def normal_item_solver(self, item, task, response):
doc = self.get_doc(response)
title = util.get_filtered_title(doc, {'h1 a'})
t = util.get_time_string_from_selectors(doc, {'span.postdate'})
t_stamp = util.get_timestamp_from_string(t) + int(
time.localtime().tm_sec)
category = doc('span.postcat a').text()
author = doc('span.postauthor a').text()
content = util.get_paragraphs_from_selector(doc, 'div p')
content = re.sub(ur'投稿:[.\n\r\t]*.*', u'', content, re.M | re.U | re.I)
content = re.sub(ur'則留言[.\n\r\t]*', u'', content, re.M | re.U | re.I)
content = re.sub(ur'大道之行也，天下為公，選賢與能，講信修睦。－－－《禮運．大同》[.\n\r\t]*', u'',
content, re.M | re.U | re.I)
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'VJMedia'
item.task_no = self.BATCH_NUMBER
for img in doc(
'#container img.size-full, #container img.size-large').items():
if img.attr('src') != '':
des = ''
if img.attr('alt'):
des = img.attr('alt')
elif img.siblings('p'):
des = img.siblings('p').text()
media = self.NewsItem.MediaItem(media_url=img.attr('src'),
type='image',
description=des,
created_at=item.fetched_at)
item.media_list.append(media)
for iframe in doc('iframe').items():
if iframe.attr('src') and re.match(r'.*youtube\.com.+',
iframe.attr('src')):
media = self.NewsItem.MediaItem(media_url=iframe.attr('src'),
type='youtube',
description='youtube',
created_at=item.fetched_at)
item.media_list.append(media)
if util.within_active_interval(6, 1200):
_comments = util.get_filtered_facebook_comments_data(
'214585295294555',
doc('div.fb-comments').attr('data-href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(
media_url=_comment['json_string'],
type='comments',
description='comments',
created_at=item.fetched_at))
def normal_item_solver(self, item, task, response):
doc = self.get_doc(response)
title = util.get_filtered_title(doc, {'title'}, ur' - 香港新浪')
t = doc('div.news-datetime').text()
t_stamp = int(
time.mktime(time.strptime(t.encode('utf-8'), '%Y年%m月%d日 %H:%M')))
scripts = doc('script').text()
category = ''
if re.findall(cat_pattern, scripts):
category = re.findall(cat_pattern, scripts)[0]
author = ''
content = util.get_paragraphs_from_selector(doc, 'div.news-body p')
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'Sina'
item.task_no = self.BATCH_NUMBER
for img in doc('.news-body img').items():
if img.attr('src') != '':
media_u = img.attr('src')
if not ad_url_pattern.findall(media_u):
des = ''
if img.parent().attr('data-caption'):
des = img.parent().attr('data-caption')
media = self.NewsItem.MediaItem(media_url=media_u,
type='image',
description=des,
created_at=item.fetched_at)
item.media_list.append(media)
for a in doc('iframe').items():
if a.attr('src') and re.match(r'.*youtube\.com.+', a.attr('src')):
media_u = a.attr('src')
if re.match(r'//.+', media_u):
media_u = 'http:' + media_u
media = self.NewsItem.MediaItem(media_url=media_u,
type='youtube',
description='youtube',
created_at=item.fetched_at)
item.media_list.append(media)
if util.within_active_interval(6, 1200):
_comments = util.get_filtered_facebook_comments_data(
'114907575364430',
doc('div.fb-comments').attr('data-href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(
media_url=_comment['json_string'],
type='comments',
description='comments',
created_at=item.fetched_at))
def normal_item_solver(self, item, task, response):
doc = self.get_doc(response)
title = util.get_filtered_title(doc, {'title'}, ur'\s*\|.*')
t = doc('div.article-content-wrap p.date').text()
t_stamp = int(
time.mktime(time.strptime(t.encode('utf-8'), '%Y/%m/%d — %H:%M')))
cat_href = cat_pattern.findall(task.url)[0]
category = doc('ul[id=mainMenuUL] a[href="' + cat_href + '"]').text()
author = ''
content = util.get_paragraphs_from_selector(doc.remove('style'),
'div.article-content p')
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'StandNews'
item.task_no = self.BATCH_NUMBER
for img in doc('.article-content-wrap img').items():
if img.attr('src') != '' and not ad_image_pattern.match(
img.attr('src')):
