def select_search_url(self, entity):
search = entity.get_search_terms(entity.names[0])
results = wpapi.full_text_search(search)
extra = ""
if entity.type == 'politician':
md = entity.politician_metadata.all()[0]
extra += md.state + ", " + md.party
print
print '----------------------', search, "(%s)" % entity.type, extra
print
for i,result in enumerate(results):
so_wide = []
for j in range(0, len(result['content']), 80):
so_wide.append(result['content'][j:j + 80])
content = "\n ".join(so_wide)
print i, result['title']
print " ", wpapi.article_url(result['title'])
print " ", content
print
while True:
selection = raw_input(search + " ?> ")
if selection == 'n':
return None
try:
i = int(selection)
return wpapi.article_url(results[i]['title'])
except ValueError:
pass
if selection.startswith("http"):
return selection
def select_search_url(self, entity):
search = entity.get_search_terms(entity.names[0])
results = wpapi.full_text_search(search)
extra = ""
if entity.type == 'politician':
md = entity.politician_metadata.all()[0]
extra += md.state + ", " + md.party
print
print '----------------------', search, "(%s)" % entity.type, extra
print
for i, result in enumerate(results):
so_wide = []
for j in range(0, len(result['content']), 80):
so_wide.append(result['content'][j:j + 80])
content = "\n ".join(so_wide)
print i, result['title']
print " ", wpapi.article_url(result['title'])
print " ", content
print
while True:
selection = raw_input(search + " ?> ")
if selection == 'n':
return None
try:
i = int(selection)
return wpapi.article_url(results[i]['title'])
except ValueError:
pass
if selection.startswith("http"):
return selection
def find_wikipedia_url(entity):
"""
Returns a tuple of (url, article excerpt, image url) for a given entity, or None if no
matching article is found.
"""
empty_result = ['', '', '']
if entity.type in ['individual', 'industry']:
return empty_result
for ename in entity.names:
# Search for exact title matches with redirects. Use for comparing
# titles later -- we might exactly match a redirect title, but not
# match the destination page at all. Full text search returns only
# the destination pages, not the redirections.
redirects = wpapi.title_search_redirects(ename.search_string())
# Full text search!
results = wpapi.full_text_search(entity.get_search_terms(ename))
for result in results:
article = wpapi.WikipediaArticle(result['title'])
# Exclude special namespaced articles (e.g. User:, Template:,
# etc.)
if (article.namespace or article.title.startswith("List of")
or article.is_disambiguation_page()):
continue
if article.title in redirects:
name_comp = wpapi.WikipediaArticle(redirects.get(
article.title)).name
else:
name_comp = article.name
if entity.type == 'politician':
if ename != PersonName(name_comp):
continue
if not article.is_politician():
continue
if not article.is_american():
continue
subject = article.get_subject()
if subject and ename != PersonName(article.get_subject()):
continue
elif entity.type == 'organization':
if ename.is_politician():
if ename.pname != PersonName(name_comp):
continue
if not article.is_politician():
continue
if not article.is_american():
continue
else:
if ename != OrganizationName(name_comp):
continue
if article.is_person():
continue
if ename.is_company() and not article.is_company():
continue
wikipedia_url = wpapi.article_url(article.title)
wikipedia_excerpt, image_url = wpapi.get_article_excerpt_and_image_url(
article.title)
return (wikipedia_url, wikipedia_excerpt, image_url)
return empty_result
