constructed->Pre-initialized
IGWMod{1.1.1-17} [In-Game wiki Mod] (IGW-Mod-1.7.10-1.1.1-17-universal.jar) Unloaded->Constructed->Pre-initialized
ImLookingAtBlood{1.2} [I'm Looking At Blood] (ImLookingAtBlood-1.6.4-1.2.jar) Unloaded->Constructed->Errored
journeymap{5.0.1} [JourneyMap] (JourneyMap5.0.1_Unlimited_MC1.7.10.jar) Unloaded->Constructed->Pre-initialized
Mantle{1.7.10-0.3.2.jenkins184} [Mantle] (Mantle-1.7.10-0.3.2.jar) Unloaded->Constructed->Pre-initialized
Morph{0.9.1} [Morph] (Morph-Beta-0.9.1.jar) Unloaded->Constructed->Pre-initialized
Natura{2.2.0} [Natura] (natura-1.7.10-2.2.0.1.jar) Unloaded->Constructed->Pre-initialized
RandomThings{2.2.2} [Random Things] (RandomThings-2.2.2.jar) Unloaded->Constructed->Pre-initialized
StevesFactoryManager{A93} [Steve's Factory Manager] (StevesFactoryManagerA93.jar) Unloaded->Constructed->Pre-initialized
TConstruct{1.7.10-1.7.1.build771} [Tinkers' Construct] (TConstruct-1.7.10-1.7.1c.jar) Unloaded->Constructed->Pre-initialized
ThaumicTinkerer{unspecified} [Thaumic Tinkerer] (ThaumicTinkerer-2.5-1.7.10-161.jar) Unloaded->Constructed->Pre-initialized
TiCTooltips{1.1.11b} [TiC Tooltips] (TiCTooltips-mc1.7.10-1.1.11b.jar) Unloaded->Constructed->Pre-initialized
TMechworks{1.7.10-86.e500b88} [Tinkers' Mechworks] (TMechworks-1.7.10-0.2.10.jar) Unloaded->Constructed->Pre-initialized
Translocator{1.1.1.14} [Translocator] (Translocator-1.7.10-1.1.1.14-universal.jar) Unloaded->Constructed->Pre-initialized
aobd{2.3.5} [Another One Bites The Dust] (AOBD-2.3.5.jar) Unloaded->Constructed->Pre-initialized
ForgeMicroblock{1.1.0.314} [Forge Microblocks] (ForgeMultipart-1.7.10-1.1.0.314-universal.jar) Unloaded->Constructed->Pre-initialized
Mantle Environment: Environment healthy.
TConstruct Environment: Environment healthy.
<issue_comment>username_1: Well, here's what to look for:
"ImLookingAtBlood{1.2} I'm Looking At Blood Unloaded->Constructed->Errored"
This is not Mantle's fault . Try redownloading that mod.<issue_closed> |
<issue_start><issue_comment>Title: Fixing spelling of initialize
username_0:
<issue_comment>username_0: I signed it!
<issue_comment>username_0: I signed it!
<issue_comment>username_1: Thanks can you please change the commit to follow our convention? sth like`style: Fix typo in jnitialize`
<issue_comment>username_0: :+1: @username_1 amended the commit message to the proper format
<issue_comment>username_1: Thanks |
<issue_start><issue_comment>Title: The tile size should be configurable.
username_0: It currently is limited to 256, which is a reasonable default, but any power of two square should be reasonable. (Instances of 8 in the code probably refer to Math.log2(*tileSize*).) In theory any width and height would also be possible but I think that generalization would make the code more complicated than necessary for now.
<issue_comment>username_1: There's also an interesting [page on "meta-tiles"](http://wiki.openstreetmap.org/wiki/Meta_tiles), which could be 2048x2048.
After some searching I have not found an actual tile provider that yields other than 256×256 tiles. Will post again here if I find one.
<issue_comment>username_2: Mapzen's API serves up 512x512 tiles:
https://mapzen.com/documentation/vector-tiles/use-service/#specify-tile-size
I'll take a stab at implementing configurable tile sizes (keeping default as 256).<issue_closed> |
<issue_start><issue_comment>Title: TRPL: dining philosophers
username_0: This is a little rough, and it needs squashed and section headers, but i'd like to get some eyes on it sooner rather than later.
<issue_comment>username_1: r? @nikomatsakis
(rust_highfive has picked a reviewer for you, use r? to override)
<issue_comment>username_0: oh, and i might not need the channel at the end: https://twitter.com/florob/status/597935275394912259
considering an extra 'cleanup' section
<issue_comment>username_0: I believe all nits are addressed. r? @username_3
<issue_comment>username_2: So, maybe I'm just terribly dense, but I still don't see why you need the channel at all. AFAICT, it does not become obsolete by joining the threads, but is unnecessary from the get go.
<issue_comment>username_3: I do also agree with @username_2 that the channel involvement here may not be pulling its weight by the time we get to the end.
<issue_comment>username_0: I added a thing at the end about removing the channel entirely.
<issue_comment>username_0: @username_2 yeah, maybe that is true. sigh.
<issue_comment>username_0: Okay, I removed the channels entirely, and use unwrap.
<issue_comment>username_3: Looks good to me! A few last minor nits and otherwise r=me
Nice chapter @username_0!
<issue_comment>username_4: :pushpin: Commit 2ba6169 has been approved by `username_3`
<!-- @username_4 r=username_3 2ba61698ccb2c43ad1ea081988ef499735bfa62d -->
<issue_comment>username_0: @username_4: r=username_3 rollup |
<issue_start><issue_comment>Title: Updates need for gradle 2.4
username_0: The library needs to be updated to support gradle 2.4 which has an updated constructor signature for the DefaultDeployerFactory.