des = ''
if img.attr('alt'):
des = img.attr('alt')
media = self.NewsItem.MediaItem(media_url=img.attr('src'),
type='image',
description=des,
created_at=item.fetched_at)
item.media_list.append(media)
for iframe in doc('.article-content-wrap iframe').items():
if iframe.attr('src') and youtube_pattern.match(
iframe.attr('src')):
media = self.NewsItem.MediaItem(media_url=iframe.attr('src'),
type='youtube',
description='youtube',
created_at=item.fetched_at)
item.media_list.append(media)
if util.within_active_interval(6, 1200):
_comments = util.get_filtered_facebook_comments_data(
'1534089350179685',
doc('div.fb-comments').attr('data-href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(
media_url=_comment['json_string'],
type='comments',
description='comments',
created_at=item.fetched_at))
def normal_item_solver(self, item, task, response):
doc = self.get_doc(response)
title = util.get_filtered_title(doc, {'title'}, ur'熱血時報 \| ')
author = doc('span.author a').text()
t = doc('time[class="published"]').text()
t_stamp = int(time.mktime(time.strptime(str(doc('time[class="published"]').text()), "%m-%d-%Y")))
category = doc('div.page-path a').text()
doc.remove('script')
doc.remove('style')
content = util.get_paragraphs_from_selector(doc, 'div.article-body p')
if content == '':
content = util.get_paragraphs_from_selector(doc, 'div.article-body')
if t_stamp >= int(time.mktime(time.strptime(str(time.localtime().tm_year) + str(time.localtime().tm_mon) + str(
time.localtime().tm_mday), '%Y%m%d'))):
t_stamp = int(time.time())
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'PassionTimes'
item.task_no = self.BATCH_NUMBER
for img in doc('.article-body img').items():
if img.attr('src') != '':
media_u = img.attr('src')
if re.match(r'//.+', media_u):
media_u = 'http:' + media_u
elif not re.match(r'http://.+', media_u):
media_u = 'http://www.passiontimes.hk' + media_u
media = self.NewsItem.MediaItem(media_url=media_u, type='image', description='',
created_at=item.fetched_at)
item.media_list.append(media)
for a in doc('iframe').items():
if a.attr('src') and re.match(r'.*youtube\.com.+', a.attr('src')):
media_u = a.attr('src')
if re.match(r'//.+', media_u):
media_u = 'http:' + media_u
media = self.NewsItem.MediaItem(media_url=media_u, type='youtube', description='youtube',
created_at=item.fetched_at)
item.media_list.append(media)
if util.within_active_interval(6, 1200):
_comments = util.get_filtered_facebook_comments_data('462543587117177',
doc('div.fb-comments').attr('data-href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(media_url=_comment['json_string'], type='comments',
description='comments', created_at=item.fetched_at)
)
def normal_item_solver(self, item, task, response):
doc = self.get_doc(response)
item.raw = doc.text()
item.title = util.get_filtered_title(doc, {'.article_tit h1'}, ur'\s*｜.*')
if item.title == '':
item.title = util.get_filtered_title(doc, {'title'}, ur'\s*｜.*')
item.t = doc('meta[name=artpdate]').attr('content')
item.t_stamp = int(time.mktime(time.strptime(item.t, "%Y-%m-%d %H:%M:%S")))
item.fetched_at = task.fetched_at
item.category = doc('meta[name=catname]').attr('content')
item.author = doc('meta[name=author]').attr('content')
content = util.get_paragraphs_from_selector(doc, 'div.article_content__module p')
if content == '':
content = util.get_paragraphs_from_selector(doc, 'li.article_summary_pt')
item.content = content
item.url = task.url
item.source = 'HK01'
item.task_no = self.BATCH_NUMBER
for img in doc('.article__body__content img').items():
if img.attr('src') != '' and not img.parents('.related_article'):
media_u = img.attr('src')
if media_u != '//cdn.hk01.com/media/dummy/default_image.png':
des = ''
if img.attr('alt'):
des = img.attr('alt')
media = self.NewsItem.MediaItem(media_url=media_u, type='image', description=des,
created_at=item.fetched_at)
item.media_list.append(media)
for img in doc('.article__body__content object[data-gallery-image="true"]').items():
if img.attr('data'):
