def parse_post(self, response):
blog = Posts()
blog['domain'] = get_domain(response.url)
blog['url'] = response.url
blog['title'] = response.xpath('/html/body/div[1]/div[1]/div/h3/text()').get().strip()
blog['author'] = get_author(blog['domain'])
blog['published_date']= dateutil.parser.parse(response.xpath('/html/body/div[1]/div[1]/h2/text()').get())
blog['content'] = "".join(response.xpath('/html/body/div[1]/div[1]/div/div[2]/p/text()').getall())
blog['content_html'] = "".join(response.xpath('/html/body/div[1]/div[1]/div/div[2]/p').getall())
# blog['language'] = get_language(blog['content'])
blog['links'] = get_links(blog['content_html'])
yield blog
#Stats requests
stat = Stats()
stat['domain'] = get_domain(response.url)
stat['url'] = response.url
stat['views'] = None
#Getting likes
post_class = response.xpath('/html/body/@class').get()
post_id = post_class[post_class.find('postid-')+7 : post_class.find('postid-')+12]
blog_id = response.xpath('//*[@id="subscribe-blog"]/p[4]/input[2]/@value').get()
likes_response = requests.get(f"https://public-api.wordpress.com/rest/v1/batch?http_envelope=1&urls[]=/me&urls[]=/sites/{blog_id}/posts/{post_id}/likes&urls[]=/sites/{blog_id}/posts/{post_id}/reblogs/mine").json()
stat['likes'] = likes_response['body'][f'/sites/{blog_id}/posts/{post_id}/likes']['found']
stat['comments'] = None
yield stat
#Comments (Looks like they're turned off)
if response.xpath('/html/body/div[1]/div[1]/div/div[3]/span/text()').get() !='Comments Off':
print("comments on??")
def parse_blog(self, response):
#HTML Content
blog_id = response.xpath('/html/head/link[@rel="shortlink"]/@href').get()
blog = Posts()
blog['domain'] = self.domain
blog['url'] = response.url
blog['title'] = response.xpath('//*[@id="block-zerohedge-page-title"]/h1/span/text()').get()
blog['author'] = response.xpath('//*[@id="block-zerohedge-content"]/article/footer/div[1]/div/div[1]/span/a/text()').get()
blog['published_date']= convert_date(response.xpath('//*[@id="block-zerohedge-content"]/article/footer/div[1]/div/div[2]/span/text()').get())
blog['content'] = " ".join(response.xpath('//*[@id="block-zerohedge-content"]/article/div/div[1]/p//text()').getall())
blog['content_html'] = "".join(response.xpath('//*[@id="block-zerohedge-content"]/article/div/div[1]/p').getall())
blog['links'] = get_links(blog['content_html'])
blog['tags'] = None
yield blog
#Stats requests
stat = Stats()
stat['domain'] = self.domain
stat['url'] = response.url
stat['views'] = requests.get('https://www.zerohedge.com/statistics-ajax?entity_ids={}'.format(blog_id)).json()[blog_id]
stat['likes'] = None
stat['comments'] = requests.get('https://www.zerohedge.com/coral-talk-comment-counts?nids={}'.format(blog_id)).json()[blog_id]
yield stat
#Comments requests
payload = {"query":"query CoralEmbedStream_Embed($assetId: ID, $assetUrl: String, $commentId: ID!, $hasComment: Boolean!, $excludeIgnored: Boolean, $sortBy: SORT_COMMENTS_BY!, $sortOrder: SORT_ORDER!) {\n me {\n id\n state {\n status {\n username {\n status\n __typename\n }\n banned {\n status\n __typename\n }\n suspension {\n until\n __typename\n }\n __typename\n }\n __typename\n }\n __typename\n }\n asset(id: $assetId, url: $assetUrl) {\n ...CoralEmbedStream_Configure_asset\n ...CoralEmbedStream_Stream_asset\n ...CoralEmbedStream_AutomaticAssetClosure_asset\n __typename\n }\n ...CoralEmbedStream_Stream_root\n ...CoralEmbedStream_Configure_root\n}\n\nfragment CoralEmbedStream_Stream_root on RootQuery {\n me {\n state {\n status {\n