[fix] pep8 : duckduckgo_definitions and wikidata engines

This commit is contained in:
dalf 2014-12-07 16:36:20 +01:00
parent 611f4e2a86
commit ffcec383b7
2 changed files with 168 additions and 90 deletions

View File

@ -3,21 +3,25 @@ from urllib import urlencode
from lxml import html from lxml import html
from searx.engines.xpath import extract_text from searx.engines.xpath import extract_text
url = 'https://api.duckduckgo.com/?{query}&format=json&pretty=0&no_redirect=1&d=1' url = 'https://api.duckduckgo.com/'\
+ '?{query}&format=json&pretty=0&no_redirect=1&d=1'
def result_to_text(url, text, htmlResult): def result_to_text(url, text, htmlResult):
# TODO : remove result ending with "Meaning" or "Category" # TODO : remove result ending with "Meaning" or "Category"
dom = html.fromstring(htmlResult) dom = html.fromstring(htmlResult)
a = dom.xpath('//a') a = dom.xpath('//a')
if len(a)>=1: if len(a) >= 1:
return extract_text(a[0]) return extract_text(a[0])
else: else:
return text return text
def html_to_text(htmlFragment): def html_to_text(htmlFragment):
dom = html.fromstring(htmlFragment) dom = html.fromstring(htmlFragment)
return extract_text(dom) return extract_text(dom)
def request(query, params): def request(query, params):
# TODO add kl={locale} # TODO add kl={locale}
params['url'] = url.format(query=urlencode({'q': query})) params['url'] = url.format(query=urlencode({'q': query}))
@ -38,16 +42,15 @@ def response(resp):
# add answer if there is one # add answer if there is one
answer = search_res.get('Answer', '') answer = search_res.get('Answer', '')
if answer != '': if answer != '':
results.append({ 'answer' : html_to_text(answer) }) results.append({'answer': html_to_text(answer)})
# add infobox # add infobox
if 'Definition' in search_res: if 'Definition' in search_res:
content = content + search_res.get('Definition', '') content = content + search_res.get('Definition', '')
if 'Abstract' in search_res: if 'Abstract' in search_res:
content = content + search_res.get('Abstract', '') content = content + search_res.get('Abstract', '')
# image # image
image = search_res.get('Image', '') image = search_res.get('Image', '')
image = None if image == '' else image image = None if image == '' else image
@ -55,29 +58,35 @@ def response(resp):
# attributes # attributes
if 'Infobox' in search_res: if 'Infobox' in search_res:
infobox = search_res.get('Infobox', None) infobox = search_res.get('Infobox', None)
if 'content' in infobox: if 'content' in infobox:
for info in infobox.get('content'): for info in infobox.get('content'):
attributes.append({'label': info.get('label'), 'value': info.get('value')}) attributes.append({'label': info.get('label'),
'value': info.get('value')})
# urls # urls
for ddg_result in search_res.get('Results', []): for ddg_result in search_res.get('Results', []):
if 'FirstURL' in ddg_result: if 'FirstURL' in ddg_result:
firstURL = ddg_result.get('FirstURL', '') firstURL = ddg_result.get('FirstURL', '')
text = ddg_result.get('Text', '') text = ddg_result.get('Text', '')
urls.append({'title':text, 'url':firstURL}) urls.append({'title': text, 'url': firstURL})
results.append({'title':heading, 'url': firstURL}) results.append({'title': heading, 'url': firstURL})
# related topics # related topics
for ddg_result in search_res.get('RelatedTopics', None): for ddg_result in search_res.get('RelatedTopics', None):
if 'FirstURL' in ddg_result: if 'FirstURL' in ddg_result:
suggestion = result_to_text(ddg_result.get('FirstURL', None), ddg_result.get('Text', None), ddg_result.get('Result', None)) suggestion = result_to_text(ddg_result.get('FirstURL', None),
ddg_result.get('Text', None),
ddg_result.get('Result', None))
if suggestion != heading: if suggestion != heading:
results.append({'suggestion': suggestion}) results.append({'suggestion': suggestion})
elif 'Topics' in ddg_result: elif 'Topics' in ddg_result:
suggestions = [] suggestions = []