media_u = img.attr('data')
des = ''
if img.attr('alt'):
des = img.attr('alt')
media = self.NewsItem.MediaItem(media_url=media_u, type='image', description=des,
created_at=item.fetched_at)
item.media_list.append(media)
for a in doc('iframe').items():
if a.attr('src') and re.match(r'.*youtube\.com.+', a.attr('src')):
media_u = a.attr('src')
if re.match(r'//.+', media_u):
media_u = 'http:' + media_u
media = self.NewsItem.MediaItem(media_url=media_u, type='youtube', description='youtube',
created_at=item.fetched_at)
item.media_list.append(media)
if util.within_active_interval(6, 1200):
_comments = util.get_filtered_facebook_comments_data('1651866545051541', doc('div.fb-comments').attr('data-href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(media_url=_comment['json_string'], type='comments', description='comments', created_at=item.fetched_at)
)
'''
def normal_item_solver(self, item, task, response):
doc = self.get_doc(response)
title = util.get_filtered_title(doc, {'title'}, ur' - .*')
t = ''
t_stamp = 0
if doc('time') or doc('span.date'):
t = util.get_time_string_from_selectors(doc, {'time', 'span.date'})
t_stamp = util.get_timestamp_from_string(t) + int(time.localtime().tm_sec)
category = re.sub(ur'.*\s+', u'', doc('.dropdown-menu li.active a').text())
author = ''
content = util.get_paragraphs_from_selector(doc, '#news-content')
if content == '':
content = util.get_paragraphs_from_selector(doc, '.content span')
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'HeadlineNews'
item.task_no = self.BATCH_NUMBER
for img in doc('.content .item img').items():
if img.attr('src') != '':
media_u = img.attr('src')
if re.match(r'//.+', media_u):
media_u = 'http:' + media_u
des = ''
if img.attr('alt'):
des = img.attr('alt')
media = self.NewsItem.MediaItem(media_url=media_u, type='image', description=des,
created_at=item.fetched_at)
item.media_list.append(media)
for a in doc('iframe').items():
if a.attr('src') and re.match(r'.*youtube\.com.+', a.attr('src')):
media_u = a.attr('src')
if re.match(r'//.+', media_u):
media_u = 'http:' + media_u
media = self.NewsItem.MediaItem(media_url=media_u, type='youtube', description='youtube',
created_at=item.fetched_at)
item.media_list.append(media)
if util.within_active_interval(12, 600):
_comments = util.get_filtered_facebook_comments_data('978368502211772',
doc('div.fb-comments').attr('data-href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(media_url=_comment['json_string'], type='comments',
description='comments', created_at=item.fetched_at)
)
def normal_item_solver(self, item, task, response):
doc = self.get_doc(response)
title = util.get_filtered_title(doc, {'h1'})
t = doc('time.published').text()
t_stamp = util.get_timestamp_from_string(
doc('time.published').attr('datetime'), '%Y-%m-%d %H:%M:%S+0800')
category = doc('#navBar li.active').text()
if category == '' and cat_pattern_in_url.findall(task.url):
category = cat_pattern_in_url.findall(task.url)[0]
author = ''
content = util.get_paragraphs_from_selector(doc, 'div.newsLeading p')
if content == '':
content = util.get_paragraphs_from_selector(doc, 'div.newsLeading')
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'NowNews'
item.task_no = self.BATCH_NUMBER
if util.within_active_interval(6, 1200):
_comments = util.get_filtered_facebook_comments_data(
'515076798590105',
doc('div.fb-comments').attr('data-href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(
media_url=_comment['json_string'],
type='comments',
description='comments',
created_at=item.fetched_at))
def normal_item_solver(self, item, task, response):
doc = self.get_doc(response)
title = util.get_filtered_title(doc, {'title'})
t = util.get_time_string_from_selectors(doc, {'div.video_date span'},
date_patterns={date_pattern})
t_stamp = util.get_timestamp_from_string(t) + int(
time.localtime().tm_sec)
category = doc('span.heretxt').text()
author = ''
content = util.get_paragraphs_from_selector(doc, 'div.video_content')