username {\n status\n __typename\n }\n banned {\n status\n __typename\n }\n suspension {\n until\n __typename\n }\n __typename\n }\n __typename\n }\n ignoredUsers {\n id\n __typename\n }\n role\n __typename\n }\n settings {\n organizationName\n __typename\n }\n ...TalkSlot_StreamFilter_root\n ...CoralEmbedStream_Comment_root\n __typename\n}\n\nfragment CoralEmbedStream_Comment_root on RootQuery {\n me {\n ignoredUsers {\n id\n __typename\n }\n __typename\n }\n ...TalkSlot_CommentInfoBar_root\n ...TalkSlot_CommentAuthorName_root\n ...TalkEmbedStream_DraftArea_root\n ...TalkEmbedStream_DraftArea_root\n __typename\n}\n\nfragment TalkEmbedStream_DraftArea_root on RootQuery {\n __typename\n}\n\nfragment CoralEmbedStream_Stream_asset on Asset {\n comment(id: $commentId) @include(if: $hasComment) {\n ...CoralEmbedStream_Stream_comment\n parent {\n ...CoralEmbedStream_Stream_singleComment\n parent {\n ...CoralEmbedStream_Stream_singleComment\n parent {\n ...CoralEmbedStream_Stream_singleComment\n parent {\n ...CoralEmbedStream_Stream_singleComment\n parent {\n ...CoralEmbedStream_Stream_singleComment\n __typename\n }\n __typename\n }\n __typename\n }\n __typename\n }\n __typename\n }\n __typename\n }\n id\n title\n url\n isClosed\n created_at\n settings {\n moderation\n infoBoxEnable\n infoBoxContent\n premodLinksEnable\n questionBoxEnable\n questionBoxContent\n questionBoxIcon\n closedTimeout\n closedMessage\n disableCommenting\n disableCommentingMessage\n charCountEnable\n charCount\n requireEmailConfirmation\n __typename\n }\n totalCommentCount @skip(if: $hasComment)\n comments(query: {limit: 50000, excludeIgnored: $excludeIgnored, sortOrder: $sortOrder, sortBy: $sortBy}) @skip(if: $hasComment) {\n nodes {\n ...CoralEmbedStream_Stream_comment\n __typename\n }\n hasNextPage\n startCursor\n endCursor\n __typename\n }\n ...TalkSlot_StreamFilter_asset\n ...CoralEmbedStream_Comment_asset\n __typename\n}\n\nfragment CoralEmbedStream_Comment_asset on Asset {\n __typename\n id\n ...TalkSlot_CommentInfoBar_asset\n ...TalkSlot_CommentReactions_asset\n ...TalkSlot_CommentAuthorName_asset\n}\n\nfragment CoralEmbedStream_Stream_comment on Comment {\n id\n status\n user {\n id\n __typename\n }\n ...CoralEmbedStream_Comment_comment\n __typename\n}\n\nfragment CoralEmbedStream_Comment_comment on Comment {\n ...CoralEmbedStream_Comment_SingleComment\n replies(query: {limit : 50000, excludeIgnored: $excludeIgnored}) {\n nodes {\n ...CoralEmbedStream_Comment_SingleComment\n replies(query: {limit : 50000, excludeIgnored: $excludeIgnored}) {\n nodes {\n ...CoralEmbedStream_Comment_SingleComment\n replies(query: {limit : 50000, excludeIgnored: $excludeIgnored}) {\n nodes {\n ...CoralEmbedStream_Comment_SingleComment\n replies(query: {limit : 50000, excludeIgnored: $excludeIgnored}) {\n nodes {\n ...CoralEmbedStream_Comment_SingleComment\n replies(query: {limit : 50000, excludeIgnored: $excludeIgnored}) {\n nodes {\n ...CoralEmbedStream_Comment_SingleComment\n __typename\n }\n hasNextPage\n startCursor\n endCursor\n __typename\n }\n __typename\n }\n hasNextPage\n startCursor\n endCursor\n __typename\n }\n __typename\n }\n hasNextPage\n startCursor\n endCursor\n __typename\n }\n __typename\n }\n hasNextPage\n startCursor\n endCursor\n __typename\n }\n __typename\n }\n hasNextPage\n startCursor\n endCursor\n __typename\n }\n __typename\n}\n\nfragment CoralEmbedStream_Comment_SingleComment on Comment {\n id\n body\n created_at\n status\n replyCount\n tags {\n tag {\n name\n __typename\n }\n __typename\n }\n user {\n id\n username\n __typename\n }\n status_history {\n type\n __typename\n }\n action_summaries {\n __typename\n count\n current_user {\n id\n __typename\n }\n }\n editing {\n edited\n editableUntil\n __typename\n }\n ...TalkSlot_CommentInfoBar_comment\n ...TalkSlot_CommentReactions_comment\n ...TalkSlot_CommentAvatar_comment\n ...TalkSlot_CommentAuthorName_comment\n ...TalkSlot_CommentContent_comment\n ...TalkEmbedStream_DraftArea_comment\n ...TalkEmbedStream_DraftArea_comment\n __typename\n}\n\nfragment TalkEmbedStream_DraftArea_comment on Comment {\n __typename\n ...TalkSlot_DraftArea_comment\n}\n\nfragment CoralEmbedStream_Stream_singleComment on Comment {\n id\n status\n user {\n id\n __typename\n }\n ...CoralEmbedStream_Comment_SingleComment\n __typename\n}\n\nfragment CoralEmbedStream_Configure_root on RootQuery {\n __typename\n ...CoralEmbedStream_Settings_root\n}\n\nfragment CoralEmbedStream_Settings_root on RootQuery {\n __typename\n}\n\nfragment CoralEmbedStream_Configure_asset on Asset {\n __typename\n ...CoralEmbedStream_AssetStatusInfo_asset\n ...CoralEmbedStream_Settings_asset\n}\n\nfragment CoralEmbedStream_AssetStatusInfo_asset on Asset {\n id\n closedAt\n isClosed\n __typename\n}\n\nfragment CoralEmbedStream_Settings_asset on Asset {\n id\n settings {\n moderation\n premodLinksEnable\n questionBoxEnable\n questionBoxIcon\n questionBoxContent\n __typename\n }\n __typename\n}\n\nfragment CoralEmbedStream_AutomaticAssetClosure_asset on Asset {\n id\n closedAt\n __typename\n}\n\nfragment TalkSlot_StreamFilter_root on RootQuery {\n ...TalkViewingOptions_ViewingOptions_root\n __typename\n}\n\nfragment TalkViewingOptions_ViewingOptions_root on RootQuery {\n __typename\n}\n\nfragment TalkSlot_CommentInfoBar_root on RootQuery {\n ...TalkModerationActions_root\n __typename\n}\n\nfragment TalkModerationActions_root on RootQuery {\n me {\n id\n __typename\n }\n __typename\n}\n\nfragment TalkSlot_CommentAuthorName_root on RootQuery {\n ...TalkAuthorMenu_AuthorName_root\n __typename\n}\n\nfragment TalkAuthorMenu_AuthorName_root on RootQuery {\n __typename\n ...TalkSlot_AuthorMenuActions_root\n}\n\nfragment TalkSlot_StreamFilter_asset on Asset {\n ...TalkViewingOptions_ViewingOptions_asset\n __typename\n}\n\nfragment TalkViewingOptions_ViewingOptions_asset on Asset {\n __typename\n}\n\nfragment TalkSlot_CommentInfoBar_asset on Asset {\n ...TalkModerationActions_asset\n ...TalkPermalink_Button_asset\n __typename\n}\n\nfragment TalkModerationActions_asset on Asset {\n id\n __typename\n}\n\nfragment TalkPermalink_Button_asset on Asset {\n url\n __typename\n}\n\nfragment TalkSlot_CommentReactions_asset on Asset {\n ...VoteButton_asset\n __typename\n}\n\nfragment VoteButton_asset on Asset {\n id\n __typename\n}\n\nfragment TalkSlot_CommentAuthorName_asset on Asset {\n ...TalkAuthorMenu_AuthorName_asset\n __typename\n}\n\nfragment TalkAuthorMenu_AuthorName_asset on Asset {\n __typename\n}\n\nfragment TalkSlot_CommentInfoBar_comment on Comment {\n ...CollapseCommentButton_comment\n ...TalkModerationActions_comment\n ...TalkPermalink_Button_comment\n ...TalkInfoBar_moveReportButton_Comment\n ...TalkInfoBar_addEdiableClass_Comment\n __typename\n}\n\nfragment CollapseCommentButton_comment on Comment {\n id\n replyCount\n __typename\n}\n\nfragment TalkModerationActions_comment on Comment {\n id\n status\n user {\n id\n __typename\n }\n tags {\n tag {\n name\n __typename\n }\n __typename\n }\n __typename\n}\n\nfragment