1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
import sys
sys.path.append('/srv/gargantext')
from admin.env import *
import re
import locale
from lxml import etree
from lxml.html import html5parser
from datetime import datetime, date
from django.utils import timezone
import dateutil.parser
import dateparser
from parsing.FileParsers.FileParser import FileParser
from parsing.NgramsExtractors import *
from admin.utils import PrintException
class EuropressFileParser(FileParser):
def _parse(self, file):
localeEncoding = "fr_FR"
codif = "UTF-8"
count = 0
if isinstance(file, str):
file = open(file, 'rb')
print(file)
contents = file.read()
encoding = self.detect_encoding(contents)
if encoding != "utf-8":
try:
contents = contents.decode("latin1", errors='replace').encode(codif)
except:
PrintException()
# try:
# contents = contents.decode(encoding, errors='replace').encode(codif)
# except Exception as error:
# print(error)
try:
html_parser = etree.HTMLParser(encoding=codif)
html = etree.fromstring(contents, html_parser)
try :
format_europresse = 50
html_articles = html.xpath('/html/body/table/tbody')
if len(html_articles) < 1:
html_articles = html.xpath('/html/body/table')
if len(html_articles) < 1:
format_europresse = 1
html_articles = html.xpath('//div[@id="docContain"]')
if len(html_articles) < 1 :
format_europresse = 50.2
html_parser = html5parser.etree.HTMLParser(encoding=codif)
html = html5parser.etree.fromstring(contents, html_parser)
html_articles = html.xpath('//article')
if len(html_articles) < 1:
print("no article found")
except :
PrintException()
if format_europresse == 50 :
name_xpath = "./tr/td/span[@class = 'DocPublicationName']"
header_xpath = "./tr/td/span[@class = 'DocHeader']"
title_xpath = "string(./tr/td/span[@class = 'TitreArticleVisu'])"
text_xpath = "./tr/td/descendant-or-self::*[not(self::span[@class='DocHeader'])]/text()"
elif format_europresse == 1 :
name_xpath = "//span[@class = 'DocPublicationName']"
header_xpath = "//span[@class = 'DocHeader']"
title_xpath = "string(//div[@class = 'titreArticleVisu'])"
text_xpath = "./descendant::*[\
not(\
self::div[@class='Doc-SourceText'] \
or self::span[@class='DocHeader'] \
or self::span[@class='DocPublicationName'] \
or self::span[@id='docNameVisu'] \
or self::span[@class='DocHeader'] \
or self::div[@class='titreArticleVisu'] \
or self::span[@id='docNameContType'] \
or descendant-or-self::span[@id='ucPubliC_lblCertificatIssuedTo'] \
or descendant-or-self::span[@id='ucPubliC_lblEndDate'] \
or self::td[@class='txtCertificat'] \
)]/text()"
doi_xpath = "//span[@id='ucPubliC_lblNodoc']/text()"
elif format_europresse == 50.2 :
name_xpath = "./header/div/span[@class = 'DocPublicationName']"
header_xpath = "./header/div/span[@class = 'DocHeader']"
title_xpath = "./header/div/span[@class = 'TitreArticleVisu'])"
text_xpath = "./header/div/descendant-or-self::*[not(self::span[@class='DocHeader'])]/text()"
except Exception as error :
PrintException()
# parse all the articles, one by one
try:
for html_article in html_articles:
hyperdata = {}
if len(html_article):
for name in html_article.xpath(name_xpath):
#print("test name.text")
if name.text is not None:
#print(name.text)
format_journal = re.compile('(.*), (.*)', re.UNICODE)
test_journal = format_journal.match(name.text)
if test_journal is not None:
hyperdata['journal'] = test_journal.group(1)
hyperdata['volume'] = test_journal.group(2)
else:
hyperdata['journal'] = name.text.encode(codif)
countbis = 0
for header in html_article.xpath(header_xpath):
# print(count)
# countbis += 1
# try:
# print('109', hyperdata['publication_date'])
# except:
# print('no date yet')
# pass
try:
text = header.text
#print("header", text)
except Exception as error:
print(error)
if isinstance(text, bytes):
text = text.decode(encoding)
if format_europresse == 50.2:
# TODO here check the split if needed: 'Brest Ville, mercredi 26 novembre 2014'
try:# # 2015-oct-08 exception added
text = text.split(', ')[1]
except:
pass
format_date_fr = re.compile('\d*\s*\w+\s+\d{4}', re.UNICODE)
format_date_fr_v2 = re.compile('\s*\w+\s+\d+\s+\w+\s+\d{4}', re.UNICODE)
if text is not None:
test_date_fr = format_date_fr.match(text)
#TODO check the v2 format here