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'CableNews'
item.task_no = self.BATCH_NUMBER
if util.within_active_interval(6, 1200):
_comments = util.get_filtered_facebook_comments_data(
'482092838576644',
doc('div.fb-comments').attr('data-href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(
media_url=_comment['json_string'],
type='comments',
description='comments',
created_at=item.fetched_at))
def normal_item_solver(self, item, task, response):
response.encoding = 'utf-8'
doc = self.get_doc(response)
title = util.get_filtered_title(doc, {'title'}, u' \|.*')
t = util.get_day_string(offset=self.OFFSET)
t_stamp = util.get_timestamp_from_string(
t) + time.localtime().tm_hour * 3600 + time.localtime().tm_min * 60
category = doc('meta[name=subsection]').attr('content')
author = ''
content = util.get_paragraphs_from_selector(doc, '#masterContent p')
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'AppleNews'
item.task_no = self.BATCH_NUMBER
if util.within_active_interval(6, 20 * 60):
_comments = util.get_filtered_facebook_comments_data(
'367495573302576',
doc('meta[property="og:url"]').attr('content'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(
media_url=_comment['json_string'],
type='comments',
description='comments',
created_at=item.fetched_at))
def normal_item_solver(self, item, task, response):
doc = self.get_doc(response)
title = util.get_filtered_title(doc, {'title'}, ur'\s*\|.*')
t = util.get_time_string_from_selectors(
doc, {'div.PolDTextBox_Date', 'div.MainNews_Date'})
t = t[-4:] + t[3:5] + t[:2]
t_stamp = util.get_timestamp_from_string(
t) + time.localtime().tm_hour * 3600 + time.localtime(
).tm_min * 60 + time.localtime().tm_sec
category = ''
if cat_pattern.findall(task.url):
category = cat_pattern.findall(task.url)[0]
category = urllib2.unquote(category)
author = ''
content = util.get_paragraphs_from_selector(doc, 'div.PolDTextBox div')
if content == '':
content = util.get_paragraphs_from_selector(
doc, 'div.MainNews_Text div')
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'SkyPost'
item.task_no = self.BATCH_NUMBER
for img in doc('.PolDetailBox img').items():
if img.attr('src') != '':
media_u = img.attr('src')
des = ''
if img.parent().siblings('.PolDCaption'):
des = img.parent().siblings('.PolDCaption').text()
elif img.parents('.NewsDPicBox02'):
des = img.parents('.NewsDPicBox02').text()
media = self.NewsItem.MediaItem(media_url=media_u,
type='image',
description=des,
created_at=item.fetched_at)
item.media_list.append(media)
for a in doc('iframe').items():
if a.attr('src') and re.match(r'.*youtube\.com.+', a.attr('src')):
media_u = a.attr('src')
if re.match(r'//.+', media_u):
media_u = 'http:' + media_u
media = self.NewsItem.MediaItem(media_url=media_u,
type='youtube',
description='youtube',
created_at=item.fetched_at)
item.media_list.append(media)
if util.within_active_interval(6, 1200):
_comments = util.get_filtered_facebook_comments_data(
'335749279848103',
doc('div.fb-comments').attr('data-href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(
media_url=_comment['json_string'],
type='comments',
description='comments',
created_at=item.fetched_at))
doc, '#articleContent p')
item.raw = doc.text()
item.title = title
item.t = t
item.t_stamp = t_stamp
item.fetched_at = task.fetched_at
item.category = category
item.author = author
item.content = content
item.url = task.url
item.source = 'BastillePost'
item.task_no = self.BATCH_NUMBER
if util.within_active_interval(6, 1200):
_comments = util.get_filtered_facebook_comments_data(
'617988554913649',
doc('#fbComments').attr('href'), task.url)
if _comments:
for _comment in _comments:
item.media_list.append(
self.NewsItem.MediaItem(
media_url=_comment['json_string'],
type='comments',
description='comments',
created_at=item.fetched_at))
@classmethod
def get_auto_configured_spider(cls, offset=0):
bastille_seed = {'http://www.bastillepost.com/hongkong/'}
r = requests.get('http://www.bastillepost.com/hongkong/')
d = pq(r.text)
