ajax_vocabulary_management.py
20 KB
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
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
# -*- coding: utf-8 -*-
import codecs
import datetime
import HTMLParser
import os
from collections import Counter
from tempfile import mkdtemp, mkstemp
from django.contrib.auth.models import User
from django.db.models import Count, Max, Sum, Q
from django.http import HttpResponse
from django.template.loader import render_to_string
from django.utils.encoding import smart_str
from common.decorators import ajax, AjaxError, render
from dictionary.forms import ManageVocabPermForm
from dictionary.models import Frame_Opinion_Value, Lemma, Lemma_Status, \
POS, Vocabulary, VocabularyFormat, WalentyStat
from dictionary.teixml import createteixml
TEXT_VOCABULARY_CLAUSE = u"""
% The Polish Valence Dictionary (Walenty)
% <date>
%
% The Polish Valence Dictionary (Walenty) is an adaptation of
% the Syntactic Dictionary of Polish Verbs by Marek Świdziński
% in its electronic version provided by Łukasz Dębowski and
% Elżbieta Hajnicz and further expanded by Witold Kieraś to
% include the most frequent verbs in the 1 million sample of
% NKJP (National Corpus of Polish).
%
% The presented resource results from an automatic conversion
% of the aforementioned dictionary, manually reviewed by Filip
% Skwarski to include correct information about a number of new
% features, including sentential subjects, passivisation, and
% control relations.
%
% The format of the new dictionary has been established by Filip
% Skwarski, Elżbieta Hajnicz, Agnieszka Patejuk, Adam Przepiórkowski,
% Marek Świdziński, and Marcin Woliński.
%
% The dictionary has been edited and compiled using a tool
% created by Bartłomiej Nitoń.
%
% The original Syntactic Dictionary of Polish Verbs derives from:
%
% Marek Świdziński
% Institute of Polish
% Warsaw University
% Warsaw, Poland
%
% © Copyright 1998,2012 by Marek Świdziński
%
% This work is distributed under a CC BY-SA license:
% http://creativecommons.org/licenses/by-sa/2.0/
%
% Parameters:
% Dictionaries: <vocabularies>
% Schema opinions: <opinions>
% Lemma statuses: <statuses>
% Owners: <owners>
% Part of speech: <part of speech>
% Opinions added: <opinions added>
%
"""
EN_MONTHS = ['January', 'February', 'March', 'April', 'May', 'June',
'July', 'August', 'September', 'October', 'November', 'December']
@ajax(method='post')
def create_vocabulary(request, form_data):
form_dict = dict((x['name'], x['value']) for x in form_data)
if not form_dict['format']:
raise AjaxError('format not selected')
voc_format_obj = VocabularyFormat.objects.get(pk=form_dict['format'])
lemmas = Lemma.objects.filter(old=False).order_by('entry')
if form_dict['vocabularies']:
lemmas = lemmas.filter(vocabulary__in=form_dict['vocabularies'])
if form_dict['lemma_statuses']:
lemmas = lemmas.filter(status__in=form_dict['lemma_statuses'])
if form_dict['owners']:
lemmas = lemmas.filter(owner__in=form_dict['owners'])
if form_dict['poss']:
lemmas = lemmas.filter(entry_obj__pos__in=form_dict['poss'])
lemmas = lemmas.distinct()
if voc_format_obj.format == u'Tekstowy':
tmp_folder = mkdtemp()
os.chdir(tmp_folder)
tmpfile, tmpfilename = mkstemp(dir=tmp_folder)
os.close(tmpfile)
file_name = create_text_walenty(tmpfilename, lemmas,
form_dict['vocabularies'], form_dict['frame_opinions'],
form_dict['lemma_statuses'], form_dict['owners'],
form_dict['poss'], form_dict['addframeopinion'])