Here is the update in the MavenPlugin that can be copied
https://github.com/gradle/gradle/blob/master/subprojects/maven/src/main/groovy/org/gradle/api/plugins/MavenPlugin.java
<issue_comment>username_1: Hi,
I've released version 1.3 of the plugin which is compatible with gradle 2.4: https://plugins.gradle.org/plugin/com.github.username_1.android-maven
Also not that I have changed the artifact name which is now android-maven-gradle-plugin, but I have kept the same plugin id.
It It should get sync to jcenter and central soon, unless the new artifact id causes a problem.<issue_closed> |
<issue_start><issue_comment>Title: Error using hasPermissionTo() method
username_0: Error using hasPermissionTo() method.
ErrorException in HasRoles.php line 215:
Call to a member function contains() on null
I had change the hasDirectPermission($permission) method for this
if (is_string($permission)) {
$permission = app(Permission::class)->findByName($permission);
if (!$permission) {
return false;
}
}
$permissions = collect($this->permissions());
return $permissions->contains('id', $permission->id);
<issue_comment>username_1: [These tests](https://github.com/spatie/laravel-permission/blob/1cb77dfe2c8c7fff7bf6dc67e9ffb75ffc007743/tests/RoleTest.php#L18-L28) indicated that hasPermissionTo is working as intended.
Could you submit a failing test or give detailed instruction on how this error can be recreated in a vanilla Laravel app?
<issue_comment>username_0: ok, I has been using a code like this in a controller:
$module='purchase.request';
$this->permisos = [
'create' => $module.'.create',
'update' => $module.'.update'
];
foreach ($this->permisos as $permiso) {
Permission::create(['name' => "$permiso"]);
$this->user->givePermissionTo(['name' => "$permiso"]);
}
dd($this->user->hasPermissionTo($this->permisos['create']));
// output
// FatalThrowableError in HasRoles.php line 217:
// Call to a member function contains() on null
<issue_comment>username_1: You're passing in an array into `givePermissionTo`. It expects a string.
Make this change:
```php
$this->user->givePermissionTo($permiso);
```<issue_closed>
<issue_comment>username_0: I'am using Laravel 5.3
<issue_comment>username_2: I am have the same problem when use can , @can directive and when I use hasPermissionTo() method
<issue_comment>username_3: Im having the same issue. Is it because the permission name has a dot period in it?
In my case, my permission name is 'module_name.create'
If I use module_name_create it working fine.
<issue_comment>username_3: I having the same problem in Laravel 5.8.
The exception thrown:
`Symfony\Component\Debug\Exception\FatalThrowableError : Call to a member function contains() on array at ....\HasPermissions.php:258`
Checked on the source:
This is the line:
`return $this->permissions->contains('id', $permission->id);`
Use gettype on `gettype($this->permissions)`, return Array.
So I assume the author is trying to use Larave's collection method: contains.
Replace that line with:
`return collect($this->permissions)->contains('id', $permission->id);`
Everything work perfectly.
Question, why $this->permissions became array and in the code is trying to assume it is an collection object??
<issue_comment>username_4: It would seem that perhaps something in your own application code is interfering with `$this->permissions`. The usual cause of this is creating a `permissions` property on the User object, which prevents the BelongsTo relationship from being used. |
<issue_start><issue_comment>Title: Enter number Amount with left handed style
username_0: Hello Guys,
I know number is is generally come with right side alignment. But actually i've some little bit problems during enter numbers with 2 decimals like
Example -
The default text style with mask is look like 0,00
Currently when i need to enter 12.00 i must need to enter amount 1200 but it will be possible that i'm just enter amount 12 and then rest of decimal .00 is automatically added at last and if user will need to change .00 values then he will change it itself by moving cursor.
But my requirement is to add 12 as a before .00 by default when i start entering amount.
Thanks in advance.
Dilip.
<issue_comment>username_0: The another way to tell this is there is any possibility to trigger this input mask when onblur event is triggered.
<issue_comment>username_1: Related to #118
Pull requests are welcome.
<issue_comment>username_1: Closing since it is a duplicate of #118.<issue_closed> |
<issue_start><issue_comment>Title: find_element_by_name notice err
username_0: because webdriver discarded find_element_by_name function, i to run it notice error in ``ios-simple.py``. but after i to using xpath get this element, it's works!
<issue_comment>username_0: hi.budy, i'm change find_element_by_name to find_element_by_accessibility_id. please review. thx.
RDs
<issue_comment>username_1: Thanks @username_0! |
<issue_start><issue_comment>Title: Updating the number of pages does not redraw
username_0: Changing the number of pages with
`self.pageControl.numberOfPages = newNum`
as result the old drawing is not cleared
<img width="247" alt="screen shot 2017-03-29 at 13 34 57" src="https://cloud.githubusercontent.com/assets/907926/24452683/a978a174-1484-11e7-8260-6181858a7b48.png">
Mauro.
<issue_comment>username_1: hi @username_0, thanks for bug report.
Which page control are you using?<issue_closed>
<issue_comment>username_1: @username_0 please try new version 0.1.1
<issue_comment>username_0: Yeah, it works!