relatedTopics.append({ 'name' : ddg_result.get('Name', ''), 'suggestions': suggestions }) relatedTopics.append({'name': ddg_result.get('Name', ''),
'suggestions': suggestions})
for topic_result in ddg_result.get('Topics', []): for topic_result in ddg_result.get('Topics', []):
suggestion = result_to_text(topic_result.get('FirstURL', None), topic_result.get('Text', None), topic_result.get('Result', None)) suggestion = result_to_text(topic_result.get('FirstURL', None),
topic_result.get('Text', None),
topic_result.get('Result', None))
if suggestion != heading: if suggestion != heading:
suggestions.append(suggestion) suggestions.append(suggestion)
@ -86,21 +95,26 @@ def response(resp):
if abstractURL != '': if abstractURL != '':
# add as result ? problem always in english # add as result ? problem always in english
infobox_id = abstractURL infobox_id = abstractURL
urls.append({'title': search_res.get('AbstractSource'), 'url': abstractURL}) urls.append({'title': search_res.get('AbstractSource'),
'url': abstractURL})
# definition # definition
definitionURL = search_res.get('DefinitionURL', '') definitionURL = search_res.get('DefinitionURL', '')
if definitionURL != '': if definitionURL != '':
# add as result ? as answer ? problem always in english # add as result ? as answer ? problem always in english
infobox_id = definitionURL infobox_id = definitionURL
urls.append({'title': search_res.get('DefinitionSource'), 'url': definitionURL}) urls.append({'title': search_res.get('DefinitionSource'),
'url': definitionURL})
# entity # entity
entity = search_res.get('Entity', None) entity = search_res.get('Entity', None)
# TODO continent / country / department / location / waterfall / mountain range : link to map search, get weather, near by locations # TODO continent / country / department / location / waterfall /
# mountain range :
# link to map search, get weather, near by locations
# TODO musician : link to music search # TODO musician : link to music search
# TODO concert tour : ?? # TODO concert tour : ??
# TODO film / actor / television / media franchise : links to IMDB / rottentomatoes (or scrap result) # TODO film / actor / television / media franchise :
# links to IMDB / rottentomatoes (or scrap result)
# TODO music : link tu musicbrainz / last.fm # TODO music : link tu musicbrainz / last.fm
# TODO book : ?? # TODO book : ??
# TODO artist / playwright : ?? # TODO artist / playwright : ??
@ -114,24 +128,25 @@ def response(resp):
# TODO programming language : ?? # TODO programming language : ??
# TODO file format : ?? # TODO file format : ??
if len(heading)>0: if len(heading) > 0:
# TODO get infobox.meta.value where .label='article_title' # TODO get infobox.meta.value where .label='article_title'
if image==None and len(attributes)==0 and len(urls)==1 and len(relatedTopics)==0 and len(content)==0: if image is None and len(attributes) == 0 and len(urls) == 1 and\
len(relatedTopics) == 0 and len(content) == 0:
results.append({ results.append({
'url': urls[0]['url'], 'url': urls[0]['url'],
'title': heading, 'title': heading,
'content': content 'content': content
}) })
else: else:
results.append({ results.append({
'infobox': heading, 'infobox': heading,
'id': infobox_id, 'id': infobox_id,
'entity': entity, 'entity': entity,
'content': content, 'content': content,
'img_src' : image, 'img_src': image,
'attributes': attributes, 'attributes': attributes,
'urls': urls, 'urls': urls,
'relatedTopics': relatedTopics 'relatedTopics': relatedTopics
}) })
return results return results

View File

@ -2,13 +2,25 @@ import json
from requests import get from requests import get
from urllib import urlencode from urllib import urlencode
resultCount=1 result_count = 1
urlSearch = 'https://www.wikidata.org/w/api.php?action=query&list=search&format=json&srnamespace=0&srprop=sectiontitle&{query}' wikidata_host = 'https://www.wikidata.org'
urlDetail = 'https://www.wikidata.org/w/api.php?action=wbgetentities&format=json&props=labels%7Cinfo%7Csitelinks%7Csitelinks%2Furls%7Cdescriptions%7Cclaims&{query}' wikidata_api = wikidata_host + '/w/api.php'