elif voc_format_obj.format == u'TEX':
file_name = create_tex_walenty(lemmas, form_dict)
elif voc_format_obj.format == u'TEI':
file_name = create_tei_walenty(lemmas, form_dict)
return {'file_name': file_name}
def create_tei_walenty(lemmas, form_dict):
tmp_folder = mkdtemp()
os.chdir(tmp_folder)
tmpfile, tmpfilename = mkstemp(dir=tmp_folder)
os.close(tmpfile)
frame_opinion_values = Frame_Opinion_Value.objects.none()
if form_dict['frame_opinions']:
frame_opinion_values = Frame_Opinion_Value.objects.filter(pk__in=form_dict['frame_opinions'])
createteixml(tmpfilename,
lemmas,
frame_opinion_values)
file_name = tmpfilename + '.xml'
os.rename(tmpfilename, file_name)
return file_name
def create_text_walenty(file_name, lemmas, vocabularies, frame_opinions,
lemma_statuses, owners, poss, add_frame_opinions):
try:
f = codecs.open(file_name, 'w+', 'utf-8-sig' )
f.write(create_copyrights_str(vocabularies, frame_opinions,
lemma_statuses, owners, poss, add_frame_opinions))
for lemma in lemmas:
founded_frame_opinions = lemma.frame_opinions.filter(value__in=frame_opinions)
#frame_chars_dict = sorted_frame_char_values_dict()
for reflex_val in lemma.get_existing_frame_char_values(u'ZWROTNOŚĆ'):
for neg_val in lemma.get_existing_frame_char_values(u'NEGATYWNOŚĆ'):
for pred_val in lemma.get_existing_frame_char_values(u'PREDYKATYWNOŚĆ'):
for aspect_val in lemma.get_existing_frame_char_values(u'ASPEKT'):
matching_frames = lemma.get_frames_by_char_values(reflex_val=reflex_val,
neg_val=neg_val,
pred_val=pred_val,
aspect_val=aspect_val).order_by('text_rep')
for frame in matching_frames:
if not lemma.phraseology_ready() and frame.phraseologic:
continue
if not frame_opinions or founded_frame_opinions.filter(frame=frame).exists():
text_rep = frame.get_position_spaced_text_rep()
if add_frame_opinions:
text_rep = add_frame_opinion_and_return_text_rep(text_rep, lemma, frame)
if frame.characteristics.filter(type=u'ZWROTNOŚĆ', value__value=u'się').exists():
text_rep = ' ' + text_rep
f.write(lemma.entry+text_rep.replace(':',': ')+'\n')
os.rename(file_name, file_name+'.txt')
file_name = file_name+'.txt'
finally:
f.close()
return file_name
def create_copyrights_str(vocabularies_pks, frame_opinions_pks,
lemma_statuses_pks, owners_pks, poss_pks,
add_frame_opinions):
date = datetime.datetime.now()
month = EN_MONTHS[date.month-1]
vocabularies = Vocabulary.objects.filter(pk__in=vocabularies_pks).order_by('name')
if not vocabularies.exists():
vocabularies = Vocabulary.objects.order_by('name')
frame_opinions = Frame_Opinion_Value.objects.filter(pk__in=frame_opinions_pks).order_by('priority')
if not frame_opinions.exists():
frame_opinions = Frame_Opinion_Value.objects.order_by('priority')
lemma_statuses = Lemma_Status.objects.filter(pk__in=lemma_statuses_pks).order_by('priority')
if not lemma_statuses.exists():
lemma_statuses = Lemma_Status.objects.order_by('priority')
owners = User.objects.filter(pk__in=owners_pks).order_by('username')
if not owners.exists():
owners = User.objects.filter(lemmas__old=False).distinct().order_by('username')
poss = POS.objects.filter(pk__in=poss_pks).order_by('priority')
if not poss.exists():
poss = POS.objects.exclude(tag=u'unk').order_by('priority')
copyrights = (TEXT_VOCABULARY_CLAUSE.replace('<date>', date.strftime(month + ' %d, %Y')).
replace('<vocabularies>', ', '.join([vocab.name for vocab in vocabularies])).
replace('<opinions>', ', '.join([opinion.value for opinion in frame_opinions])).