Sorry I didn't check for an update first. |
<issue_start><issue_comment>Title: Wrong optimisation
username_0: Hello,
consider the following problem: maximise `P = 3x + 4y` subject to:
```
x + y ≤ 4
2x + y ≤ 5
x ≥ 0, y ≥ 0
```
the maximum `P=16` can be obtained with: `x = 0, y = 4`
The same linear optimisation in `cassowary`:
```javascript
var c = require('cassowary');
var solver = new c.SimplexSolver();
var p = new c.Variable({name: 'p'})
, x = new c.Variable({name: 'x'})
, y = new c.Variable({name: 'y'});
solver.addConstraint(new c.Equation(p, c.plus(c.times(x, 3), c.times(y, 4))));
solver.addConstraint(new c.Inequality(c.plus(x, y), c.LEQ, 4));
solver.addConstraint(new c.Inequality(c.plus(c.times(x, 2), y), c.LEQ, 5));
solver.addConstraint(new c.Inequality(x, c.GEQ, 0));
solver.addConstraint(new c.Inequality(y, c.GEQ, 0));
solver.optimize(p);
console.log('p: ', p) //15
console.log('x: ', x) //1
console.log('y: ', y) //3
```
results in `P=15, x=1, y=3`, which is not the optimum solution. Any idea on what I'm doing wrong?
Thanks
Daniele
<issue_comment>username_1: Hello Daniele,
This reply might come a bit late but I think I can solve your problem. You have 2 issues:
1 - You need to call ```solver.resolve()``` in order to run the optimisation (I think that by default only a feasible solution will be computed, not the optimal one)
2 - Cassowary will solve a minimization problem, therefore you need to inverse the sign of p to get the value of ```P=16``` that you were looking for, for instance by changing your constraint ```P = -3x - 4y```
The following code seem to work:
```javascript
var solver = new c.SimplexSolver();
var p = new c.Variable({name: 'p'})
, x = new c.Variable({name: 'x'})
, y = new c.Variable({name: 'y'});
solver.addConstraint(new c.Equation(p, c.plus(c.times(x, -3), c.times(y, -4))));
solver.addConstraint(new c.Inequality(c.plus(x, y), c.LEQ, 4));
solver.addConstraint(new c.Inequality(c.plus(c.times(x, 2), y), c.LEQ, 5));
solver.addConstraint(new c.Inequality(x, c.GEQ, 0));
solver.addConstraint(new c.Inequality(y, c.GEQ, 0));
solver.optimize(p);
solver.resolve();
console.log('p: ', p) //-16
console.log('x: ', x) //0
console.log('y: ', y) //4
``` |
<issue_start><issue_comment>Title: Fix mysensors callback race
username_0: ## Description:
* Fix possible race at startup in mysensors callback.
* Update devices via persistence before starting gateway to avoid
two threads calling the same callback at the same time.
* Call add_devices max once per callback.
**Related issue (if applicable):**
Reported in gitter:
https://gitter.im/home-assistant/home-assistant/devs?at=58a821edde50490822e334a0
## Checklist:
If the code communicates with devices, web services, or third-party tools:
- [x] Local tests with `tox` run successfully. **Your PR cannot be merged unless tests pass**
[ex-requir]: https://github.com/home-assistant/home-assistant/blob/dev/homeassistant/components/keyboard.py#L14
[ex-import]: https://github.com/home-assistant/home-assistant/blob/dev/homeassistant/components/keyboard.py#L54
<issue_comment>username_0: I'm getting this error every time I shut down `hass` when I'm testing https://github.com/home-assistant/home-assistant/pull/6264:
```
17-02-28 23:06:44 ERROR (Dummy-14) [homeassistant.core] Error doing job: Future exception was never retrieved
Traceback (most recent call last):
File "/usr/lib/python3.5/concurrent/futures/thread.py", line 55, in run
result = self.fn(*self.args, **self.kwargs)
File "/home/martin/Dev/home-assistant/homeassistant/components/mysensors.py", line 191, in gw_start
gateway.event_callback('persistence', node_id)
File "/home/martin/Dev/home-assistant/homeassistant/components/mysensors.py", line 331, in node_update
callback(self, node_id)
File "/home/martin/Dev/home-assistant/homeassistant/components/mysensors.py", line 277, in mysensors_callback
devices[key].schedule_update_ha_state(True)
File "/home/martin/Dev/home-assistant/homeassistant/helpers/entity.py", line 288, in schedule_update_ha_state
self.hass.add_job(self.async_update_ha_state(force_refresh))
AttributeError: 'NoneType' object has no attribute 'add_job'
```
I need to investigate this some more.
<issue_comment>username_1: You need start your gateway on EVENT_HOMEASSISTANT_START event. After this event, you are sure that all device exists on statemachine.
<issue_comment>username_1: also possible, use the new dispatcher to broadcast data from callback and on entity register dispatcher in `async_added_to_hass`
<issue_comment>username_0: I think it was a false alarm. I manually repatched this change on top of #6264 when testing, instead of rebasing, and forgot to apply the whole patch.
Testing again after rebasing on dev works fine.
@username_1 thanks, I'll look into those new goodies (dispatcher and `async_added_to_hass`) for another PR. |
<issue_start><issue_comment>Title: Charset for Russian and Chinese
username_0: Hi,
I have problems with Russian and Chinese language in the frontend. It shows '????????'
Can you help me to solve it please?
Moisés, can you add my Skype (aleksandr.galimon) please?
Thanks<issue_closed> |
<issue_start><issue_comment>Title: unknown maker filter
username_0: 
Should we delete this filter? We can either clean the elasticsearch database or filter this value on the server side. Any thoughts?
<issue_comment>username_1: It’s on the list of approved changes to the new index, just waiting for it to be implemented.
<issue_comment>username_0: This issue will be resolved with the new elasticsearch index<issue_closed> |
<issue_start><issue_comment>Title: Allow people to hide the user on posts?
username_0: I honestly don't know that we care about this one too much, but as there are character aliases to hide a character's name (or whatever), also allow people to hide (not change the display of, as that is very easily abusable) the user on a post, to make it less clear while reading a thread who owns which characters.
