Datasets:

Modalities:
Text
Formats:
json
Libraries:
Datasets
pandas
License:
scampion commited on
Commit
1f981be
1 Parent(s): 7b2d120

Update update.py

Browse files
Files changed (1) hide show
  1. update.py +28 -20
update.py CHANGED
@@ -47,6 +47,8 @@ import requests
47
  from joblib import expires_after
48
  from pdfminer.high_level import extract_text
49
 
 
 
50
 
51
  def clean_text(func):
52
  """
@@ -73,7 +75,9 @@ def get_eurovoc_terms_and_id():
73
  eurovoc_terms_and_id = {}
74
  response = requests.get('http://publications.europa.eu/resource/dataset/eurovoc',
75
  headers={'Accept': 'application/xml',
76
- 'Accept-Language': 'en'}
 
 
77
  )
78
  data = xmltodict.parse(response.content)
79
  for term in data['xs:schema']['xs:simpleType']['xs:restriction']['xs:enumeration']:
@@ -98,6 +102,7 @@ def get_sparql_query(d):
98
 
99
  def get_json_response(d):
100
  url = "https://publications.europa.eu/webapi/rdf/sparql"
 
101
  params = {"default-graph-uri": "",
102
  "query": get_sparql_query(d),
103
  "format": "application/sparql-results+json",
@@ -105,7 +110,7 @@ def get_json_response(d):
105
  "debug": "on",
106
  "run": "Run Query"}
107
 
108
- response = requests.get(url, params)
109
  assert response.status_code == 200
110
  return response.json()
111
 
@@ -123,7 +128,7 @@ def get_docs(d):
123
  results = get_json_response(d)
124
  for r in results['results']['bindings']:
125
  terms = r['subjects']['value'].replace(u'\xa0', u' ').split(',')
126
- r['eurovoc_concepts'] = terms #list(get_concepts_id(terms))
127
  r['url'] = r['cellarURIs']['value']
128
  r['title'] = r['title']['value']
129
  r['date'] = r['date']['value']
@@ -144,20 +149,19 @@ def get_docs_text(d):
144
 
145
  def get_body(r):
146
  try:
147
- if 'pdf' in r['formats']:
148
- r['text'] = get_pdf_body(r)
149
- elif 'docx' in r['formats']:
150
- r['text'] = get_docx_body(r)
151
- elif 'doc' in r['formats']:
152
- r['text'] = get_doc_body(r)
153
- elif 'xhtml' in r['formats']:
154
- r['text'] = get_xhtml_body(r)
155
- else:
156
- log.warning(f"⚠️ Could not find a parser for {r['formats']}")
157
- return r
158
  except Exception as e:
159
- log.error(str(e) + str(r) )
160
-
161
 
162
 
163
  @clean_text
@@ -166,26 +170,28 @@ def get_pdf_body(r):
166
  url = r['url']
167
  language = r['lang']
168
  accept = 'application/pdf'
169
- response = requests.get(url, headers={'Accept': accept, 'Accept-Language': language})
170
  if response.status_code == 300:
171
  return " ".join(_multiple_choice(get_pdf_body, response, accept, language))
172
  elif response.status_code == 200:
173
  mem = BytesIO(response.content)
174
  return extract_text(mem)
175
 
 
176
  @clean_text
177
  @memory.cache()
178
  def get_xhtml_body(r):
179
  url = r['url']
180
  language = r['lang']
181
  accept = 'application/xhtml+xml'
182
- response = requests.get(url, headers={'Accept': accept, 'Accept-Language': language})
183
  if response.status_code == 300:
184
  return " ".join(_multiple_choice(get_xhtml_body, response, accept, language))
185
  elif response.status_code == 200:
186
  soup = BeautifulSoup(response.content, 'html.parser')
187
  return soup.get_text()
188
 
 
189
  def get_docx_body(r):
190
  accept = 'application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml'
191
  url = r['url']
@@ -209,6 +215,7 @@ def get_doc_body(r):
209
  print(f"⚠️ Could not download {r} --- {accept} {e}")
210
  return ""
211
 
 
212
  def _multiple_choice(func, response, accept, language):
213
  soup = BeautifulSoup(response.text, 'html.parser')
214
  for link in soup.find_all('a'):
@@ -216,10 +223,11 @@ def _multiple_choice(func, response, accept, language):
216
  url = link.attrs['href']
217
  yield func(url, accept, language)
218
 
 
219
  @clean_text
220
  @memory.cache()
221
  def _get_doc_body(url, accept, language='en'):
222
- response = requests.get(url, headers={'Accept': accept, 'Accept-Language': language})
223
  if response.status_code == 300:
224
  return " ".join(_multiple_choice(_get_doc_body, response, accept, language))
225
  elif response.status_code == 200:
@@ -246,6 +254,6 @@ if __name__ == '__main__':
246
  ofiles[ym].flush()
247
  except Exception as e:
248
  log.error('Day ' + str(d) + ' ' + str(e))
249
- print ('Day ' + str(d) + ' ' + str(e))
250
  for f in ofiles.values():
251
  f.close()
 
