def get(self):
wikipedia.set_lang(u'ru')
try:
page = wikipedia.page(u'Проект:Города_России/Списки_улиц/Казани')
streets = []
for link in page.links:
nlink = unicode(link).encode('utf-8').strip().decode('utf-8')
norm_name = normalize(nlink)
try:
street_info = StreetInfo.get_by_norm_name(norm_name)
if not street_info:
street_info = StreetInfo()
street_page = wikipedia.page(nlink)
street_info.name = nlink
street_info.norm_name = norm_name
street_info.info = unicode(street_page.summary).encode('utf-8').strip()
street_info.images = [Image(url=x) for x in street_page.images]
street_info.city = u'Казань'.encode('utf-8').strip()
street_info.lang=u'ru'.encode('utf-8').strip()
street_info.put()
except Exception, e:
print nlink.encode('utf-8')
except DeadlineExceededError:
pass
self.response.headers['Content-Type'] = "text/html; charset=utf-8"
self.response.write(json.dumps({'success':True}))
def test_change_through_multiple_langs(self):
# set language to Swedish
wikipedia.set_lang("sv")
self.assertEqual(wikipedia.API_URL,
'http://sv.wikipedia.org/w/api.php')
# switch language to German
wikipedia.set_lang("de")
self.assertEqual(wikipedia.API_URL,
'http://de.wikipedia.org/w/api.php')
# switch language to Russian
wikipedia.set_lang("ru")
self.assertEqual(wikipedia.API_URL,
'http://ru.wikipedia.org/w/api.php')
# switch language to Polish
wikipedia.set_lang("pl")
self.assertEqual(wikipedia.API_URL,
'http://pl.wikipedia.org/w/api.php')
# switch language to Portuguese
wikipedia.set_lang("pt")
self.assertEqual(wikipedia.API_URL,
'http://pt.wikipedia.org/w/api.php')
def searchIntent(self, session: DialogSession):
search = self._extractSearchWord(session)
if not search:
self._whatToSearch(session, 'whatToSearch')
return
wikipedia.set_lang(self.LanguageManager.activeLanguage)
try:
with Online():
result = wikipedia.summary(search, sentences=3)
except OfflineError:
self.endDialog(sessionId=session.sessionId, text=self.randomTalk('offline', skill='system'))
except wikipedia.DisambiguationError as e:
self.logWarning(msg=e)
self._whatToSearch(session, 'ambiguous')
except wikipedia.WikipediaException as e:
self.logWarning(msg=e)
self._whatToSearch(session, 'noMatch')
except Exception as e:
self.logWarning(msg=e, printStack=True)
else:
if not result:
self._whatToSearch(session, 'noMatch')
else:
self.endDialog(sessionId=session.sessionId, text=result)
def title(query: str) -> str:
wikipedia.set_lang('en')
search = wikipedia.search(query)[0]
return json.dumps(
filterResult(
wikipedia.page(
search['title']
)))
def test_zh(self):
wikipedia.set_lang('zh')
wikipedia.set_proxy({
'http': 'http://localhost:1080',
'https': 'https://localhost:1080'
})
wikipedia.set_request_lang('zh-CN,zh;q=0.9,en;q=0.8,da;q=0.7')
print(wikipedia.summary('流行性感冒'))
def import_images():
image_collection.remove(source='wiki')
wikipedia.set_lang('ru')
root_page = wikipedia.page('Экспонаты эрмитажа')
for link in root_page.links:
import_images_from_page(link)
def pageid(query: str) -> str:
print(query)
wikipedia.set_lang('en')
search = wikipedia.search(query)[0]
return json.dumps(
filterResult(
wikipedia.page(
None, search['pageid']
)))
def __init__(self, parent=None):
super(SoundThread, self).__init__(parent)
self.name = ""
home = expanduser("~")
self.filename = home + "/Curie.mp3"
wikipedia.set_lang("es")
self.player = QMediaPlayer(None, QMediaPlayer.StreamPlayback)
media = QMediaContent(QUrl.fromLocalFile(self.filename))
self.player.setMedia(media)
def test_set_lang_then_back_to_eng(self):
# set language to Spanish
wikipedia.set_lang("es")
self.assertEqual(wikipedia.API_URL,
'http://es.wikipedia.org/w/api.php')
# revert language back to English
wikipedia.set_lang("en")
self.assertEqual(wikipedia.API_URL,
'http://en.wikipedia.org/w/api.php')
def main():
choice = ''
known_language = ''
