[fix] urls merge in infobox (#593)

TODO:
    merge attributes
This commit is contained in:
marc 2016-06-24 00:38:17 -05:00
parent a0a1284998
commit c2e4014287
2 changed files with 25 additions and 21 deletions

View File

@ -99,9 +99,8 @@ def response(resp):
return [] return []
# link to wikipedia article # link to wikipedia article
# parenthesis are not quoted to make infobox mergeable with wikidata's
wikipedia_link = url_lang(resp.search_params['language']) \ wikipedia_link = url_lang(resp.search_params['language']) \
+ 'wiki/' + quote(title.replace(' ', '_').encode('utf8')).replace('%28', '(').replace('%29', ')') + 'wiki/' + quote(title.replace(' ', '_').encode('utf8'))
results.append({'url': wikipedia_link, 'title': title}) results.append({'url': wikipedia_link, 'title': title})

View File

@ -18,7 +18,17 @@ def result_content_len(content):
def compare_urls(url_a, url_b): def compare_urls(url_a, url_b):
if url_a.netloc != url_b.netloc or url_a.query != url_b.query: # ignore www. in comparison
if url_a.netloc.startswith('www.'):
host_a = url_a.netloc.replace('www.', '', 1)
else:
host_a = url_a.netloc
if url_b.netloc.startswith('www.'):
host_b = url_b.netloc.replace('www.', '', 1)
else:
host_b = url_b.netloc
if host_a != host_b or url_a.query != url_b.query:
return False return False
# remove / from the end of the url if required # remove / from the end of the url if required
@ -37,15 +47,17 @@ def merge_two_infoboxes(infobox1, infobox2):
urls1 = infobox1.get('urls', None) urls1 = infobox1.get('urls', None)
if urls1 is None: if urls1 is None:
urls1 = [] urls1 = []
infobox1['urls'] = urls1
urlSet = set() for url2 in infobox2.get('urls', []):
for url in infobox1.get('urls', []): unique_url = True
urlSet.add(url.get('url', None)) for url1 in infobox1.get('urls', []):
if compare_urls(urlparse(url1.get('url', '')), urlparse(url2.get('url', ''))):
unique_url = False
break
if unique_url:
urls1.append(url2)
for url in infobox2.get('urls', []): infobox1['urls'] = urls1
if url.get('url', None) not in urlSet:
urls1.append(url)
if 'img_src' in infobox2: if 'img_src' in infobox2:
img1 = infobox1.get('img_src', None) img1 = infobox1.get('img_src', None)
@ -97,7 +109,6 @@ class ResultContainer(object):
self.results = defaultdict(list) self.results = defaultdict(list)
self._merged_results = [] self._merged_results = []
self.infoboxes = [] self.infoboxes = []
self._infobox_ids = {}
self.suggestions = set() self.suggestions = set()
self.answers = set() self.answers = set()
self._number_of_results = [] self._number_of_results = []
@ -138,14 +149,13 @@ class ResultContainer(object):
add_infobox = True add_infobox = True
infobox_id = infobox.get('id', None) infobox_id = infobox.get('id', None)
if infobox_id is not None: if infobox_id is not None:
existingIndex = self._infobox_ids.get(infobox_id, None) for existingIndex in self.infoboxes:
if existingIndex is not None: if compare_urls(urlparse(existingIndex.get('id', '')), urlparse(infobox_id)):
merge_two_infoboxes(self.infoboxes[existingIndex], infobox) merge_two_infoboxes(existingIndex, infobox)
add_infobox = False add_infobox = False
if add_infobox: if add_infobox:
self.infoboxes.append(infobox) self.infoboxes.append(infobox)
self._infobox_ids[infobox_id] = len(self.infoboxes) - 1
def _merge_result(self, result, position): def _merge_result(self, result, position):
result['parsed_url'] = urlparse(result['url']) result['parsed_url'] = urlparse(result['url'])
@ -155,11 +165,6 @@ class ResultContainer(object):
result['parsed_url'] = result['parsed_url']._replace(scheme="http") result['parsed_url'] = result['parsed_url']._replace(scheme="http")
result['url'] = result['parsed_url'].geturl() result['url'] = result['parsed_url'].geturl()
result['host'] = result['parsed_url'].netloc
if result['host'].startswith('www.'):
result['host'] = result['host'].replace('www.', '', 1)
result['engines'] = [result['engine']] result['engines'] = [result['engine']]
# strip multiple spaces and cariage returns from content # strip multiple spaces and cariage returns from content