sporada/nopaque
1
0
Fork 0
mirror of https://gitlab.ub.uni-bielefeld.de/sfb1288inf/nopaque.git synced 2025-06-12 17:10:41 +00:00
Code Issues Projects Releases Wiki Activity

push

Browse Source
This commit is contained in:
Patrick Jentsch
2019-11-27 13:29:43 +01:00
parent 5411bd91bd
commit 54f3036d89
2 changed files with 71 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files

47
app/corpora/events.py
View File

@ -47,15 +47,44 @@ def corpus_analysis(message):
room=request.sid)
return
""" Prepare and execute a query """
logger.warning('Payload: {}'.format(message))
corpus_name = 'CORPUS'
query = message['query']
client.select_corpus(corpus_name)
client.query_subcorpus(query)
results = client.show_query_results(
result_len=int(message['hits_per_page']),
context_len=int(message['context']))
socketio.emit('corpus_analysis', results, room=request.sid)
corpus = 'CORPUS'
query = (message['query'])
query_subcorpus = 'Results'
client.cqp_query(corpus, query_subcorpus, query)
data = {'matches': [], 'cpos_lookup': {}, 'text_loopup': {}}
""" Evaluate query results """
match_corpus = '{}:{}'.format(corpus, query_subcorpus)
match_num = min(int(message['hits_per_page']), client.cqp_subcorpus_size(match_corpus))
match_boundaries = zip(client.cqp_dump_subcorpus(match_corpus,
0x10,
0, match_num - 1),
client.cqp_dump_subcorpus(match_corpus,
0x11,
0, match_num - 1))
context = 15
corpus_len = 10000
for match_start, match_end in match_boundaries:
data['matches'].append({'lc': list(range(max(0, match_start - int(message['context'])), match_start)),
'hit': list(range(match_start, match_end + 1)),
'rc': list(range(match_end + 1, min(corpus_len, match_end + 1 + int(message['context']))))})
cpos_list = []
for match in data['matches']:
cpos_list += match['lc'] + match['hit'] + match['rc']
cpos_list = list(set(cpos_list))
pos_list = client.cl_cpos2str('{}.pos'.format(corpus), cpos_list)
simple_pos_list = client.cl_cpos2str('{}.simple_pos'.format(corpus), cpos_list)
text_id_list = client.cl_cpos2struc('{}.text_title'.format(corpus), cpos_list)
word_list = client.cl_cpos2str('{}.word'.format(corpus), cpos_list)
for cpos, pos, simple_pos, text_id, word in zip(cpos_list, pos_list, simple_pos_list, text_id_list, word_list):
data['cpos_lookup'][cpos] = {'pos': pos, 'simple_pos': simple_pos, 'text_id': text_id, 'word': word}
text_author_list = client.cl_struc2str('{}.text_author'.format(corpus), text_id_list)
text_publishing_year_list = client.cl_struc2str('{}.text_publishing_year'.format(corpus), text_id_list)
text_title_list = client.cl_struc2str('{}.text_title'.format(corpus), text_id_list)
for text_id, text_author, text_publishing_year, text_title in zip(text_id_list, text_author_list, text_publishing_year_list, text_title_list):
data['text_loopup'][text_id] = {'author': text_author, 'publishing_year': text_publishing_year, 'title': text_title}
socketio.emit('corpus_analysis', data, room=request.sid)
def corpus_analysis_session_handler(app, corpus_id, session_id):