while choice.lower() not in ['y', 'yes', 'n', 'no']:
choice = raw_input('Default language - English. You wanna to change this? [y/n]: ')
if choice.lower() in ['yes', 'y']:
known_language = raw_input('You know your language abbreviation? [y/n]: ')
if known_language.lower() in ['no', 'n']:
lang_abbreviation = raw_input('Enter a native language naming: ')
find_lang(lang_abbreviation)
wikipedia.set_lang(raw_input('Enter your choice: '))
name = raw_input('Page for search: ')
MyPrettyPrinter().pprint(wikipedia.summary(name))
def searchIntent(self, session: DialogSession, **_kwargs):
search = self._extractSearchWord(session)
if not search:
self._whatToSearch(session, 'whatToSearch')
return
wikipedia.set_lang(self.LanguageManager.activeLanguage)
result = wikipedia.summary(search, sentences=3)
if not result:
self._whatToSearch(session, 'noMatch')
else:
self.endDialog(sessionId=session.sessionId, text=result)
def get(self):
wikipedia.set_lang(u"ru")
page = wikipedia.page(u"Проект:Города_России/Списки_улиц/Казани")
streets = []
for link in page.links:
nlink = unicode(link).encode("utf-8").strip()
try:
street_page = wikipedia.page(nlink)
streets.append(
{"name": nlink, "info": street_page.summary, "images": street_page.images, "city": u"Казань"}
)
except Exception, e:
print nlink
def get(self):
self.response.headers["Content-Type"] = "text/html; charset=utf-8"
wikipedia.set_lang(u"ru")
page = wikipedia.page(u"Проект:Города_России/Списки_улиц/Казани")
text = page.content
alphabet = re.split("\n\n\n==\s*...\s*==\n", text)
byline = []
for line in alphabet:
byline.append(re.split("\n", line))
byline.remove(byline[0])
# byline[0] - names with number
# byline[1..]- names with corresponding letter
for line in byline[0]:
self.response.write(json.dumps(line))
def update_on_priv_msg(self, data, connection):
if data['message'].find('.w ') == -1:
return
i18n_server = i18n()
w = wikipedia.set_lang(i18n_server.get_text('wiki_lang', lang=self.config.lang))
q = data['message'].split(' ')
query = ''
for word in q:
if word.strip() != '.w':
query += word + ' '
w = wikipedia.search(query)
if w.__len__() == 0: # TODO BUG BELOW, ERROR MESSAGE NOT SHOWN!
connection.send_back(data['nick'] + ', ' +
i18n_server.get_text('wiki_fail',
lang=self.config.lang),
data)
return
try:
page = wikipedia.WikipediaPage(w.pop(0))
except wikipedia.DisambiguationError as error:
print('disambiguation page')
page = wikipedia.WikipediaPage(error.args[1][0])
connection.send_back(data['nick'] + ' ' + page.url, data)
index = 51 + page.summary[50:350].rfind('. ')
if index == 50 or index > 230:
index = page.summary[0:350].rfind(' ')
connection.send_back(page.summary[0:index], data)
else:
connection.send_back(page.summary[0:index], data)
def update_on_priv_msg(self, data):
if data['message'].find('.w ') == -1:
return
i18n_server = i18n()
w = wikipedia.set_lang(i18n_server.get_text('wiki_lang'))
q = data['message'].split(' ')
query = ''
for word in q:
if word.strip() != '.w':
query += word + ' '
w = wikipedia.search(query)
if w.__len__() == 0:
Connection.singleton().send_back(data['nick'] + ', ' + i18n_server.get_text('wiki_fail'), data)
return
try:
page = wikipedia.WikipediaPage(w.pop(0))
except wikipedia.DisambiguationError as error:
print('disambiguation page')
page = wikipedia.WikipediaPage(error.args[1][0])
Connection.singleton().send_back(data['nick'] + ' ' + page.url, data)
index = 51 + page.summary[50:230].find('. ')
if index == -1 or index > 230:
Connection.singleton().send_back(page.summary[0:230], data)
else:
Connection.singleton().send_back(page.summary[0:index], data)
def extract_actor_from_wikipedia(lastname, firstname):
wikipedia.set_lang("fr")
searchs = wikipedia.search(lastname + " " + firstname)
for search in searchs:
page = wikipedia.page(search)
rc = {"links": list({"title": "wikipedia", "url": page.url})}