Yep, MWF request. |
<issue_start><issue_comment>Title: Fix installation instructions for Ubuntu
username_0: The previous url was incorrect and returns error while doing sudo apt-get update
<!--Thanks for your contribution. See [CONTRIBUTING](CONTRIBUTING.md)
for this project's contribution guidelines. Remove these comments
as you go.
DO NOT edit files and directories listed in _data/not_edited_here.yaml.
These are maintained in upstream repos and changes here will be lost.
Help us merge your changes more quickly by adding details and setting metadata
(such as labels, milestones, and reviewers) over at the right-hand side.-->
### Proposed changes
Fix repository path for Ubuntu installation instructions. The original instructions cause problems while performing ```sudo apt-get update```
```
W: The repository 'https://apt.dockerproject.org/repo/pool xenial Release' does not have a Release file.
N: Data from such a repository can't be authenticated and is therefore potentially dangerous to use.
N: See apt-secure(8) manpage for repository creation and user configuration details.
E: Failed to fetch https://apt.dockerproject.org/repo/pool/dists/xenial/main/binary-amd64/Packages 403 Forbidden
E: Some index files failed to download. They have been ignored, or old ones used instead.
```
### Unreleased project version (optional)
<!--If this change only applies to an unreleased version of a project, note
that here and base your work on the `vnext-` branch for your project. If
this doesn't apply to this PR, you can remove this whole section.
Set a milestone if appropriate. -->
### Related issues (optional)
<!--Refer to related PRs or issues: #1234, or 'Fixes #1234' or 'Closes #1234'.
Or link to full URLs to issues or pull requests in other Github projects -->
<issue_comment>username_0: Nvm, seems someone else already fixed it! |
<issue_start><issue_comment>Title: Compiling Erlang with mix, doesn't generate beam files if using {:d, VAR, VALUE} in erlc_options
username_0: I'm using mix to compile some Erlang modules, and I've got the following in `src/foo.erl`:
```
-module(foo).
-export([bar/0]).
bar() -> ok.
```
This compiles fine with `mix compile`.
I then add `erlc_options: [{:d, 'BAZ', 'quux'}]` to `mix.exs` and run `mix clean`.
Now when I run `mix compile`, mix reports the following:
```
Compiling 1 file (.erl)
Generated bug app
```
There are no `.beam` files in the `_build/dev/lib/ebin` directory and the `bug.app` file has an empty `{modules,[]}` term.
Running `mix compile` again results in:
```
Compiling 1 file (.erl)
```
...and it never generates a beam file.
Setting `erlc_options: []` causes it to start working correctly again.
----
My mix.exs file looks like this:
```
defmodule Bug.Mixfile do
use Mix.Project
def project do
[app: :bug,
version: "0.1.0",
#erlc_options: [],
erlc_options: [{:d, 'BAZ', 'quux'}],
deps: []]
end
def application do
[]
end
end
```
<issue_comment>username_1: @username_0 can you please push a small project to github that we can run `mix compile` on and verify? That will make our life much easier if we already have a mechanism to directly reproduce the issue. Thank you!
<issue_comment>username_0: ```
$ mix -v
Erlang/OTP 20 [erts-9.0] [source] [64-bit] [smp:12:12] [ds:12:12:10] [async-threads:10] [hipe] [kernel-poll:false]
Mix 1.4.4
```
...incidentally.
<issue_comment>username_1: So the error is because you should write: `erlc_options: [{:d, :BAZ, :quux}]`, i.e. as Elixir atoms. I will make sure to not silently discard on those cases but the error message will just say "ArgumentError", because unfortunately that's all Erlang gives us.<issue_closed> |
<issue_start><issue_comment>Title: Draggable textdraw bug
username_0: Hello,
First, thanks a lot for your wonderfull work !
I used it for a program that allows to draw zones on a map and place detector (technical application).
Recently I had a problem with draggable elements that.
I developped first the program with version v13.07.03.
Since yesterday (the code has not been changed) it's impossible to drag any of the elements (arc or text). The cursor change on mouseover but it's impossible to drag things (click events are OK).
I then switch to v20.0.0.
Elements are now draggable, but I have a problem with Y position of dragged text.
When I drag texts horizontally, it's ok. But when I drag them vertically, they move twice slower than the mouse (if yPos of the mouse is 500, Ypos of the drown text is 250). That problem only appear on drown text (I also draw arc, they move correctly).
Here is an exemple of the code I use:
The section "DRAW ZONES" works properly : I can move the corners of the drown zone.
The problem appear with text drawn in the "DRAW ZONE NUMBER" part
Thanks in advance for your help :) !