TalkPermalink_Button_comment on Comment {\n id\n __typename\n}\n\nfragment TalkInfoBar_moveReportButton_Comment on Comment {\n id\n __typename\n}\n\nfragment TalkInfoBar_addEdiableClass_Comment on Comment {\n id\n editing {\n __typename\n editableUntil\n }\n __typename\n}\n\nfragment TalkSlot_CommentReactions_comment on Comment {\n ...TalkDisableDeepReplies_disableDeepReplies_Comment\n ...VoteButton_comment\n __typename\n}\n\nfragment TalkDisableDeepReplies_disableDeepReplies_Comment on Comment {\n id\n __typename\n}\n\nfragment VoteButton_comment on Comment {\n id\n action_summaries {\n __typename\n ... on UpvoteActionSummary {\n count\n current_user {\n id\n __typename\n }\n __typename\n }\n ... on DownvoteActionSummary {\n count\n current_user {\n id\n __typename\n }\n __typename\n }\n }\n __typename\n}\n\nfragment TalkSlot_CommentAvatar_comment on Comment {\n ...UserAvatar_comment\n __typename\n}\n\nfragment UserAvatar_comment on Comment {\n user {\n avatar\n __typename\n }\n __typename\n}\n\nfragment TalkSlot_CommentAuthorName_comment on Comment {\n ...TalkAuthorMenu_AuthorName_comment\n __typename\n}\n\nfragment TalkAuthorMenu_AuthorName_comment on Comment {\n __typename\n id\n user {\n username\n __typename\n }\n ...TalkSlot_AuthorMenuActions_comment\n}\n\nfragment TalkSlot_CommentContent_comment on Comment {\n ...TalkPluginRichText_CommentContent_comment\n __typename\n}\n\nfragment TalkPluginRichText_CommentContent_comment on Comment {\n body\n richTextBody\n __typename\n}\n\nfragment TalkSlot_DraftArea_comment on Comment {\n ...TalkPluginRichText_Editor_comment\n __typename\n}\n\nfragment TalkPluginRichText_Editor_comment on Comment {\n body\n richTextBody\n __typename\n}\n\nfragment TalkSlot_AuthorMenuActions_root on RootQuery {\n ...TalkIgnoreUser_IgnoreUserAction_root\n __typename\n}\n\nfragment TalkIgnoreUser_IgnoreUserAction_root on RootQuery {\n me {\n id\n __typename\n }\n __typename\n}\n\nfragment TalkSlot_AuthorMenuActions_comment on Comment {\n ...TalkIgnoreUser_IgnoreUserAction_comment\n ...TalkDrupalUserId_DrupalProfile_comment\n __typename\n}\n\nfragment TalkIgnoreUser_IgnoreUserAction_comment on Comment {\n user {\n id\n __typename\n }\n ...TalkIgnoreUser_IgnoreUserConfirmation_comment\n __typename\n}\n\nfragment TalkIgnoreUser_IgnoreUserConfirmation_comment on Comment {\n user {\n id\n username\n __typename\n }\n __typename\n}\n\nfragment TalkDrupalUserId_DrupalProfile_comment on Comment {\n user {\n id\n __typename\n }\n __typename\n}\n","variables":{"assetId":"","assetUrl":blog['url'],"commentId":"","hasComment":False,"excludeIgnored":False,"sortBy":"CREATED_AT","sortOrder":"DESC"},"operationName":"CoralEmbedStream_Embed"}
yield scrapy.Request('https://talk.zerohedge.com/api/v1/graph/ql',
method = 'POST',
body=json.dumps(payload),
headers={'Content-Type':'application/json'},
callback=self.process_comments)
def parse_blog(self, response):
#There are dead links on the pages that can't be processed
if not "The page you are looking for cannot be found or is no longer available." in response.text:
blog = Posts()
blog['domain'] = self.domain
blog['url'] = response.url
blog['title'] = response.xpath(
'//*[@id="Headline"]/h1/text()').get()
blog['author'] = response.xpath(
'//*[@id="Headline"]/p/span/a/text()').get()
blog['published_date'] = parse(
response.xpath(
'//*[@id="Headline"]/p/span/text()').get().replace(
' by: ', ''))
blog['content'] = "".join(
response.xpath('//*[@id="Col2"]/article/div[2]//text()').