if lastname in page.title and firstname in page.title:
rc = dict({"links": [], "name": firstname + " " + lastname})
for img in page.images:
if img.endswith(".jpg"): rc["photo"] = img
save_domains = [
"unifrance.org", "www.lefilmfrancais", "www.allocine.fr",
"catalogue.bnf.fr", "www.allmovie.com"
]
libs = [
"UniFrance", "Le Film Francais", "Allocine", "La BNF",
"All movie"
]
try:
for ref in page.references:
domain = urlparse(ref).netloc
try:
idx = save_domains.index(domain)
rc["links"].append({"title": libs[idx], "url": ref})
except:
pass
except:
pass
html: wikipedia.BeautifulSoup = wikipedia.BeautifulSoup(
page.html(), "html5lib")
#Recherche de la section des films
# for link in html.findAll('a', attrs={'href': wikipedia.re.compile("^http://")}):
# if "film" in link.text:
# pass
rc["summary"] = page.summary
rc["title"] = page.title
rc["url"] = page.url
return rc
return None
def searchIntent(self, session: DialogSession):
search = self._extractSearchWord(session)
if not search:
self._whatToSearch(session, 'whatToSearch')
return
wikipedia.set_lang(self.LanguageManager.activeLanguage)
try:
result = wikipedia.summary(search, sentences=3)
except wikipedia.DisambiguationError as e:
self.logWarning(msg='Ambiguous result')
self._whatToSearch(session, 'ambiguous')
except wikipedia.WikipediaException as e:
self.logWarning(msg='No match')
self._whatToSearch(session, 'noMatch')
except Exception as e:
self.logWarning(msg=str(e), printStack=True)
else:
if not result:
self._whatToSearch(session, 'noMatch')
else:
self.endDialog(sessionId=session.sessionId, text=result)
def update_on_priv_msg(self, data):
if data['message'].find('.w ') == -1:
return
i18n_server = i18n()
w = wikipedia.set_lang(i18n_server.get_text('wiki_lang'))
q = data['message'].split(' ')
query = ''
for word in q:
if word.strip() != '.w':
query += word + ' '
w = wikipedia.search(query)
if w.__len__() == 0:
Connection.singleton().send_channel(data['nick'] + ', ' + i18n_server.get_text('wiki_fail'))
return
page = wikipedia.WikipediaPage(w.pop(0))
Connection.singleton().send_channel(data['nick'] + ' ' + page.url)
Connection.singleton().send_channel(page.summary[0:230])
def test_lang(self):
''' ensure the url gets updated correctly '''
wikipedia.set_lang("fr")
self.assertEqual(wikipedia.WIKIPEDIA_GLOBALS['API_URL'],
'http://fr.wikipedia.org/w/api.php')
def geo(lat: float, lon: float) -> str:
wikipedia.set_lang('en')
return json.dumps(
filterResult(
wikipedia.page(
wikipedia.geosearch(lat, lon))))
def searchIntent(self, intent: str, session: DialogSession) -> bool:
slots = session.slots
sessionId = session.sessionId
customData = session.customData
search = customData.get('userInput', slots.get('what'))
if not search:
self.continueDialog(sessionId=sessionId,
text=self.randomTalk('whatToSearch'),
intentFilter=[self._INTENT_USER_ANSWER],
previousIntent=self._INTENT_SEARCH,
customData={
'module': self.name,
})
return True
wikipedia.set_lang(self.LanguageManager.activeLanguage)
engine = customData.get('engine', 'wikipedia')
try:
if engine == 'wikipedia':
result = wikipedia.summary(search, sentences=3)
else:
result = wikipedia.summary(search, sentences=3)
if result:
self.endDialog(sessionId=sessionId, text=result)
else:
self.continueDialog(
sessionId=sessionId,
text=self.TalkManager.randomTalk('noMatch').format(search),
intentFilter=[self._INTENT_USER_ANSWER],
previousIntent=self._INTENT_SEARCH,
customData={
'module': self.name,
'engine': engine
})
except wikipedia.DisambiguationError:
self.continueDialog(
sessionId=sessionId,
text=self.TalkManager.randomTalk('ambiguous').format(search),
intentFilter=[self._INTENT_USER_ANSWER],
previousIntent=self._INTENT_SEARCH,
customData={
'module': self.name,
'engine': engine
})
except wikipedia.WikipediaException:
self.continueDialog(