```
function drawZones() {
// INIT
$canvas.removeLayers();
zoneIDs = $("#zoneIDs").val();
zones = zoneIDs.split(';');
nbZones = zones.length - 1;
$canvas.clearCanvas();
colorsBgrdTxt = $('#bgrd_colors').val();
colorsBgrdTab = colorsBgrdTxt.split(';');
radius = $('#d_poignee_zone').val();
var couleurIndice = 1;
// DRAW ZONES
for (var zoneIndice = 0; zoneIndice < nbZones; zoneIndice += 1) {
if (zones[zoneIndice] != selectedZone || selectedZone == 0) {
$canvas.removeLayer("layer" + zones[zoneIndice]);
elt = $("#zone" + zones[zoneIndice] + "Pts");
var obj = {
layer: true,
name: "layer" + zones[zoneIndice],
strokeStyle: colorsBgrdTab[zoneIndice],
strokeWidth: 3,
rounded: true,
closed: true
};
ptsStr = $(elt).val();
pts = ptsStr.split(';');
nbPts = pts.length - 1;
for (var p = 0; p < nbPts; p += 1) {
coord = pts[p].split(',');
obj['x' + (p + 1)] = parseFloat(coord[0]);
obj['y' + (p + 1)] = parseFloat(coord[1]);
}
$canvas.drawLine(obj);
// Remove corners
var line = $canvas.getLayer("layer" + zones[zoneIndice]);
for (var i = 1; i <= nbPts; i += 1) {
$canvas.removeLayer("arc" + zones[zoneIndice] + "_" + i);
}
}
else {
couleurIndice = zoneIndice;
}
}
// DRAW SELECTED ZONE FOR IT TO GET THEI HEIGHER Z-INDEX
if (selectedZone != 0) {
$canvas.removeLayer("layer" + selectedZone);
elt = $("#zone" + selectedZone + "Pts");
var obj = {
layer: true,
name: "layer" + selectedZone,
strokeStyle: colorsBgrdTab[couleurIndice],
strokeWidth: 3,
fillStyle: colorsBgrdTab[couleurIndice],
opacity: .7,
rounded: true,
closed: true
};
ptsStr = $(elt).val();
[Truncated]
$('body').css('cursor', 'default');
},
drag: function(joint) {
var j = joint.data.zoneId;
xNumZonePos = joint.x;
yNumZonePos = joint.y;
$("#zone" + j + "Num").val(xNumZonePos + ',' + yNumZonePos);
},
dragstop : function(joint) {
recordNumZoneCoord();
}
});
if (mouseEvent == 'deletePoint')
$canvas.removeLayer("arc" + selectedZone + "_" + i);
mouseEvent = '';
drawDetectors(selectedZone);
}
};
```
<issue_comment>username_1: Interesting. Keep in mind that you upgraded from v13.07.03 to v20.0.0 (with a ton of versions in-between), so there is bound to be at least a few jCanvas code patterns that don't work anymore (or that should be done differently now). You also have a ton of code here, but I will try to help you as best I can.
To start, how often are you calling `drawZones()`? One thing I notice is that you are calling `clearCanvas()`, `removeLayer()` and `removeLayers()` in your function. If you only intend to hide those layers temporarily, or if you want to update them with new values, you should really be using `setLayer()` and `drawLayers()` (as needed) instead. See [this section of the docs](https://projects.calebevans.me/jcanvas/docs/clearCanvas/#a-note-about-layers), as well as [this section](https://projects.calebevans.me/jcanvas/docs/manipulateLayers/) about manipulating layer properties in general.<issue_closed>
<issue_comment>username_1: Hello,
First, thanks a lot for your wonderfull work !
I used it for a program that allows to draw zones on a map (image loadedd) and place detectors (technical application).
Recently I had a problem with draggable elements.
I developped first the program with version v13.07.03.
Since yesterday (the code has not been changed) it's impossible to drag any of the elements (arc or text). The cursor change on mouseover but it's impossible to drag things (click events are OK).
I then switch to v20.0.0.
Elements are now draggable, but I have a problem with Y position of dragged text.
When I drag texts horizontally, it's ok. But when I drag them vertically, they move twice slower than the mouse (if yPos of the mouse is 500, Ypos of the drown text is 250). That problem only appear on drown text (I also draw arc, they move correctly).
Here is an exemple of the code I use. That exemple allow to draw a zone (with corners that are draggable) and set a zone name (that is also draggable).
The section "DRAW ZONES" works properly : I can move the corners of the drown zone.
The problem appear with text drawn in the "DRAW ZONE NUMBER" part
Thanks in advance for your help :) !
```js
function drawZones() {
// INIT
$canvas.removeLayers();
zoneIDs = $("#zoneIDs").val();
zones = zoneIDs.split(';');
nbZones = zones.length - 1;
$canvas.clearCanvas();
colorsBgrdTxt = $('#bgrd_colors').val();
colorsBgrdTab = colorsBgrdTxt.split(';');
radius = $('#d_poignee_zone').val();
var couleurIndice = 1;
// DRAW ZONES
for (var zoneIndice = 0; zoneIndice < nbZones; zoneIndice += 1) {
if (zones[zoneIndice] != selectedZone || selectedZone == 0) {
$canvas.removeLayer("layer" + zones[zoneIndice]);
elt = $("#zone" + zones[zoneIndice] + "Pts");
var obj = {
layer: true,
name: "layer" + zones[zoneIndice],
strokeStyle: colorsBgrdTab[zoneIndice],
strokeWidth: 3,
rounded: true,
closed: true
};
ptsStr = $(elt).val();
pts = ptsStr.split(';');
nbPts = pts.length - 1;
for (var p = 0; p < nbPts; p += 1) {
coord = pts[p].split(',');
obj['x' + (p + 1)] = parseFloat(coord[0]);
obj['y' + (p + 1)] = parseFloat(coord[1]);
}
$canvas.drawLine(obj);
// Remove corners
var line = $canvas.getLayer("layer" + zones[zoneIndice]);
for (var i = 1; i <= nbPts; i += 1) {
$canvas.removeLayer("arc" + zones[zoneIndice] + "_" + i);
}
}
else {
couleurIndice = zoneIndice;
}
}
// DRAW SELECTED ZONE FOR IT TO GET THEI HEIGHER Z-INDEX
if (selectedZone != 0) {
$canvas.removeLayer("layer" + selectedZone);
elt = $("#zone" + selectedZone + "Pts");
var obj = {
layer: true,
name: "layer" + selectedZone,
strokeStyle: colorsBgrdTab[couleurIndice],
strokeWidth: 3,
fillStyle: colorsBgrdTab[couleurIndice],
opacity: .7,
rounded: true,
closed: true
};
ptsStr = $(elt).val();
[Truncated]
$('body').css('cursor', 'default');
},
drag: function(joint) {
var j = joint.data.zoneId;
xNumZonePos = joint.x;
yNumZonePos = joint.y;
$("#zone" + j + "Num").val(xNumZonePos + ',' + yNumZonePos);
},
dragstop : function(joint) {
recordNumZoneCoord();
}
});
if (mouseEvent == 'deletePoint')
$canvas.removeLayer("arc" + selectedZone + "_" + i);
mouseEvent = '';
drawDetectors(selectedZone);
}
};
```
<issue_comment>username_1: @username_0 any update on this issue?