47
  from joblib import expires_after
48
  from pdfminer.high_level import extract_text
49
 
50
+ user_agent = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36'
51
+
52
 
53
  def clean_text(func):
54
  """
 
75
  eurovoc_terms_and_id = {}
76
  response = requests.get('http://publications.europa.eu/resource/dataset/eurovoc',
77
  headers={'Accept': 'application/xml',
78
+ 'Accept-Language': 'en',
79
+ 'User-Agent': user_agent
80
+ }
81
  )
82
  data = xmltodict.parse(response.content)
83
  for term in data['xs:schema']['xs:simpleType']['xs:restriction']['xs:enumeration']:
 
102
 
103
  def get_json_response(d):
104
  url = "https://publications.europa.eu/webapi/rdf/sparql"
105
+ headers = {'User-Agent': user_agent}
106
  params = {"default-graph-uri": "",
107
  "query": get_sparql_query(d),
108
  "format": "application/sparql-results+json",
 
110
  "debug": "on",
111
  "run": "Run Query"}
112
 
113
+ response = requests.get(url, headers=headers, params=params)
114
  assert response.status_code == 200
115
  return response.json()
116
 
 
128
  results = get_json_response(d)
129
  for r in results['results']['bindings']:
130
  terms = r['subjects']['value'].replace(u'\xa0', u' ').split(',')
131
+ r['eurovoc_concepts'] = terms # list(get_concepts_id(terms))
132
  r['url'] = r['cellarURIs']['value']
133
  r['title'] = r['title']['value']
134
  r['date'] = r['date']['value']
 
149
 
150
  def get_body(r):
151
  try:
152
+ if 'pdf' in r['formats']:
153
+ r['text'] = get_pdf_body(r)
154
+ elif 'docx' in r['formats']:
155
+ r['text'] = get_docx_body(r)
156
+ elif 'doc' in r['formats']:
157
+ r['text'] = get_doc_body(r)
158
+ elif 'xhtml' in r['formats']:
159
+ r['text'] = get_xhtml_body(r)
160
+ else:
161
+ log.warning(f"⚠️ Could not find a parser for {r['formats']}")
162
+ return r
163
  except Exception as e:
164
+ log.error(str(e) + str(r))
 
165
 
166
 
167
  @clean_text
 
170
  url = r['url']
171
  language = r['lang']
172
  accept = 'application/pdf'
173
+ response = requests.get(url, headers={'Accept': accept, 'Accept-Language': language, 'User-Agent': user_agent})
174
  if response.status_code == 300:
175
  return " ".join(_multiple_choice(get_pdf_body, response, accept, language))
176
  elif response.status_code == 200:
177
  mem = BytesIO(response.content)
178
  return extract_text(mem)
179
 
180
+
181
  @clean_text
182
  @memory.cache()
183
  def get_xhtml_body(r):
184
  url = r['url']
185
  language = r['lang']
186
  accept = 'application/xhtml+xml'
187
+ response = requests.get(url, headers={'Accept': accept, 'Accept-Language': language, 'User-Agent': user_agent})
188
  if response.status_code == 300:
189
  return " ".join(_multiple_choice(get_xhtml_body, response, accept, language))
190
  elif response.status_code == 200:
191
  soup = BeautifulSoup(response.content, 'html.parser')
192
  return soup.get_text()
193
 
194
+
195
  def get_docx_body(r):
196
  accept = 'application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml'
197
  url = r['url']
 
215
  print(f"⚠️ Could not download {r} --- {accept} {e}")
216
  return ""
217
 
218
+
219
  def _multiple_choice(func, response, accept, language):
220
  soup = BeautifulSoup(response.text, 'html.parser')
221
  for link in soup.find_all('a'):
 
223
  url = link.attrs['href']
224
  yield func(url, accept, language)
225
 
226
+
227
  @clean_text
228
  @memory.cache()
229
  def _get_doc_body(url, accept, language='en'):
230
+ response = requests.get(url, headers={'Accept': accept, 'Accept-Language': language, 'User-Agent': user_agent})
231
  if response.status_code == 300:
232
  return " ".join(_multiple_choice(_get_doc_body, response, accept, language))
233
  elif response.status_code == 200:
 
254
  ofiles[ym].flush()
255
  except Exception as e:
256
  log.error('Day ' + str(d) + ' ' + str(e))
257
+ print('Day ' + str(d) + ' ' + str(e))
258
  for f in ofiles.values():
259
  f.close()