sessionId=sessionId,
text=self.TalkManager.randomTalk('noMatch').format(search),
intentFilter=[self._INTENT_USER_ANSWER],
previousIntent=self._INTENT_SEARCH,
customData={
'module': self.name,
'engine': engine
})
except Exception as e:
self._logger.error(f'Error: {e}')
self.endDialog(sessionId=sessionId,
text=self.TalkManager.randomTalk('error',
module='system'))
return True
def main():
kalk_mode = False
aut_reg_mode = False
quest_mode = False
weather_mode = False
wiki_mode = False
quest_stage = 0
answer_quest_continue = False
wikipedia.set_lang('ru')
for event in longpoll.listen():
if event.type == VkEventType.MESSAGE_NEW:
if event.to_me:
request = event.text.lower()
if request in ['stop', 'стоп']: # остановить все режимы
kalk_mode = False
aut_reg_mode = False
quest_mode = False
weather_mode = False
wiki_mode = False
write_msg(event.user_id, 'Все режимы остановленны.', create_empty_keyboard())
elif request in ["что ты можешь", 'помощь', 'help']:
write_msg(event.user_id, "Вот, что я могу:", create_help_keyboard())
elif request in ['калькулятор', 'kalk'] and not wiki_mode: # Калькулятор
kalk_mode = not kalk_mode
if kalk_mode:
aut_reg_mode = False
quest_mode = False
weather_mode = False
wiki_mode = False
write_msg(event.user_id, 'Режим калькулятор запущен.', create_empty_keyboard())
write_msg(event.user_id,
"Введите выражение. Используйте знаки из допустимого списка: "
"+, -, *, /, **, (, ), %, //. Для остановки режима введите 'stop', 'стоп' "
"или слово, использованное для старта.",
create_empty_keyboard())
else:
write_msg(event.user_id, 'Режим калькулятор остановлен.', create_empty_keyboard())
elif request in ['регион', 'aut_reg', 'номер'] and not wiki_mode: # Регион регистрации автомобиля
aut_reg_mode = not aut_reg_mode
if aut_reg_mode:
kalk_mode = False
quest_mode = False
weather_mode = False
wiki_mode = False
write_msg(event.user_id,
"Режим поиска региона регистрации автомобиля запущен. "
"Для остановки режима введите 'stop', 'стоп' или слово, использованное для старта.",
create_empty_keyboard())
write_msg(event.user_id, "Введите номер региона. Исспользуйте только цифры.",
create_empty_keyboard())
else:
write_msg(event.user_id, 'Режим поиска региона регистрации автомобиля остановлен.',
create_empty_keyboard())
elif request in ['квест', 'quest', 'текстовый квест'] and not wiki_mode: # текстовый квест
quest_mode = not quest_mode
if quest_mode:
kalk_mode = False
aut_reg_mode = False
weather_mode = False
wiki_mode = False
write_msg(event.user_id, "Режим текстового квеста запущен. Для остановки режима введите "
"'stop', 'стоп' или слово, использованное для старта.",
create_empty_keyboard())
if quest_stage != 0:
write_msg(event.user_id, 'Хотите продолжить с момента, на котором остановились?',
create_answer_keyboard('Да', 'Нет'))
answer_quest_continue = True
continue
quest(event.user_id, quest_stage)
else:
write_msg(event.user_id, 'Режим текстового квеста остановлен.', create_empty_keyboard())
elif request in ['погода', 'weather'] and not wiki_mode: # погода
weather_mode = not weather_mode
if weather_mode:
kalk_mode = False
aut_reg_mode = False
quest_mode = False
wiki_mode = False
write_msg(event.user_id,
"Режим погода запущен. "
"Для остановки режима введите 'stop', 'стоп' или слово, использованное для старта.",
create_empty_keyboard())
write_msg(event.user_id, "Введите название города, погоду в котором вы хотите узнать.",
create_empty_keyboard())
else:
write_msg(event.user_id, 'Режим погода остановлен.', create_empty_keyboard())
elif request == 'wiki' or request == 'поиск': # поиск в wikipedia
wiki_mode = not wiki_mode
if wiki_mode:
kalk_mode = False
aut_reg_mode = False
quest_mode = False
weather_mode = False
write_msg(event.user_id,
"Режим wikipedia запущен. "
"Для остановки режима введите 'stop', 'стоп' или слово, использованное для старта.",