urlMap = 'https://www.openstreetmap.org/?lat={latitude}&lon={longitude}&zoom={zoom}&layers=M' url_search = wikidata_api \
+ '?action=query&list=search&format=json'\
+ '&srnamespace=0&srprop=sectiontitle&{query}'
url_detail = wikidata_api\
+ '?action=wbgetentities&format=json'\
+ '&props=labels%7Cinfo%7Csitelinks'\
+ '%7Csitelinks%2Furls%7Cdescriptions%7Cclaims'\
+ '&{query}'
url_map = 'https://www.openstreetmap.org/'\
+ '?lat={latitude}&lon={longitude}&zoom={zoom}&layers=M'
def request(query, params): def request(query, params):
params['url'] = urlSearch.format(query=urlencode({'srsearch': query, 'srlimit': resultCount})) params['url'] = url_search.format(
query=urlencode({'srsearch': query,
'srlimit': result_count}))
return params return params
@ -23,7 +35,8 @@ def response(resp):
language = resp.search_params['language'].split('_')[0] language = resp.search_params['language'].split('_')[0]
if language == 'all': if language == 'all':
language = 'en' language = 'en'
url = urlDetail.format(query=urlencode({'ids': '|'.join(wikidata_ids), 'languages': language + '|en'})) url = url_detail.format(query=urlencode({'ids': '|'.join(wikidata_ids),
'languages': language + '|en'}))
htmlresponse = get(url) htmlresponse = get(url)
jsonresponse = json.loads(htmlresponse.content) jsonresponse = json.loads(htmlresponse.content)
@ -32,6 +45,7 @@ def response(resp):
return results return results
def getDetail(jsonresponse, wikidata_id, language): def getDetail(jsonresponse, wikidata_id, language):
results = [] results = []
urls = [] urls = []
@ -40,60 +54,103 @@ def getDetail(jsonresponse, wikidata_id, language):
result = jsonresponse.get('entities', {}).get(wikidata_id, {}) result = jsonresponse.get('entities', {}).get(wikidata_id, {})
title = result.get('labels', {}).get(language, {}).get('value', None) title = result.get('labels', {}).get(language, {}).get('value', None)
if title == None: if title is None:
title = result.get('labels', {}).get('en', {}).get('value', None) title = result.get('labels', {}).get('en', {}).get('value', None)
if title == None: if title is None:
return results return results
description = result.get('descriptions', {}).get(language, {}).get('value', None) description = result\
if description == None: .get('descriptions', {})\
description = result.get('descriptions', {}).get('en', {}).get('value', '') .get(language, {})\
.get('value', None)
if description is None:
description = result\
.get('descriptions', {})\
.get('en', {})\
.get('value', '')
claims = result.get('claims', {}) claims = result.get('claims', {})
official_website = get_string(claims, 'P856', None) official_website = get_string(claims, 'P856', None)
if official_website != None: if official_website is not None:
urls.append({ 'title' : 'Official site', 'url': official_website }) urls.append({'title': 'Official site', 'url': official_website})
results.append({ 'title': title, 'url' : official_website }) results.append({'title': title, 'url': official_website})
wikipedia_link_count = 0 wikipedia_link_count = 0
if language != 'en': if language != 'en':
wikipedia_link_count += add_url(urls, 'Wikipedia (' + language + ')', get_wikilink(result, language + 'wiki')) wikipedia_link_count += add_url(urls,
'Wikipedia (' + language + ')',
get_wikilink(result, language +
'wiki'))
wikipedia_en_link = get_wikilink(result, 'enwiki') wikipedia_en_link = get_wikilink(result, 'enwiki')
wikipedia_link_count += add_url(urls, 'Wikipedia (en)', wikipedia_en_link) wikipedia_link_count += add_url(urls,
'Wikipedia (en)',
wikipedia_en_link)
if wikipedia_link_count == 0: if wikipedia_link_count == 0:
misc_language = get_wiki_firstlanguage(result, 'wiki') misc_language = get_wiki_firstlanguage(result, 'wiki')
if misc_language != None: if misc_language is not None:
add_url(urls, 'Wikipedia (' + misc_language + ')', get_wikilink(result, misc_language + 'wiki')) add_url(urls,
'Wikipedia (' + misc_language + ')',
get_wikilink(result, misc_language + 'wiki'))