replace('<statuses>', ', '.join([status.status for status in lemma_statuses])).
replace('<owners>', ', '.join([owner.username for owner in owners])).
replace('<part of speech>', ', '.join([pos.name for pos in poss])).
replace('<opinions added>', ('True' if add_frame_opinions else 'False')))
return copyrights
def create_tex_walenty(lemmas, form_dict):
q_frame_opinions = []
if form_dict['frame_opinions']:
for pk in form_dict['frame_opinions']:
q_frame_opinions.append(Q(value__pk=pk))
tmp_folder = mkdtemp()
os.chdir(tmp_folder)
tmpfile, tmpfilename = mkstemp(dir=tmp_folder)
h = HTMLParser.HTMLParser()
# Pass the TeX template through Django templating engine and into the temp file
os.write(tmpfile, smart_str(h.unescape(render_to_string('tex/slowal.tex', {'lemmas': lemmas,
'q_frame_opinions': q_frame_opinions,
'download_dict' : form_dict}))))
os.close(tmpfile)
file_name = tmpfilename + '.tex'
os.rename(tmpfilename, file_name)
return file_name
def add_frame_opinion_and_return_text_rep(text_rep, lemma, frame):
try:
frame_op = lemma.frame_opinions.filter(frame__pk=frame.pk).all()[0].value.value
except:
frame_op = 'unk'
frame_form_list = text_rep.split(':')
frame_form_list.insert(1, frame_op)
text_rep = ':'.join(frame_form_list)
return text_rep
def download_vocabulary(request, file_name):
fullpath = '/' + file_name
download_file_name = '%s_%s' % ('walenty', datetime.datetime.now().strftime('%Y%m%d'))
with open(fullpath, "r") as f:
data = f.read()
response = HttpResponse(data, mimetype='text/txt')
if file_name.endswith('.txt'):
response['Content-Disposition'] = 'attachment; filename=%s.txt' % download_file_name
elif file_name.endswith('.tex'):
response['Content-Disposition'] = 'attachment; filename=%s.tex' % download_file_name
elif file_name.endswith('.xml'):
response['Content-Disposition'] = 'attachment; filename=%s.xml' % download_file_name
os.remove(fullpath)
os.rmdir(os.path.split(fullpath)[0])
return response
@render('vocab_perm_manage_form.html')
@ajax(method='get', encode_result=False)
def vocab_perm_manage_form(request, vocabulary_name):
vocabulary_obj = Vocabulary.objects.get(name=vocabulary_name)
form = ManageVocabPermForm(editors=vocabulary_obj.editors,
viewers=vocabulary_obj.viewers)
return {'form': form}
@ajax(method='post')
def vocab_perm_manage_form_submit(request, form_data):
form_dict = dict((x['name'], x['value']) for x in form_data)
vocabulary = Vocabulary.objects.get(name=form_dict['vocabulary_name'])
vocabulary.editors.clear()
vocabulary.viewers.clear()
for user_pk in form_dict['editors']:
selected_user = User.objects.get(pk=user_pk)
vocabulary.editors.add(selected_user)
vocabulary.viewers.add(selected_user)
for user_pk in form_dict['viewers']:
vocabulary.viewers.add(User.objects.get(pk=user_pk))
return {}
@render('vocabulary_stats.html')
@ajax(method='get', encode_result=False)
def get_vocabulary_stats(request, vocabulary_name):
if vocabulary_name:
voc = Vocabulary.objects.get(name=vocabulary_name)
lemmas = voc.lemmas.filter(old=False)
else:
lemmas = Lemma.objects.filter(old=False)
lemma_statuses = Lemma_Status.objects.order_by('priority')
poss = POS.objects.exclude(tag=u'unk').order_by('priority')
top_labels = create_top_labels(poss)