create_empty_keyboard())
write_msg(event.user_id, "Введите ваш запрос:",
create_empty_keyboard())
else:
write_msg(event.user_id, 'Режим wikipedia остановлен.', create_empty_keyboard())
elif kalk_mode:
kalk(event.user_id, request)
elif aut_reg_mode:
find_aut_reg(event.user_id, request)
elif quest_mode:
if answer_quest_continue is True:
answer_quest_continue = False
if request == 'да':
quest(event.user_id, quest_stage)
elif request == 'нет':
quest_stage = 0
quest(event.user_id, quest_stage)
elif request == 'пойти в дверь №1' and quest_stage == 0:
quest_stage = 1
quest(event.user_id, quest_stage)
elif request == 'пойти в дверь №2' and quest_stage == 0:
quest_stage = 2
quest(event.user_id, quest_stage)
elif request == 'выпить из пузырька' and quest_stage == 1:
quest_stage = 5
quest(event.user_id, quest_stage)
elif request == 'попробовать протиснуться в дверцу' and quest_stage == 1:
quest_stage = 6
quest(event.user_id, quest_stage)
elif request == 'нырнуть за ключом' and quest_stage == 2:
quest_stage = 7
quest(event.user_id, quest_stage)
elif request == 'попробовать открыть дверь' and quest_stage == 2:
quest_stage = 8
quest(event.user_id, quest_stage)
elif request == 'начать с начала':
quest_stage = 0
quest(event.user_id, quest_stage)
elif request == 'закончить':
quest_stage = 0
quest_mode = False
write_msg(event.user_id, 'Режим текстового квеста остановлен.', create_empty_keyboard())
elif weather_mode:
weather(event.user_id, request)
elif wiki_mode:
wiki_search(event.user_id, request)
# общение с пользователем
elif request == "привет" or request == 'hello':
write_msg(event.user_id, "Здравствуйте.", create_empty_keyboard())
elif request == "пока" or request == 'goodbye':
write_msg(event.user_id, "До свидания.", create_empty_keyboard())
else:
write_msg(event.user_id, "Не понимаю вашего ответа... Если нужна помощь введите 'help' "
"или 'помощь'.", create_quest_keyboard('Help'))
class WikipediaScraping:
"""
:Date: 2018-02-16
:Version: 1.2
:Author: Edwin Puertas - Pontificia Universidad Javeriana
:Copyright: Por definir
:Organization: Centro de Excelencia y Apropiación de Big Data y Data Analytics - CAOBA
This class extracts wikipedia content.
"""
wikipedia.set_lang('es')
def __init__(self):
"""
:Date: 2017-02-08
:Author: Edwin Puertas - Pontificia Universidad Javeriana
:Copyright: Por definir
:Organization: Centro de Excelencia y Apropiación de Big Data y Data Analytics - CAOBA
This function return inicialized wikipedia object article
:return: Wikipedia object
"""
self.support = Support()
self.media_wiki = wikipediaapi.Wikipedia('es')
print('WikipediaScraping')
def search_articles(self, list_word):
art_list = []
try:
for word in list_word:
art_list += wikipedia.search(word)
art_list = self.support.normalized_list(art_list)
return art_list
except:
Logging.write_standard_error(sys.exc_info())
def get_content(self, word):
try:
wiki = wikipediaapi.Wikipedia(
language='es', extract_format=wikipediaapi.ExtractFormat.WIKI)
article = wiki.page(word)
list_tmp = [
'Historia', 'Véase también', 'Referencias', 'Enlaces externos'
]
text = ''
list_subtitle = []
if article.exists():
text = article.summary
for s in article.sections:
if not (s.title in list_tmp):
text += s.title + '\n' + s.text
list_subtitle.append(s.title)
text = self.support.clean_text(text)
return list_subtitle, text
except:
Logging.write_standard_error(sys.exc_info())
def get_article(self, word):
"""
:Date: 2017-05-09
:Version: 1.2
:Author: Edwin Puertas - Pontificia Universidad Javeriana
:Organization: Centro de Excelencia y Apropiación de Big Data y Data Analytics - CAOBA
This function return wikipedia object article with sets values
:param word: Name of the wikipedia article to extract
:type word: Text
:rtype: Dictionary
:return: Dictionary of synonyms
"""
art = {}
result = None
try:
article = self.media_wiki.page(word)
if article.exists():
art['Id'] = article.pageid # Id
art['Title'] = article.title #Title
art['SubTitle'] = self.get_content(art['Title'])[0] #SubTitle
art['Summary'] = article.summary #Summary
art['Content'] = self.get_content(art['Title'])[1] #Content
art['Links'] = self.normalized_links(article.links) #Links
art['Categories'] = self.get_categories(
art['Title']) #Categories
art['URL'] = article.canonicalurl #URL
art['Weight'] = len(art['Links']) # Peso del Artículos
art['Frequency'] = Counter(tb(article.summary +
art['Content']))
result = art
return result
except:
Logging.write_standard_error(sys.exc_info())
def get_nearby_articles(self, list_articles):
"""
:Date: 2017-05-09
:Version: 1.2
:Author: Edwin Puertas - Pontificia Universidad Javeriana
:Organization: Centro de Excelencia y Apropiación de Big Data y Data Analytics - CAOBA
This function return list of wikipedia article nearby to seed words
:param list_articles: List of the wikipedia article
:type list_articles: List
:rtype: Dictionary
:return: Dictionary of nearby articles
"""
try:
dic_temp = {}
for art in list_articles:
key = self.media_wiki.page(art)
if key.exists():
dic_temp[str(key.title).lower()] = self.get_categories(
key.title)
return dic_temp
except:
Logging.write_standard_error(sys.exc_info())
def get_categories(self, word):
"""
:Date: 2017-05-09
:Version: 1.2
:Author: Edwin Puertas - Pontificia Universidad Javeriana
:Organization: Centro de Excelencia y Apropiación de Big Data y Data Analytics - CAOBA
This function return list of wikipedia article nearby to seed words
:param word: Name of the wikipedia article to extract
:type word: Text
:rtype: List
:return: List of categories by article
"""
try:
list_categories = self.media_wiki.page(word).categories
new_list = []
for item in list_categories:
item = str(item).lower()
item = self.support.clean_text(item)
if not ('wiki' in item) and item != '':
item = item[10:len(item)]
item = item.strip()
new_list.append(item)
new_list = self.support.normalized_list(new_list)
return new_list
except:
Logging.write_standard_error(sys.exc_info())
def normalized_links(self, list_links):
"""
:Date: 2017-05-09
:Version: 1.2
:Author: Edwin Puertas - Pontificia Universidad Javeriana
:Organization: Centro de Excelencia y Apropiación de Big Data y Data Analytics - CAOBA
This function return list of wikipedia article nearby to seed words
:param list_links: list of word
:type list: Text
:rtype: List
:return: List any repeats words
"""
list_tmp = []
try:
for item in list_links:
text = self.support.clean_text(item).strip()
if text != '':
list_tmp.append(text)
list_tmp = self.support.normalized_list(list_tmp)
return list_tmp
except:
Logging.write_standard_error(sys.exc_info())
def compare_categories(self, main_categories, second_categories):
"""
:Date: 2017-05-09
:Version: 1.2
:Author: Edwin Puertas - Pontificia Universidad Javeriana
:Organization: Centro de Excelencia y Apropiación de Big Data y Data Analytics - CAOBA
This function return True or False if the words in main_categories exist in second_categories.
:param main_categories: list of words by main categories
:type list: Text
:param second_categories: list of words by second categories
:type list: Text
:rtype: Boolean
:return: True or False
"""
try:
val = False
list_tmp = []
for i in main_categories:
if (i not in list_tmp) and (i in second_categories):
list_tmp.append(i)
if len(list_tmp) > 0:
val = True
return val
except:
Logging.write_standard_error(sys.exc_info())
def creted_corpus_wikipedia(self, seed_words):
"""
:Date: 2017-05-09
:Version: 1.2
:Author: Edwin Puertas - Pontificia Universidad Javeriana
:Organization: Centro de Excelencia y Apropiación de Big Data y Data Analytics - CAOBA
This function create a corpus by articles in Wikipedia by seed words.