if language != 'en': if language != 'en':
add_url(urls, 'Wiki voyage (' + language + ')', get_wikilink(result, language + 'wikivoyage')) add_url(urls,
add_url(urls, 'Wiki voyage (en)', get_wikilink(result, 'enwikivoyage')) 'Wiki voyage (' + language + ')',
get_wikilink(result, language + 'wikivoyage'))
add_url(urls,
'Wiki voyage (en)',
get_wikilink(result, 'enwikivoyage'))
if language != 'en': if language != 'en':
add_url(urls, 'Wikiquote (' + language + ')', get_wikilink(result, language + 'wikiquote')) add_url(urls,
add_url(urls, 'Wikiquote (en)', get_wikilink(result, 'enwikiquote')) 'Wikiquote (' + language + ')',
get_wikilink(result, language + 'wikiquote'))
add_url(urls, 'Commons wiki', get_wikilink(result, 'commonswiki')) add_url(urls,
'Wikiquote (en)',
get_wikilink(result, 'enwikiquote'))
add_url(urls, 'Location', get_geolink(claims, 'P625', None)) add_url(urls,
'Commons wiki',
get_wikilink(result, 'commonswiki'))
add_url(urls, 'Wikidata', 'https://www.wikidata.org/wiki/' + wikidata_id + '?uselang='+ language) add_url(urls,
'Location',
get_geolink(claims, 'P625', None))
add_url(urls,
'Wikidata',
'https://www.wikidata.org/wiki/'
+ wikidata_id + '?uselang=' + language)
musicbrainz_work_id = get_string(claims, 'P435') musicbrainz_work_id = get_string(claims, 'P435')
if musicbrainz_work_id != None: if musicbrainz_work_id is not None:
add_url(urls, 'MusicBrainz', 'http://musicbrainz.org/work/' + musicbrainz_work_id) add_url(urls,
'MusicBrainz',
'http://musicbrainz.org/work/'
+ musicbrainz_work_id)
musicbrainz_artist_id = get_string(claims, 'P434') musicbrainz_artist_id = get_string(claims, 'P434')
if musicbrainz_artist_id != None: if musicbrainz_artist_id is not None:
add_url(urls, 'MusicBrainz', 'http://musicbrainz.org/artist/' + musicbrainz_artist_id) add_url(urls,
'MusicBrainz',
'http://musicbrainz.org/artist/'
+ musicbrainz_artist_id)
musicbrainz_release_group_id = get_string(claims, 'P436') musicbrainz_release_group_id = get_string(claims, 'P436')
if musicbrainz_release_group_id != None: if musicbrainz_release_group_id is not None:
add_url(urls, 'MusicBrainz', 'http://musicbrainz.org/release-group/' + musicbrainz_release_group_id) add_url(urls,
'MusicBrainz',
'http://musicbrainz.org/release-group/'
+ musicbrainz_release_group_id)
musicbrainz_label_id = get_string(claims, 'P966') musicbrainz_label_id = get_string(claims, 'P966')
if musicbrainz_label_id != None: if musicbrainz_label_id is not None:
add_url(urls, 'MusicBrainz', 'http://musicbrainz.org/label/' + musicbrainz_label_id) add_url(urls,
'MusicBrainz',
'http://musicbrainz.org/label/'
+ musicbrainz_label_id)
# musicbrainz_area_id = get_string(claims, 'P982') # musicbrainz_area_id = get_string(claims, 'P982')
# P1407 MusicBrainz series ID # P1407 MusicBrainz series ID
@ -102,42 +159,43 @@ def getDetail(jsonresponse, wikidata_id, language):
# P1407 MusicBrainz series ID # P1407 MusicBrainz series ID
postal_code = get_string(claims, 'P281', None) postal_code = get_string(claims, 'P281', None)
if postal_code != None: if postal_code is not None:
attributes.append({'label' : 'Postal code(s)', 'value' : postal_code}) attributes.append({'label': 'Postal code(s)', 'value': postal_code})
date_of_birth = get_time(claims, 'P569', None) date_of_birth = get_time(claims, 'P569', None)
if date_of_birth != None: if date_of_birth is not None:
attributes.append({'label' : 'Date of birth', 'value' : date_of_birth}) attributes.append({'label': 'Date of birth', 'value': date_of_birth})
date_of_death = get_time(claims, 'P570', None) date_of_death = get_time(claims, 'P570', None)
if date_of_death != None: if date_of_death is not None:
attributes.append({'label' : 'Date of death', 'value' : date_of_death}) attributes.append({'label': 'Date of death', 'value': date_of_death})
if len(attributes)==0 and len(urls)==2 and len(description)==0: if len(attributes) == 0 and len(urls) == 2 and len(description) == 0:
results.append({ results.append({