<issue_comment>username_0: Hello Caleb,
I'm back pretty late... Sorry for that
Actually, the problem has disapered for most of users (not for me). Very surprising !!!
I tried many things but did not solved and mainly understood the problem.
The application is used at the moment and then I don't want to take any risk.
It won't be used this summer, and then I will try again to solve the problem.
Best regards,
Sébastien
<issue_comment>username_1: @username_0 for what it's worth, what web browser do the majority of your users use? What web browser are *you* using to test the application?
<issue_comment>username_0: Caleb531,
I tried with FF, Chrome and IE.
Users use most of the time FF or Chrome.
Regards,
Sébastien.
<issue_comment>username_1: Hmm, and you experience the issue when you yourself test with either FF or Chrome?
<issue_comment>username_0: Actually I do not use the application.
Even if it's for my company, I just developped it but do not use it.
In a few weeks now, I will move it to another hosting (the hosting is shared at the moment).
Perhaps problems come from modifications that could have been done on the hosting.
I'll let you know...
<issue_comment>username_1: Alright, sure! I'll close this issue for now, but if the issue persists after you change hosts, let me know and I'll re-open the issue. :)<issue_closed> |
<issue_start><issue_comment>Title: SPF?
username_0: Hi, I have an issue with spf. You told us that mioght cause problems and it does, when Spf is on player moving left and right couple second, somethimes is restart at the half of the video but loadig faster. When Spf is turned off than player opens normaly but tkes couple minutes that the page load.
Any solutions, any adjustments or fix for that?
Thank you. |
<issue_start><issue_comment>Title: Instruments crashes when clicking buttons - IOS Only
username_0: On IOS I'm having issues when clicking buttons sometimes, the app crashes. I cannot get it happen manually for the developer, so we do not think it related to the APP code.
I'm simply trying to click a Save button after I have clicked 2 buttons to hide some events. You can see it around line 1150 in the log file. This is happening all the time.
info: [debug] [INST] 2016-01-25 21:30:57 +0000 Fail: The target application appears to have died
The appium debug log is located here for investgation.
https://gist.github.com/username_0/c4a018b19d753aa34701
<issue_comment>username_1: Hi @username_0, you clicked the wrong xpath, perhaps? Can you try click xpath using of Appium Inspector. Please check this.
I think if you can click a button, you can click all of them.
<issue_comment>username_0: The mapping of the button is correct. Sometimes it works, sometimes it does not. The problem is "the target application appears to have died" thus the button element not available anymore.
org.openqa.selenium.WebDriverException: Instruments died while responding to command
<issue_comment>username_2: I am hitting the same issue that "Instruments died while responding to command", It's more likely an Apple's bug...
<issue_comment>username_3: +1<issue_closed> |
<issue_start><issue_comment>Title: Build error of file permission
username_0: $(CL_KERNELS_SH) has no permission problem.
diff
---------------------------------------
# Copy the OpenCL kernels into C++ char strings
$(CL_KERNELS_CPP) : $(CL_HEADERS) $(CL_KERNELS)
+ chmod +x $(CL_KERNELS_SH)
$(CL_KERNELS_SH)
proto: $(PROTO_GEN_CC) $(PROTO_GEN_HEADER)
<issue_comment>username_1: Ok, good!
Note that this will be changed soon (or later) anyways, removing the need for the script.<issue_closed>
<issue_comment>username_1: Should be fixed in the latest version. |
<issue_start><issue_comment>Title: What's really new
username_0: Replaces #194 and #237, because apparently I forgot everything I used to know about how git works.
Closes #191.
<issue_comment>username_1: Great, thanks @username_0
<issue_comment>username_1: Oh crap -- actually I just realized that a reference to this file needs to be added to https://github.com/LABSN/expyfun/blob/master/doc/source/expyfun.rst for it to render properly |
<issue_start><issue_comment>Title: luci-base: Update Simplified Chinese translation
username_0:
<issue_comment>username_1: @username_0 Hi. You efforts are appreciated, but could I trouble you to update the commit message to indicate what part of LuCI for which you are updating strings; we'd like to avoid non-specific commit messages. Title is fine, but the commit message should be a little more specific. Just saying something like 'updates strings in luci-base' is fine. Hope you don't mind!
<issue_comment>username_2: no response from the PR opener, negative feedback.
closing. |
<issue_start><issue_comment>Title: flushregs on RRCmds because they can invalidate the GDB caches
username_0: What do you think about this fix?
https://sourceware.org/gdb/onlinedocs/gdb/Maintenance-Commands.html
We might get hit later by another cache but so far this seems to work ok.
I wrote a testcase to make sure we do the right thing when hitting the end of the program and we don't. I've added it as a todo for now.
<issue_comment>username_1: This definitely won't handle changes to the shared library list. Though neither would the more elaborate scheme I proposed :-(. And there may be other issues lurking, or things may break in future gdb releases.