:param seed_words: list of seed words
:type list: Text
:rtype: XML
:return: Corpus of Wikipedia Articles
"""
parent_categories = []
list_article_invalid = []
try:
print('Searching Wikipedia with the seed {0}'.format(
str(seed_words)))
print('Searching Articles in Wikipedia, wait a moment please.')
list_nearby_parent_articles = self.search_articles(seed_words)
print('Found articles:{0}'.format(list_nearby_parent_articles))
print('Building main domain region, wait a moment please.')
#Saved articles seed words
list_articles = []
list_titles = []
for word in seed_words:
art = self.get_article(word)
if art is not None:
list_articles.append({
'resource': 'Wikipedia',
'title': art['Title'],
'subtitle': art['SubTitle'],
'content': art['Content']
})
list_titles.append(art['Title'])
parent_categories += self.get_categories(str(word))
parent_categories = self.support.normalized_list(parent_categories)
print('Main domain region {0}'.format(parent_categories))
for art in list_nearby_parent_articles:
child = self.get_article(art)
child_categories = child['Categories']
if self.compare_categories(parent_categories,
child_categories):
if (art is not None) and not (child['Title']
in list_titles):
print(
'Retrieving Information from [{0}], [{1}]'.format(
child['Title'], child['URL']))
list_articles.append({
'resource': 'Wikipedia',
'title': child['Title'],
'subtitle': child['SubTitle'],
'content': child['Content']
})
list_titles.append(child['Title'])
dict_grand_child = {}
print('Validating articles children of {0}'.format(
str(child['Title'])))
print(
'Validating domain region for articles {0}, wait a moment please.'
.format(len(child['Links'])))
dict_grand_child = self.get_nearby_articles(
child['Links'])
for k, v in dict_grand_child.items():
grand_child_categories = v
if self.compare_categories(parent_categories,
grand_child_categories):
grand_child = self.get_article(k)
if grand_child is not None:
if not (grand_child['Title']
in list_titles):
print(
'Retrieving Information from [{0}], [{1}]'
.format(grand_child['Title'],
grand_child['URL']))
list_articles.append({
'resource':
'Wikipedia',
'title':
grand_child['Title'],
'subtitle':
grand_child['SubTitle'],
'content':
grand_child['Content']
})
list_titles.append(
grand_child['Title'])
else:
print('Article [{0}] excluded!'.format(
grand_child['Title']))
if grand_child['Title'] != '':
list_article_invalid.append(
grand_child['Title'])
excluded_items = len(list_article_invalid)
recovered_items = len(list_articles)
total_items = excluded_items + recovered_items
print(
'\n# Article excluded: [{0}] \n# Articles recovered: [{1}] \nTotal Articles: [{2}]'
.format(excluded_items, recovered_items, total_items))
return list_articles
except:
Logging.write_standard_error(sys.exc_info())
def test_lang(self):
''' ensure the url gets updated correctly '''
wikipedia.set_lang("fr")
self.assertEqual(wikipedia.WIKIPEDIA_GLOBALS['API_URL'], 'http://fr.wikipedia.org/w/api.php')
def test_lang(self):
wikipedia.set_lang("fr")
self.assertEqual(wikipedia.API_URL, 'https://fr.wikipedia.org/w/api.php')
import vk_api
from vk_api.bot_longpoll import VkBotLongPoll, VkBotEventType
import random
from wikipedia import wikipedia
vk_session = vk_api.VkApi(token=TOKEN)
longpoll = VkBotLongPoll(vk_session, GROUP_ID)
vk = vk_session.get_api()
wikipedia.set_lang('ru')
def wiki_response(request_text):
return str(wikipedia.page(request_text).content[:1000])
def help():
return f"What do you want to ask Wikipedia?"
def main():
flag_wiki, flag_help = False, True
for event in longpoll.listen():
if event.type == VkBotEventType.MESSAGE_NEW and flag_help:
flag_wiki = not flag_wiki
flag_help = not flag_help
vk.messages.send(user_id=event.obj.message['from_id'],
message=help(),
random_id=random.randint(0, 2**64))
def test_lang(self):
wikipedia.set_lang("fr")
self.assertEqual(wikipedia.api_url, 'http://fr.wikipedia.org/w/api.php')
def test_set_lang_it(self):
wikipedia.set_lang("zh")
self.assertEqual(wikipedia.API_URL,
'http://zh.wikipedia.org/w/api.php')