'url': urls[0]['url'], 'url': urls[0]['url'],
'title': title, 'title': title,
'content': description 'content': description
}) })
else: else:
results.append({ results.append({
'infobox' : title, 'infobox': title,
'id' : wikipedia_en_link, 'id': wikipedia_en_link,
'content' : description, 'content': description,
'attributes' : attributes, 'attributes': attributes,
'urls' : urls 'urls': urls
}) })
return results return results
def add_url(urls, title, url): def add_url(urls, title, url):
if url != None: if url is not None:
urls.append({'title' : title, 'url' : url}) urls.append({'title': title, 'url': url})
return 1 return 1
else: else:
return 0 return 0
def get_mainsnak(claims, propertyName): def get_mainsnak(claims, propertyName):
propValue = claims.get(propertyName, {}) propValue = claims.get(propertyName, {})
if len(propValue) == 0: if len(propValue) == 0:
@ -157,7 +215,7 @@ def get_string(claims, propertyName, defaultValue=None):
mainsnak = e.get('mainsnak', {}) mainsnak = e.get('mainsnak', {})
datavalue = mainsnak.get('datavalue', {}) datavalue = mainsnak.get('datavalue', {})
if datavalue != None: if datavalue is not None:
result.append(datavalue.get('value', '')) result.append(datavalue.get('value', ''))
if len(result) == 0: if len(result) == 0:
@ -177,7 +235,7 @@ def get_time(claims, propertyName, defaultValue=None):
mainsnak = e.get('mainsnak', {}) mainsnak = e.get('mainsnak', {})
datavalue = mainsnak.get('datavalue', {}) datavalue = mainsnak.get('datavalue', {})
if datavalue != None: if datavalue is not None:
value = datavalue.get('value', '') value = datavalue.get('value', '')
result.append(value.get('time', '')) result.append(value.get('time', ''))
@ -190,7 +248,7 @@ def get_time(claims, propertyName, defaultValue=None):
def get_geolink(claims, propertyName, defaultValue=''): def get_geolink(claims, propertyName, defaultValue=''):
mainsnak = get_mainsnak(claims, propertyName) mainsnak = get_mainsnak(claims, propertyName)
if mainsnak == None: if mainsnak is None:
return defaultValue return defaultValue
datatype = mainsnak.get('datatype', '') datatype = mainsnak.get('datatype', '')
@ -209,21 +267,25 @@ def get_geolink(claims, propertyName, defaultValue=''):
# 1 --> 6 # 1 --> 6
# 0.016666666666667 --> 9 # 0.016666666666667 --> 9
# 0.00027777777777778 --> 19 # 0.00027777777777778 --> 19
# wolframalpha : quadratic fit { {13, 5}, {1, 6}, {0.0166666, 9}, {0.0002777777,19}} # wolframalpha :
# quadratic fit { {13, 5}, {1, 6}, {0.0166666, 9}, {0.0002777777,19}}
# 14.1186-8.8322 x+0.625447 x^2 # 14.1186-8.8322 x+0.625447 x^2
if precision < 0.0003: if precision < 0.0003:
zoom = 19 zoom = 19
else: else:
zoom = int(15 - precision*8.8322 + precision*precision*0.625447) zoom = int(15 - precision*8.8322 + precision*precision*0.625447)
url = urlMap.replace('{latitude}', str(value.get('latitude',0))).replace('{longitude}', str(value.get('longitude',0))).replace('{zoom}', str(zoom)) url = url_map\
.replace('{latitude}', str(value.get('latitude', 0)))\
.replace('{longitude}', str(value.get('longitude', 0)))\
.replace('{zoom}', str(zoom))
return url return url
def get_wikilink(result, wikiid): def get_wikilink(result, wikiid):
url = result.get('sitelinks', {}).get(wikiid, {}).get('url', None) url = result.get('sitelinks', {}).get(wikiid, {}).get('url', None)
if url == None: if url is None:
return url return url
elif url.startswith('http://'): elif url.startswith('http://'):
url = url.replace('http://', 'https://') url = url.replace('http://', 'https://')
@ -231,8 +293,9 @@ def get_wikilink(result, wikiid):
url = 'https:' + url url = 'https:' + url
return url return url
def get_wiki_firstlanguage(result, wikipatternid): def get_wiki_firstlanguage(result, wikipatternid):
for k in result.get('sitelinks', {}).keys(): for k in result.get('sitelinks', {}).keys():
if k.endswith(wikipatternid) and len(k)==(2+len(wikipatternid)): if k.endswith(wikipatternid) and len(k) == (2+len(wikipatternid)):
return k[0:2] return k[0:2]
return None return None