So I guess this will probably mostly work, but it could be fragile. When we document this, we should document that.
<issue_comment>username_2: Does the `sharedlibrary` work for that purpose or does GDB ignore it because it thinks nothing changed?
<issue_comment>username_1: I don't know. What you really want is something that makes gdb rescan the shared library list and only reload symbols for libraries that have changed. I remember looking for something like that and being unable to find it.
<issue_comment>username_0: This will fix issue #1806.
<issue_comment>username_1: I disabled the `history` test because it fails on Debian 8.5 due to this bug.
<issue_comment>username_0: Thanks, this shouldn't take very long to address. I'll wrap this up when I'm done moving. |
<issue_start><issue_comment>Title: Add configuration setting to set selion log file limits.
username_0: With the recent changes to `SimpleLogger` (which is used by `SeLionLogger`) https://github.com/paypal/SeLion/commit/003b48052acddf8e7c3ba7f7499cbe893eda8b46 we have an opportunity to add a new configuration option that will control the log file limits. For example, we can now limit the number of log files and/or the size before rolling to a new file. We should define a default behavior, add it to our `SeLionLoggerSettings`, and give the user the ability to override the default behavior.
<issue_comment>username_1: @username_0 : I am getting this done. Will send out a pull request after testing
<issue_comment>username_0: This was fixed with #151<issue_closed> |
<issue_start><issue_comment>Title: Possible brace expansion issue
username_0: I was setting up some local benchmark tests and ran into a possible issue with brace expansion. Can you verify whether or not I correctly set the `look-up` options?
<issue_comment>username_1: I think you would need to either pass `matchBase` on the options, or use a globstar `**`. e.g.
```js
**/.{,config/}waldo
```
<issue_comment>username_1: let me know if neither of those work. if not, we have a bug
<issue_comment>username_0: I am currently passing matchBase. I'll give `**/` a try. Full source available in the [develop branch of `find-config`](https://github.com/username_0/find-config/blob/develop/test/look-up.bench.js).
<issue_comment>username_0: Still no dice after adding `**/`.
<issue_comment>username_1: oh, wait. that's a dotfile pattern. try passing `dot: true`
<issue_comment>username_1: and maybe some combination of those other options. since it's also matching on an absolute path
<issue_comment>username_0: Unfortunately still getting `null`. I know that the target file exists because the `findup-sync` and `find-config` tests are written exactly the same way and are in the same folder.
```js
var lookup = require('look-up'),
path = require('path'),
options = {
cwd: path.join(__dirname, 'fixtures/a/b'),
unixify: true,
dot: true,
nodupes: true,
matchBase: true,
braces: true,
brackets: true,
extglob: true,
nonull: true,
cache: true
};
console.log(lookup('.{,config/}waldo', options));
console.log(lookup('**/.{,config/}waldo', options));
```
<issue_comment>username_1: hmm, I'll need to look into it. I doubt it's brace expansion. it's possible, but I've used this pattern lot's of times. it's more likely a bug here.
thanks for reporting
<issue_comment>username_1: wait, try one more thing. don't pass any other options besides `dot: true`.
<issue_comment>username_1: and `cwd`, sorry
<issue_comment>username_0: Still `null`, sorry to say.
```js
var lookup = require('look-up'),
path = require('path'),
options = {
cwd: path.join(__dirname, 'fixtures/a/b'),
dot: true
};
console.log(lookup('.{,config/}waldo', options));
console.log(lookup('**/.{,config/}waldo', options));
```
<issue_comment>username_1: k. thanks for checking. I'll look into it
<issue_comment>username_0: This works:
```js
var lookup = require('look-up'),
path = require('path'),
options = {
cwd: path.join(__dirname, 'fixtures/a/b'),
dot: true
};
function test() {
return lookup(['.waldo', '.config/waldo'], options);
}
console.log(test());
```
<issue_comment>username_0: I've narrowed this down. It's a tree walking issue, not a brace expansion issue. There is only one path to drill down into the test fixtures (`a/b/c/d/e/f/g/h/i/j/`). So, when looking for `**/c/package.json` from `j/`, it will be found by the readdir calls and match the glob pattern because `c/` is an ancestor of `j/`, and `package.json` is a child of `c/`. So far so good.
However, if you fork the directory tree, we find the issue. Consider this structure where `d/` and `k/` are siblings:
```
a/ - b/ - c/ - d/ - e/ - f/ - g/ - h/ - i/ - j/
`- k/ - l/ - package.json
```
When looking for `**/c/k/l/package.json` from `j/`, nothing will be found because the loop never calls `readdir` on the contents of `k/` and below. This is what's happening in the case of `.{,config/}waldo`. Because I'm looking into `.config/` which is not an ancestor of the starting cwd, `waldo` is never found because `readdir` is never called on `.config/`.
I'm not sure what the fix is, but would you be open to a PR to add a couple fixtures and a failing test?
<issue_comment>username_1: of course, that would be great, thanks! I think I know how to solve, let me think about it I'll try to get a fix pushed up soon<issue_closed>
<issue_comment>username_1: closing since this lib was deprecated in favor of findup-sync |
<issue_start><issue_comment>Title: Full update of document
username_0: Current API (11.0.1) only allows to do partial update of documents. For my application I would like to have ability to use "full" update of document like in ES REST API. Something like this:
```
PUT localhost:9200/test/type1/1
{
"counter" : 1,
"tags" : ["red"]
}
```
Any ideas, how could it be done with current API, or is this really not supported?<issue_closed>
<issue_comment>username_1: You are correct, the update API is only for modifying documents. The index API is for replacing them. |
<issue_start><issue_comment>Title: Use https:// Git URL in ~/vm-setup/.git/config
username_0: Make sure we use the https:// URL before building the .ova image
1. because you can't `update-vm --pull` without setting up your keypair first with ssh:// URLs
1. because it's more firewall friendly to use HTTPS |
<issue_start><issue_comment>Title: Implement $paths
username_0: We need the ability to build a document to multiple paths. Example syntax below.