all_statuses_line = create_status_stats_line(lemmas, poss, 'wszystkie')
vocabulary_stats_tab_lines = [top_labels, all_statuses_line]
stats_by_pos = []
for lemma_status in lemma_statuses:
stat_lemmas = lemmas.filter(status__status=lemma_status.status)
stats_line = create_status_stats_line(stat_lemmas, poss, lemma_status.status)
vocabulary_stats_tab_lines.append(stats_line)
stats_by_pos.append(stats_line)
return {'vocabulary_stats': vocabulary_stats_tab_lines}
def create_top_labels(poss):
top_labels = ['Hasła / Schematy', 'wszystkie']
for pos in poss:
top_labels.append(pos.name)
return top_labels
def create_status_stats_line(lemmas, poss, status):
all_count = {'lemmas': lemmas.count(),
'schemas': count_schemas(lemmas)}
status_line = [status, all_count]
for pos in poss:
pos_lemmas = lemmas.filter(entry_obj__pos=pos)
pos_count = {'lemmas': pos_lemmas.count(),
'schemas': count_schemas(pos_lemmas)}
status_line.append(pos_count)
return status_line
def count_schemas(lemmas):
schemas_count = lemmas.annotate(num_frames=Count('frames')).aggregate(Sum('num_frames'))['num_frames__sum']
if not schemas_count:
schemas_count = 0
return schemas_count
@render('other_stats.html')
@ajax(method='get', encode_result=False)
def get_other_stats(request):
return {'stats': WalentyStat.objects.order_by('label')}
def get_stats(statuses, pos):
stats_dict = Counter({u'phrases': 0,
u'poss': 0,
u'lemmas': 0,
u'sub_lemmas': 0,
u'schemata': 0,
u'frames': 0,
u'cer_schemata': 0,
u'uncer_schemata': 0,
u'bad_schemata': 0,
u'arch_schemata': 0,
u'col_schemata': 0,
u'vul_schemata': 0,
u'coor_schemata': 0,
u'lex_schemata': 0,
u'coor_lemmas': 0,
u'lex_lemmas': 0})
lemmas = Lemma.objects.filter(old=False,
entry_obj__pos__tag=pos).filter(status__in=statuses).distinct()
stats_dict[u'lemmas'] = lemmas.count()
for lemma in lemmas.order_by('entry').all():
stats_dict[u'cer_schemata'] += lemma.frame_opinions.filter(value__value=u'pewny').count()
stats_dict[u'uncer_schemata'] += lemma.frame_opinions.filter(value__value=u'wątpliwy').count()
stats_dict[u'bad_schemata'] += lemma.frame_opinions.filter(value__value=u'zły').count()
stats_dict[u'arch_schemata'] += lemma.frame_opinions.filter(value__value=u'archaiczny').count()
stats_dict[u'col_schemata'] += lemma.frame_opinions.filter(value__value=u'potoczny').count()
stats_dict[u'vul_schemata'] += lemma.frame_opinions.filter(value__value=u'wulgarny').count()
stats_dict[u'schemata'] += lemma.frames.count()
stats_dict[u'frames'] += lemma.entry_obj.actual_frames().count()
stats_dict = stats_dict + Counter(get_sub_entries_dict(lemma))
has_phraseology = False
has_coordination = False
for frame in lemma.frames.all():
stats_dict[u'poss'] += frame.positions.count()
flat_frames = frame.positions.annotate(num_args=Count('arguments')).aggregate(Max('num_args'))['num_args__max']
if flat_frames > 1:
stats_dict[u'coor_schemata'] += 1
has_coordination = True
for pos in frame.positions.all():
stats_dict[u'phrases'] += pos.arguments.count()
if frame.phraseologic:
stats_dict[u'lex_schemata'] += 1
has_phraseology = True
if has_phraseology:
stats_dict[u'lex_lemmas'] += 1
if has_coordination:
stats_dict[u'coor_lemmas'] += 1
return stats_dict
def get_sub_entries_dict(lemma):
sub_entries_dict = {'sub_lemmas': 0}