getall()).strip().replace('\n', ' ')
blog['content_html'] = response.xpath(
'//*[@id="Col2"]/article/div[2]').get()
blog['links'] = get_links(blog['content_html'])
blog['tags'] = tags_to_json(
response.xpath(
'//*[@id="Headline"]/p/span/i/a/text()').getall())
yield blog
#Stats
stat = Stats()
stat['domain'] = self.domain
stat['url'] = response.url
stat['views'] = get_views(response.url)
stat['likes'] = None
comment_data = get_comments(response.url)
stat['comments'] = comment_data['total']
yield stat
for comment in comment_data['comments']:
#Comments
parsed_comment = Comments()
parsed_comment['domain'] = self.domain
parsed_comment['url'] = response.url
parsed_comment['comment_id'] = comment['id']
parsed_comment['username'] = comment['author'][
'username'] if 'username' in comment[
'author'] else comment['author']['name']
parsed_comment['user_id'] = comment['author'][
'id'] if 'id' in comment['author'] else None
parsed_comment['comment'] = comment['raw_message']
parsed_comment['comment_original'] = comment['message']
parsed_comment['links'] = get_links(comment['message'])
parsed_comment['upvotes'] = comment['likes']
parsed_comment['downvotes'] = comment['dislikes']
parsed_comment['published_date'] = parse(comment['createdAt'])
parsed_comment['reply_count'] = len([
x for x in comment_data['comments']
if str(x['parent']) == str(comment['id'])
])
parsed_comment['reply_to'] = comment['parent']
yield parsed_comment
def parse_blog(self, response):
url = unquote(response.url)
blog = Posts()
try:
blog['domain'] = self.domain
blog['url'] = url
blog['title'] = response.xpath(
'//h1[contains(@class, "css-1ln1egd")]/text()').get()
blog['author'] = get_author(response)
blog['published_date'] = parse(
response.xpath(
'//*[contains(@class, "css-1fboxhy")]/a/text()').get())
blog['content'] = " ".join(
response.xpath('//*[contains(@class, "css-118w07p")]/p//text()'
).getall()).strip().replace('\n', '')
blog['content_html'] = response.xpath(
'//*[contains(@class, "css-118w07p")]').get()
blog['links'] = get_links(
response.xpath('//*[contains(@class, "css-118w07p")]').get())
blog['tags'] = None
if blog['title']: yield blog
except Exception as e:
print(":here")
#Stats
stat = Stats()
stat['domain'] = self.domain
stat['url'] = url
stat['views'] = None
stat['likes'] = None
comments = get_comments(response.url)
stat['comments'] = comments['total'] if comments else None
yield stat
#Comments
for comment in comments['comments']:
if comment[
'body'] != None: #sometimes there are empty reply comments
parsed_comment = Comments()
parsed_comment['domain'] = self.domain
parsed_comment['url'] = url
parsed_comment['comment_id'] = comment['id']
parsed_comment['username'] = comment['user']['username']
parsed_comment['user_id'] = comment['user']['id']
parsed_comment['comment'] = comment['body']
original_comment = comment['richTextBody'] if comment[
'richTextBody'] is not None else comment['body']
parsed_comment['comment_original'] = original_comment
parsed_comment['links'] = get_links(original_comment)
parsed_comment['upvotes'] = comment['action_summaries'][0][
'count'] if comment['action_summaries'] else None
parsed_comment['downvotes'] = None
parsed_comment['published_date'] = parse(comment['created_at'])
parsed_comment['reply_count'] = comment['replyCount']
parsed_comment['reply_to'] = comments['replies'][
parsed_comment['comment_id']] if parsed_comment[
'comment_id'] in comments['replies'] else None
yield parsed_comment
def parse_blog(self, response):
blog = Posts()
blog['domain'] = self.domain
blog['url'] = response.url
blog['title'] = response.xpath(
'//*[@id="Blog1"]/div[1]/div/div/div/div[1]/h3/text()').get(
).strip()
blog['author'] = response.xpath(
'//*[@id="Blog1"]/div[1]/div/div/div/div[1]/div[3]/div[1]/span[1]/span/text()'
).get()
blog['published_date'] = parse(
response.xpath(
'//*[@id="Blog1"]/div[1]/div/div/div/div[1]/div[3]/div[1]/span[2]/a/abbr/@title'
).get())
blog['content'] = " ".join(
response.xpath(
"//div[contains(@class, 'post-body entry-content')]//text()").