Before implementing, we need to figure out how to handle the case `{{g.doc('/content/pages/foo.yaml').url.path}}` – which path would that use? Need a way to specify the path formatter value as an argument – so need to take that into consideration when designing the feature.
```
$paths:
- /foo/
- /bar/
- /baz/
```
<issue_comment>username_1: I think we need to getter define the use case for this. It seems you want to parameterize the document based on url variables, but it is unclear on why we would really want to do that.<issue_closed> |
<issue_start><issue_comment>Title: latest development build: hs.inspect(hs.application) crashes Hammerspoon
username_0: I'll try and load the latest formal release of Hammerspoon later, and test this, but with the latest development build, typing either of the following into the console results in a crash:
hs.inspect(hs.appliaction)
hs.inspect(hs.window)
Even limiting the depth to 1 -- hs.inspect(hs.application,{depth=1}) -- doesn't make a difference; it still crashes. Using hs.inspect on other hs modules (hs.screen, etc.) seems to work fine, though I've only spot checked these.
I sometimes use this when I forget the exact spelling of a method I want to use, so I can't exactly recall the last time I used it and it worked... it may have been a while, since I alternate between this method and using doc.hs.application (or help("hs.application")) (both of which still work).
Since we have the help/tostring method of seeing the official docs, it's less important to me that this works than that it doesn't crash -- ideally the user should have to really try hard to make things crash, and even then it shouldn't happen with core code!
I'll try some regressions later to determine when it started -- I've been burning through development versions lately.
Top of the crash log:
Exception Type: EXC_BAD_ACCESS (SIGSEGV)
Exception Codes: KERN_INVALID_ADDRESS at 0x0000000000000000
VM Regions Near 0:
-->
__TEXT 000000010de04000-000000010de96000 [ 584K] r-x/rwx SM=COW /Volumes/VOLUME/*/Hammerspoon.app/Contents/MacOS/Hammerspoon
Application Specific Information:
Performing @selector(tryMessage:) from sender HSGrowingTextField 0x6000001b7220
Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0 internal.so 0x000000010e0eb60f uielement_eq + 31
1 org.hammerspoon.Hammerspoon 0x000000010de159c5 luaD_precall + 458 (ldo.c:340)
2 org.hammerspoon.Hammerspoon 0x000000010de15cf1 luaD_call + 68 (ldo.c:421)
3 org.hammerspoon.Hammerspoon 0x000000010de26699 luaT_callTM + 175 (ltm.c:96)
4 org.hammerspoon.Hammerspoon 0x000000010de28094 luaV_equalobj + 473 (lvm.c:436)
5 org.hammerspoon.Hammerspoon 0x000000010de29e1a luaV_execute + 5922 (lvm.c:1037)
6 org.hammerspoon.Hammerspoon 0x000000010de15cfd luaD_call + 80 (ldo.c:422)
7 org.hammerspoon.Hammerspoon 0x000000010de154ae luaD_rawrunprotected + 89 (ldo.c:145)
8 org.hammerspoon.Hammerspoon 0x000000010de160b7 luaD_pcall + 68 (ldo.c:644)
9 org.hammerspoon.Hammerspoon 0x000000010de0e2d1 lua_pcallk + 121 (lapi.c:945)
10 org.hammerspoon.Hammerspoon 0x000000010de10c56 luaB_pcall + 86 (lbaselib.c:440)
11 org.hammerspoon.Hammerspoon 0x000000010de159c5 luaD_precall + 458 (ldo.c:340)
12 org.hammerspoon.Hammerspoon 0x000000010de28e23 luaV_execute + 1835 (lvm.c:1084)
13 org.hammerspoon.Hammerspoon 0x000000010de15cfd luaD_call + 80 (ldo.c:422)
14 org.hammerspoon.Hammerspoon 0x000000010de154ae luaD_rawrunprotected + 89 (ldo.c:145)
15 org.hammerspoon.Hammerspoon 0x000000010de160b7 luaD_pcall + 68 (ldo.c:644)
16 org.hammerspoon.Hammerspoon 0x000000010de0e2d1 lua_pcallk + 121 (lapi.c:945)
17 org.hammerspoon.Hammerspoon 0x000000010de08bc3 MJLuaRunString + 209 (MJLua.m:144)
18 org.hammerspoon.Hammerspoon 0x000000010de07a34 -[MJConsoleWindowController tryMessage:] + 153 (MJConsoleWindowController.m:100)<issue_closed> |
<issue_start><issue_comment>Title: throw ArgumentNullException on UIApplication.Main (args, null, "AppDelegate")
username_0: throw ArgumentNullException on UIApplication.Main (args, null, "AppDelegate");
when adding FormsPlugin.Iconize.iOS.IconControls.Init() to AppDelegate
Plugin.Iconize.Iconize.With(new Plugin.Iconize.Fonts.MaterialModule());
global::Xamarin.Forms.Forms.Init();
**FormsPlugin.Iconize.iOS.IconControls.Init();**
LoadApplication(new App());
<issue_comment>username_1: Add this to your Info.plist file (as mentioned here: [Setup](https://github.com/username_2/Xamarin.Plugins/tree/master/Iconize)
```
<key>UIAppFonts</key>
<array>
<string>iconize-material.ttf</string>
</array>
```<issue_closed> |