test_date_fr_v2 = format_date_fr_v2.match(text)
format_date_en = re.compile('\w+\s+\d+,\s+\d{4}', re.UNICODE)
test_date_en = format_date_en.match(text)
format_sect = re.compile('(\D+),', re.UNICODE)
test_sect = format_sect.match(text)
format_page = re.compile(', p. (\w+)', re.UNICODE)
test_page = format_page.match(text)
else:
test_date_fr = None
test_date_fr_v2 = None
test_date_en = None
test_sect = None
test_page = None
if test_date_fr is not None or test_date_fr_v2 is not None:
self.localeEncoding = "fr_FR"
locale.setlocale(locale.LC_ALL, "fr_FR.utf-8")
if encoding != "utf-8":
text = text.replace('י', 'é')
text = text.replace('ű', 'û')
text = text.replace(' aot ', ' août ')
try:
hyperdata['publication_date'] = dateparser.parse(text, languages=['fr'])
except:
try :
hyperdata['publication_date'] = datetime.strptime(text, '%d %B %Y')
except :
try:
hyperdata['publication_date'] = datetime.strptime(text, '%B %Y')
except :
try:
locale.setlocale(locale.LC_ALL, "fr_FR")
hyperdata['publication_date'] = datetime.strptime(text, '%d %B %Y')
# hyperdata['publication_date'] = dateutil.parser.parse(text)
except :
# TODO format to parse: ' mercredi 26 novembre 2014'
try :
hyperdata['publication_date'] = datetime.strptime(text, ' %A %d %B %Y')
except Exception as error:
print(error, text)
pass
if test_date_en is not None:
localeEncoding = "en_GB.UTF-8"
locale.setlocale(locale.LC_ALL, localeEncoding)
try :
hyperdata['publication_date'] = datetime.strptime(text, '%B %d, %Y')
except :
try :
hyperdata['publication_date'] = datetime.strptime(text, '%B %Y')
except :
pass
if test_sect is not None:
hyperdata['section'] = test_sect.group(1).encode(codif)
if test_page is not None:
hyperdata['page'] = test_page.group(1).encode(codif)
# try:
# print('183', hyperdata['publication_date'])
# except:
# print('no date yet')
# pass
#
hyperdata['title'] = html_article.xpath(title_xpath).encode(codif)
hyperdata['abstract'] = html_article.xpath(text_xpath)
line = 0
br_tag = 10
for i in html_articles[count].iter():
# print line, br, i, i.tag, i.attrib, i.tail
if i.tag == "span":
if "class" in i.attrib:
if i.attrib['class'] == 'TitreArticleVisu':
line = 1
br_tag = 2
if line == 1 and i.tag == "br":
br_tag -= 1
if line == 1 and br_tag == 0:
try:
hyperdata['authors'] = str.title(etree.tostring(i, method="text", encoding=codif)).encode(codif)#.split(';')
except:
hyperdata['authors'] = 'not found'
line = 0
br_tag = 10
try:
if hyperdata['publication_date'] is not None or hyperdata['publication_date'] != '':
try:
back = hyperdata['publication_date']
except Exception as e:
#print(e)
pass
else:
try:
hyperdata['publication_date'] = back
except Exception as e:
print(e)
except :
hyperdata['publication_date'] = timezone.now()
#if lang == 'fr':
#hyperdata['language_iso2'] = 'fr'
#elif lang == 'en':
# hyperdata['language_iso2'] = 'en'
hyperdata['publication_year'] = hyperdata['publication_date'].strftime('%Y')
hyperdata['publication_month'] = hyperdata['publication_date'].strftime('%m')
hyperdata['publication_day'] = hyperdata['publication_date'].strftime('%d')
#hyperdata.pop('publication_date')
if len(hyperdata['abstract'])>0 and format_europresse == 50:
hyperdata['doi'] = str(hyperdata['abstract'][-9])
hyperdata['abstract'].pop()
# Here add separator for paragraphs
hyperdata['abstract'] = str(' '.join(hyperdata['abstract']))
hyperdata['abstract'] = str(re.sub('Tous droits réservés.*$', '', hyperdata['abstract']))
elif format_europresse == 1:
hyperdata['doi'] = ' '.join(html_article.xpath(doi_xpath))
hyperdata['abstract'] = hyperdata['abstract'][:-9]
# Here add separator for paragraphs
hyperdata['abstract'] = str(' '.join(hyperdata['abstract']))
else:
hyperdata['doi'] = "not found"
# try:
# hyperdata['length_words'] = len(hyperdata['abstract'].split(' '))
# except:
# PrintException()
hyperdata['length_letters'] = len(hyperdata['abstract'])
hyperdata['bdd'] = u'europresse'
hyperdata['url'] = u''
#hyperdata_str = {}
for key, value in hyperdata.items():
hyperdata[key] = value.decode() if isinstance(value, bytes) else value
yield hyperdata
count += 1
file.close()
except Exception as error:
print(error)
pass
if __name__ == "__main__":
e = EuropressFileParser()
e.parse(str(sys.argv[1]))