getall()).replace('\n', '')
blog['content_html'] = response.xpath(
"//div[contains(@class, 'post-body entry-content')]").get()
blog['links'] = get_links(
response.xpath(
"//div[contains(@class, 'post-body entry-content')]").get())
tags = response.xpath(
'//*[contains(@id, "Blog1")]/div[1]/div/div/div/div[1]/div[3]/div[2]/span//text()'
).getall()
blog['tags'] = tags_to_json(
list(
filter(
lambda a: a != ',\n' and a != '\n' and 'Labels:' not in a,
tags)))
yield blog
#Stats
stat = Stats()
stat['domain'] = self.domain
stat['url'] = response.url
stat['views'] = None
stat['likes'] = None
comment_num = response.xpath('//div[@id="comments"]/h4/text()').get()
if "No" not in comment_num:
stat['comments'] = int(re.search(r'\d+', comment_num).group())
else:
stat['comments'] = None
yield stat
#Comments
if "No" not in comment_num:
for c in response.xpath('//*[@id="top-ra"]//li'):
parsed_comment = Comments()
parsed_comment['domain'] = self.domain
parsed_comment['url'] = response.url
parsed_comment['comment_id'] = c.xpath(
'//li[contains(@class, "comment")]/@id').get()
if c.xpath(
'//li[contains(@class, "comment")]/div[2]/div/cite/a/text()'
).get() is not None:
parsed_comment['username'] = c.xpath(
'//li[contains(@class, "comment")]/div[2]/div/cite/a/text()'
).get()
parsed_comment['user_id'] = c.xpath(
'//li[contains(@class, "comment")]/div[2]/div/cite/a/@href'
).get().replace('https://www.blogger.com/profile/', "")
else:
parsed_comment['username'] = c.xpath(
'//li[contains(@class, "comment")]/div[2]/div/cite/text()'
).get()
parsed_comment['user_id'] = None
parsed_comment['comment'] = " ".join(
c.xpath(
'//li[contains(@class, "comment")]/div[2]/p//text()').
getall())
parsed_comment['comment_original'] = c.xpath(
'//li[contains(@class, "comment")]/div[2]/p').get()
parsed_comment['links'] = get_links(
c.xpath(
'//li[contains(@class, "comment")]/div[2]/p').get())
parsed_comment['upvotes'] = None
parsed_comment['downvotes'] = None
parsed_comment['published_date'] = None
parsed_comment['reply_count'] = None
parsed_comment['reply_to'] = None
yield parsed_comment
def parse_article(self, response):
blog = Posts()
blog['domain'] = self.domain
blog['url'] = response.url
blog['title'] = response.xpath(
'//*[@id="js-post-container"]/div/div[1]/header/h1/text()').get()
blog['author'] = response.xpath(
'//*[@id="js-post-container"]/div/div[1]/header/div[2]/a/span/span[1]/text()'
).get()
blog['published_date'] = get_date(
response.xpath(
'//*[@id="js-post-container"]/div/div[1]/header/div[3]/p/text()'
).get())
blog['content'] = ' '.join(
response.xpath(
'//*[@id="js-post-container"]/div/div[1]/div[1]/div/p/text()').
getall()).strip()
blog['content_html'] = " ".join(
response.xpath(
'//*[@id="js-post-container"]/div/div[1]/div[1]/div').getall())
blog['links'] = get_links(blog['content_html'])
yield blog
#Comments requests
article_url = response.url.replace(
'https://www.buzzfeednews.com/article/', '')
comments, authors = get_comments(article_url)
reply_dic = get_reply_dic(comments)
if comments: #Catches no comments
for c in comments:
parsed_comment = Comments()
parsed_comment['domain'] = self.domain
parsed_comment['url'] = response.url
parsed_comment['comment_id'] = c['id']
parsed_comment['username'] = [
x['name'] for x in authors if c['authorID'] == x['id']
][0]
parsed_comment['user_id'] = c['authorID']
parsed_comment['comment'] = c['body']['text']
parsed_comment['comment_original'] = c['body']['text']
parsed_comment['links'] = get_links(c['body']['text'])
parsed_comment['upvotes'] = c['likeCount']
parsed_comment['downvotes'] = None
parsed_comment['published_date'] = dateutil.parser.parse(
c['timestamp']['text'])
if 'public_replies' in c:
parsed_comment['reply_count'] = len([
x for x in comments
if 'targetID' in x and x['targetID'] == c['id']
])
else:
parsed_comment['reply_count'] = 0
if c['id'] in reply_dic:
parsed_comment['reply_to'] = reply_dic[c['id']]
else:
parsed_comment['reply_to'] = None
yield parsed_comment
#Stats
stat = Stats()
stat['domain'] = self.domain
stat['url'] = response.url
stat['views'] = None
stat['likes'] = None
if comments is None:
stat['comments'] = 0
else:
stat['comments'] = len(comments)
yield stat
def parse_blog(self, response):
# Posts
blog = Posts()
blog['domain'] = self.domain
blog['url'] = response.url
blog['title'] = parse_title(response)
date, author = parse_author_date(response)
author = parse_author(response)
blog['author'] = str(author)[0:99] if author else None
blog['published_date'] = parse_datetime(date) if date else None
content, content_html = parse_content(response)
