SaulLu commited on
Commit
2e5e727
1 Parent(s): 821b145

add script

Browse files
Files changed (2) hide show
  1. README.md +23 -0
  2. wikipedia_html_enterprise.py +177 -0
README.md ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # This is an helper script to load an html enterprise dataset into a datasets object
2
+
3
+ ## How to use
4
+
5
+ 1. Download a NS0 dump at https://dumps.wikimedia.org/other/enterprise_html/runs/20230220/
6
+
7
+ 2. Untar it
8
+
9
+ For example with:
10
+ ```
11
+ mkdir enwiki-NS6-20230220-ENTERPRISE-HTML
12
+ tar -I pigz -vxf enwiki-NS6-20230220-ENTERPRISE-HTML.json.tar.gz -C enwiki-NS6-20230220-ENTERPRISE-HTML
13
+ ```
14
+
15
+ 3. Load it:
16
+ ```python
17
+ from datasets import load_dataset
18
+
19
+ local_path=... # Path to directory where you extracted the NS0 dump
20
+ shard_id=...
21
+
22
+ ds = load_dataset("SaulLu/wikipedia_html_enterprise", shard=shard_id, data_dir=local_path)
23
+ ```
wikipedia_html_enterprise.py ADDED
@@ -0,0 +1,177 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # coding=utf-8
2
+ # Copyright 2020 The TensorFlow Datasets Authors and the HuggingFace Datasets Authors.
3
+ #
4
+ # Licensed under the Apache License, Version 2.0 (the "License");
5
+ # you may not use this file except in compliance with the License.
6
+ # You may obtain a copy of the License at
7
+ #
8
+ # http://www.apache.org/licenses/LICENSE-2.0
9
+ #
10
+ # Unless required by applicable law or agreed to in writing, software
11
+ # distributed under the License is distributed on an "AS IS" BASIS,
12
+ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
+ # See the License for the specific language governing permissions and
14
+ # limitations under the License.
15
+
16
+ # Lint as: python3
17
+ """Wikipedia dataset containing cleaned articles of all languages."""
18
+
19
+
20
+ import bz2
21
+ import codecs
22
+ import json
23
+ import re
24
+ import xml.etree.cElementTree as etree
25
+ from urllib.parse import quote
26
+ import mwparserfromhell
27
+ from multiprocess import Process, Manager
28
+ from tqdm import tqdm
29
+ import multiprocessing
30
+ import datasets
31
+ from functools import partial
32
+ from pathlib import Path
33
+
34
+ logger = datasets.logging.get_logger(__name__)
35
+
36
+
37
+ _CITATION = """"""
38
+
39
+ _DESCRIPTION = """"""
40
+
41
+ _LICENSE = (
42
+ "This work is licensed under the Creative Commons Attribution-ShareAlike "
43
+ "3.0 Unported License. To view a copy of this license, visit "
44
+ "http://creativecommons.org/licenses/by-sa/3.0/ or send a letter to "
45
+ "Creative Commons, PO Box 1866, Mountain View, CA 94042, USA."
46
+ )
47
+
48
+ _INFO_FILE = "dumpstatus.json"
49
+
50
+
51
+ _VERSION = datasets.Version("2.0.0", "")
52
+ _NUM_SPLITS = 68
53
+
54
+ class WikipediaConfig(datasets.BuilderConfig):
55
+ """BuilderConfig for Wikipedia."""
56
+
57
+ def __init__(self, shard=None, version=_VERSION, **kwargs):
58
+ """BuilderConfig for Wikipedia.
59
+
60
+ Args:
61
+ split: int, split number.
62
+ **kwargs: keyword arguments forwarded to super.
63
+ """
64
+ super().__init__(
65
+ name=f"shard_{shard}",
66
+ description=f"Wikipedia dataset for split {shard}",
67
+ version=version,
68
+ **kwargs,
69
+ )
70
+ self.shard = shard
71
+ print(f"Split: {self.shard}")
72
+
73
+
74
+ class Wikipedia(datasets.GeneratorBasedBuilder):
75
+ """Wikipedia dataset."""
76
+
77
+ # Use mirror (your.org) to avoid download caps.
