def topics(request):
# show results grouped by entity/topic, a result set for each one...
# first get top N entities for the text search
# for each entity do another sub-query
text_query=request.REQUEST.get('q')
if text_query is not None:
if len(text_query)==0:
text_query=None
entity_list=[]
results=Index.topics(text_query,page_size=5,num_topics=12)
return render(request,'topics.html',{'topics':results})
def topics(request):
# show results grouped by entity/topic, a result set for each one...
# first get top N entities for the text search
# for each entity do another sub-query
text_query = request.GET.get('q')
if text_query is not None:
if len(text_query) == 0:
text_query = None
entity_list = []
results = Index.topics(text_query, page_size=5, num_topics=12)
return render(request, 'topics.html', {'topics': results})
def index(request):
text_query=request.REQUEST.get('q')
if text_query is not None:
if len(text_query)==0:
text_query=None
print 'get entity list'
entity_list=get_entity_list(request)
if text_query is None and len(entity_list)==0:
text_query='*'
page_number=int(request.REQUEST.get('p',1))
page_size=20
facet_max=25
print 'get results'
results=Index.search(text_query,entity_list,page_number=page_number,facet_max=facet_max,highlight=True,highlight_inline=True,start_tag="<span style='background:yellow'>",end_tag='</span>')
print 'got results'
for result in results['results']:
if result.entities.filter(name='LeftWing').exists():
result.wing='left'
else:
if result.entities.filter(name='RightWing').exists():
result.wing='right'
else:
result.wing='none'
# adjust facet links to be relative to current path
facetpath=request.path
if not facetpath.endswith('/'):
facetpath=facetpath+'/'
facetquery=''
if text_query is not None and len(text_query)>0 and text_query!='*':
facetquery='?q='+text_query #TODO: encode query
title='"'+text_query+'"'
else:
if len(entity_list)>0:
title=entity_list[-1].name
else:
title='Search'
all_titles=''
print 'get word cloud'
for doc in results['results']:
all_titles=all_titles+' '+doc.title
all_words=nltk.word_tokenize(all_titles.lower())
# TODO: include patterns from entities, remove noise words
tagcloud=[{'text':word,'freq':all_words.count(word),'link':request.path+'?q='+word} for word in set(all_words) if not word in stopwords and word.isalpha()]
results['breadcrumbs']=get_breadcrumbs(entity_list,text_query)
results['facetpath']=facetpath
results['facetquery']=facetquery
results['title']=title
results['tagcloud']=tagcloud
print 'render'
return render(request,'index.html',results)
def index(request):
text_query = request.GET.get('q')
if text_query is not None:
if len(text_query) == 0:
text_query = None
print('get entity list')
entity_list = get_entity_list(request)
if text_query is None and len(entity_list) == 0:
text_query = '*'
page_number = int(request.GET.get('p', 1))
page_size = 20
facet_max = 25
print('get results')
results = Index.search(text_query,
entity_list,
page_number=page_number,
facet_max=facet_max,
highlight=True,
highlight_inline=True,
start_tag="<span style='background:yellow'>",
end_tag='</span>')
print('got results')
for result in results['results']:
if result.entities.filter(name='LeftWing').exists():
result.wing = 'left'
else:
if result.entities.filter(name='RightWing').exists():
result.wing = 'right'
else:
result.wing = 'none'
# adjust facet links to be relative to current path
facetpath = request.path
if not facetpath.endswith('/'):
facetpath = facetpath + '/'
facetquery = ''
if text_query is not None and len(text_query) > 0 and text_query != '*':
facetquery = '?q=' + text_query #TODO: encode query
title = '"' + text_query + '"'
else:
if len(entity_list) > 0:
title = entity_list[-1].name
else:
title = 'Search'
all_titles = ''
print('get word cloud')
for doc in results['results']:
all_titles = all_titles + ' ' + doc.title
all_words = nltk.word_tokenize(all_titles.lower())
# TODO: include patterns from entities, remove noise words
tagcloud = [{
'text': word,
'freq': all_words.count(word),
'link': request.path + '?q=' + word
} for word in set(all_words) if not word in stopwords and word.isalpha()]
results['breadcrumbs'] = get_breadcrumbs(entity_list, text_query)
results['facetpath'] = facetpath
results['facetquery'] = facetquery
results['title'] = title
results['tagcloud'] = tagcloud
print('render')
return render(request, 'index.html', results)