if content_html:
blog['content'] = content
blog['content_html'] = content_html
blog['links'] = get_links(content_html)
tags = response.xpath('//*[@id="article-tags"]/div//text()').extract()
blog['tags'] = tags_to_json(list(filter(lambda x: '\n' not in x and '\t' not in x, tags))) if tags else None
yield blog
# Comments
comment_data = facebook_comments(response.url, 318812448281278)
comments = comment_data['comments']
authors = comment_data['authors']
reply_dic = comment_data['reply_dic']
if comments: #Catches no comments
for c in comments:
parsed_comment = Comments()
parsed_comment['domain'] = self.domain
parsed_comment['url'] = response.url
parsed_comment['comment_id'] = c['id']
parsed_comment['username'] = [x['name'] for x in authors if c['authorID'] == x['id']][0]
parsed_comment['user_id'] = c['authorID']
parsed_comment['comment'] = c['body']['text']
parsed_comment['comment_original'] = None
parsed_comment['links'] = get_links(c['body']['text'])
parsed_comment['upvotes'] = c['likeCount']
parsed_comment['downvotes'] = None
parsed_comment['published_date'] = parse_datetime(c['timestamp']['text'])
if 'public_replies' in c:
parsed_comment['reply_count'] = len([x for x in comments if 'targetID' in x and x['targetID'] == c['id']])
else:
parsed_comment['reply_count'] = 0
if c['id'] in reply_dic:
parsed_comment['reply_to'] = reply_dic[c['id']]
else:
parsed_comment['reply_to'] = None
yield parsed_comment
# #Stats
stat = Stats()
stat['domain'] = self.domain
stat['url'] = response.url
stat['views'] = None
stat['likes'] = None
if comments is None:
stat['comments'] = 0
else:
stat['comments'] = len(comments)
yield stat
def parse_blog(self, response):
# Posts
script = response.xpath(
"//script[contains(., 'identity')]/text()").extract_first()
try:
data = json.loads(script)
except Exception as e:
print(str(e) + f"\n{str(response.url)}")
data = {}
if data:
blog = Posts()
blog['domain'] = get_domain(response.url)
blog['url'] = response.url
blog['title'] = response.css(
'.article-title::text').extract_first()
author, date = parse_authors_date(data) if data else None
blog['author'] = author.replace(" ",
"").strip() if author else None
blog['published_date'] = date if date else None
blog['tags'] = tags_to_json(
data['page']['tags']) if 'page' in data else None
blog['content'] = get_content(response)
blog['content_html'] = " ".join(
response.xpath('//*[@class="article-content"]').extract())
blog['links'] = get_links(" ".join(
response.xpath('//*[@class="article-content"]').extract()))
yield blog
else:
print('here')
pass
# Comments
article_id = data['page']['articleId'] if 'page' in data else None
comments = get_torontosun_comments(article_id)
if comments: #Catches no comments
for c in comments:
if 'content' in c and c['content']: #Skipping empty comments
parsed_comment = Comments()
parsed_comment['domain'] = self.domain
parsed_comment['url'] = response.url
parsed_comment['comment_id'] = c['content_uuid']
parsed_comment[
'username'] = None #Could not find API to get this
parsed_comment['user_id'] = c['actor_uuid']
parsed_comment[
'comment'] = c['content'] if 'content' in c else None
parsed_comment['comment_original'] = None
parsed_comment['links'] = get_links(
c['content']) if 'content' in c else None
parsed_comment['upvotes'] = c['total_likes']
parsed_comment['downvotes'] = c['total_dislikes']
parsed_comment['published_date'] = parse_datetime(
time.strftime('%m/%d/%Y %H:%M:%S',
time.gmtime(c['date_created'] / 1000.)))
if c['total_replies'] > 0:
parsed_comment['reply_count'] = c['total_replies']
else:
parsed_comment['reply_count'] = 0
if c['content_container_uuid'] != c['thread_uuid'] and c[
'content_container_uuid'] != c['parent_uuid']:
parsed_comment['reply_to'] = c['thread_uuid']
else:
parsed_comment['reply_to'] = None
yield parsed_comment
#Stats
stat = Stats()
stat['domain'] = self.domain
stat['url'] = response.url
stat['views'] = None
stat['likes'] = None
if comments is None:
stat['comments'] = 0
else:
stat['comments'] = len(comments)
yield stat
def parse_blog(self, response):
# Posts
pid = response.css('article').xpath(
"@id").extract_first().strip().replace(' ', '').split('-')[1]
data = get_stats_data(pid)
blog = Posts()
blog['domain'] = self.domain
blog['url'] = response.url
blog['title'] = response.css("h1.entry-title::text").extract_first()
blog['author'] = parse_author(data, response)
blog['published_date'] = parse(
response.css(
"span.entry-meta-date.updated a::text").extract_first())
blog['content'] = "".join(
response.xpath("//div[contains(@class, 'entry-content')]//text()").