78
+ BUILDER_CONFIG_CLASS = WikipediaConfig
79
+ BUILDER_CONFIG = [WikipediaConfig(shard=str(id)) for id in range(_NUM_SPLITS)]
80
+
81
+ def _info(self):
82
+ return datasets.DatasetInfo(
83
+ description=_DESCRIPTION,
84
+ features=datasets.Features(
85
+ {
86
+ "identifier": datasets.Value("string"),
87
+ "name": datasets.Value("string"),
88
+ "namespace_name": datasets.Value("string"),
89
+ "namespace_identifier": datasets.Value("string"),
90
+ "categories": [
91
+ {
92
+ "name": datasets.Value("string"),
93
+ "url": datasets.Value("string"),
94
+ }
95
+ ],
96
+ "date_modified": datasets.Value("string"),
97
+ "url": datasets.Value("string"),
98
+ "html": datasets.Value("string"),
99
+ "wikitext": datasets.Value("string"),
100
+ "in_language": datasets.Value("string"),
101
+ "main_entity": {
102
+ "identifier": datasets.Value("string"),
103
+ "url": datasets.Value("string"),
104
+ },
105
+ "is_part_of" : {
106
+ "name": datasets.Value("string"),
107
+ "identifier": datasets.Value("string"),
108
+ },
109
+ "license":[ {
110
+ "name": datasets.Value("string"),
111
+ "url": datasets.Value("string"),
112
+ "identifier": datasets.Value("string"),
113
+ }]
114
+ }
115
+ ),
116
+ # No default supervised_keys.
117
+ supervised_keys=None,
118
+ homepage="https://dumps.wikimedia.org",
119
+ citation=_CITATION,
120
+ )
121
+
122
+ def _split_generators(self, dl_manager):
123
+ data_paths = [
124
+ Path(self.config.data_dir) / f"enwiki_{self.config.shard}.ndjson"
125
+ ]
126
+ return [
127
+ datasets.SplitGenerator( # pylint:disable=g-complex-comprehension
128
+ name=datasets.Split.TRAIN, gen_kwargs={"filepaths": data_paths}
129
+ )
130
+ ]
131
+
132
+ def _generate_examples(self, filepaths, ):
133
+
134
+
135
+ print("Parsing and cleaning Wikipedia examples")
136
+
137
+ for filepath in filepaths:
138
+ with open(filepath, 'r') as f:
139
+ for line in tqdm(f):
140
+ example = json.loads(line)
141
+ clean_example = {}
142
+ clean_example['name'] = example['name']
143
+ clean_example['identifier'] = example['identifier']
144
+ clean_example['date_modified'] = example['date_modified']
145
+ clean_example['namespace_name'] = example['namespace']["name"]
146
+ clean_example['namespace_identifier'] = example['namespace']["identifier"]
147
+ clean_example["categories"] = example.get("categories", None)
148
+ clean_example['url'] = example['url']
149
+ clean_example['html'] = f'{example["article_body"]["html"]}'
150
+ clean_example['wikitext'] = example['article_body']['wikitext']
151
+ clean_example['in_language'] = example['in_language']
152
+ clean_example['main_entity'] = example.get('main_entity', None)
153
+ clean_example['is_part_of'] = example['is_part_of']
154
+ clean_example['license'] = example['license']
155
+ yield clean_example['identifier'], clean_example
156
+ # num_processes = 16
157
+ # with multiprocessing.Pool(processes=num_processes) as pool:
158
+
159
+ # results = pool.imap_unordered(partial(parse_and_clean), filepaths)
160
+ # for result in results:
161
+ # for example in result:
162
+ # yield example
163
+
164
+ def parse_and_clean(filepath):
165
+ examples = []
166
+ with open(filepath, 'r') as f:
167
+ for line in tqdm(f):
168
+ example = json.loads(line)
169
+ clean_example = {}
170
+ clean_example['id'] = example['identifier']
171
+ clean_example['date_modified'] = example['date_modified']
172
+ clean_example['url'] = example['url']
173
+ clean_example['html'] = f'{example["article_body"]["html"]}'
174
+ clean_example['wikitext'] = example['article_body']['wikitext']
175
+
176
+ examples.append(clean_example)
177
+ return examples