for reflex in lemma.get_existing_frame_char_values(u'ZWROTNOŚĆ'):
for neg in lemma.get_existing_frame_char_values(u'NEGATYWNOŚĆ'):
for pred in lemma.get_existing_frame_char_values(u'PREDYKATYWNOŚĆ'):
for aspect in lemma.get_existing_frame_char_values(u'ASPEKT'):
matching_frames = lemma.get_frames_by_char_values(reflex_val=reflex,
neg_val=neg,
pred_val=pred,
aspect_val=aspect)
if matching_frames.exists():
sub_entries_dict[u'sub_lemmas'] += 1
subentry_key = u'Liczba podhaseł postaci: (%s,%s,%s,%s)' % (reflex.value, neg.value,
pred.value, aspect.value)
if not subentry_key in sub_entries_dict:
sub_entries_dict[subentry_key] = 0
sub_entries_dict[subentry_key] += 1
return sub_entries_dict
def write_stats(stats_path, stats):
try:
outfile = codecs.open(stats_path, 'wt', 'utf-8')
outfile.write(u'Łączna liczba haseł:\t%d\n\n' % stats['lemmas'])
outfile.write(u'Łączna liczba pozycji w schematach:\t%d\n' % stats['poss'])
outfile.write(u'Łączna liczba realizacji w schematach:\t%d\n\n' % stats['phrases'])
outfile.write(u'Łączna liczba podhaseł:\t%d\n' % stats['sub_lemmas'])
outfile.write(u'Liczba podhaseł postaci (ZWROTNOŚĆ, NEGATYWNOŚĆ, PREDYKATYWNOŚĆ, ASPEKT)\n')
write_subschemas_stats(outfile, stats)
outfile.write(u'Łączna liczba schematów:\t%d\n' % stats['schemata'])
outfile.write(u'Liczba schematów pewnych:\t%d\n' % stats['cer_schemata'])
outfile.write(u'Liczba schematów wątpliwych:\t%d\n' % stats['uncer_schemata'])
outfile.write(u'Liczba schematów złych:\t%d\n' % stats['bad_schemata'])
outfile.write(u'Liczba schematów archaicznych:\t%d\n' % stats['arch_schemata'])
outfile.write(u'Liczba schematów potocznych:\t%d\n' % stats['col_schemata'])
outfile.write(u'Liczba schematów wulgarnych:\t%d\n\n' % stats['vul_schemata'])
outfile.write(u'Łączna liczba schematów z koordynacją:\t%d\n' % stats['coor_schemata'])
outfile.write(u'Łączna liczba schematów zleksykalizowanych:\t%d\n\n' % stats['lex_schemata'])
outfile.write(u'Łączna liczba haseł zawierających pozycje z koordynacją:\t%d\n' % stats['coor_lemmas'])
outfile.write(u'Łączna liczba haseł zawierających schematy zleksykalizowane:\t%d\n\n' % stats['lex_lemmas'])
outfile.write(u'Łączna liczba ram semantycznych:\t%d\n\n' % stats['frames'])
finally:
outfile.close()
def write_subschemas_stats(stats_file, stats):
subschemas_stats = ['%s:\t%d\n' % (k, v) for k,v in stats.iteritems() if k.startswith(u'Liczba podhaseł postaci:')]
subschemas_stats.sort()
for stat in subschemas_stats:
stats_file.write(stat)
stats_file.write('\n')
def update_walenty_stats(stats):
WalentyStat.objects.all().delete()
WalentyStat(label=u'Łączna liczba haseł', value=str(stats['lemmas'])).save()
WalentyStat(label=u'Łączna liczba pozycji w schematach', value=str(stats['poss'])).save()
WalentyStat(label=u'Łączna liczba realizacji w schematach', value=str(stats['phrases'])).save()
WalentyStat(label=u'Łączna liczba schematów', value=str(stats['schemata'])).save()
WalentyStat(label=u'Łączna liczba schematów z koordynacją', value=str(stats['coor_schemata'])).save()
WalentyStat(label=u'Łączna liczba schematów zleksykalizowanych', value=str(stats['lex_schemata'])).save()
WalentyStat(label=u'Łączna liczba ram semantycznych', value=str(stats['frames'])).save()