extract()).strip().replace('\n', ' ').replace('\t', ' ')
blog['content_html'] = "".join(
response.xpath(
"//div[contains(@class, 'entry-content')]").extract())
blog['links'] = get_links("".join(
response.xpath(
"//div[contains(@class, 'entry-content')]").extract()))
blog['tags'] = tags_to_json(list(
data['tags'].keys())) if data['tags'] else None
yield blog
#Stats
stat = Stats()
stat['domain'] = self.domain
stat['url'] = response.url
stat['views'] = None
stat['likes'] = data['like_count']
stat['comments'] = data['comment_count']
yield stat
#Comments
pids__ = [
comment.strip().replace(' ', '').split('-')[1]
for comment in response.css("li").xpath("@id").extract()
]
req = build_batch(pids__, 70000375)
res = make_api_request(req)
if res:
for res_url in res:
res_values = res[res_url]
#-Parsing comments
parsed_comment = Comments()
parsed_comment['domain'] = self.domain
parsed_comment['url'] = response.url
parsed_comment['comment_id'] = f"bc_comment_{res_values['ID']}"
usr_name = res_values['author']['name'].lower(
) if res_values['author']['name'] else None
parsed_comment['username'] = usr_name
parsed_comment['user_id'] = get_user_id(
self.comment_users, usr_name)
parsed_comment['comment'] = res_values['raw_content']
parsed_comment['comment_original'] = res_values['content']
parsed_comment['links'] = get_links(res_values['content'])
stats_links = res_values['meta']['links']
parsed_comment['upvotes'] = parse_comments(
stats_links['likes'])
parsed_comment['downvotes'] = None
parsed_comment['published_date'] = res_values['date']
#-Getting replies
replies = get_wordpress_replies(res, res_values['ID'],
70000375)
parsed_comment['reply_count'] = replies[0]
parsed_comment['reply_to'] = replies[1]
yield parsed_comment
def parse_article(self, response):
blog = Posts()
blog['domain'] = self.domain
blog['url'] = response.url
blog['title'] = response.xpath(
'//*[@class="news-article-header__title"]/text()').get()
blog['author'] = response.xpath(
'//*[@class="news-byline-full__info-wrapper"]/span/text()').get()
blog['published_date'] = get_date(
response.xpath(
'//*[@class="news-article-header__timestamps-posted"]/text()').
get())
blog['content'] = " ".join(
response.xpath(
'//*[@id="js-post-container"]/div/div[1]/div[1]/div/p/text()').
getall()).strip()
blog['content_html'] = " ".join(
response.xpath(
'//*[@id="js-post-container"]/div/div[1]/div[1]/div').getall())
blog['links'] = get_links(blog['content_html'])
blog['tags'] = tags_to_json(parse_tags(response))
yield blog
#Comments requests
article_url = format_comment_url(response.url)
comment_data = facebook_comments(article_url, '162111247988300')
comments = comment_data['comments']
authors = comment_data['authors']
reply_dic = comment_data['reply_dic']
if comments: #Catches no comments
for c in comments:
parsed_comment = Comments()
parsed_comment['domain'] = self.domain
parsed_comment['url'] = response.url
parsed_comment['comment_id'] = c['id']
parsed_comment['username'] = [
x['name'] for x in authors if c['authorID'] == x['id']
][0]
parsed_comment['user_id'] = c['authorID']
parsed_comment['comment'] = c['body']['text']
parsed_comment['comment_original'] = None
parsed_comment['links'] = get_links(c['body']['text'])
parsed_comment['upvotes'] = c['likeCount']
parsed_comment['downvotes'] = None
parsed_comment['published_date'] = parse_datetime(
c['timestamp']['text'])
if 'public_replies' in c:
parsed_comment['reply_count'] = len([
x for x in comments
if 'targetID' in x and x['targetID'] == c['id']
])
else:
parsed_comment['reply_count'] = 0
if c['id'] in reply_dic:
parsed_comment['reply_to'] = reply_dic[c['id']]
else:
parsed_comment['reply_to'] = None
yield parsed_comment
#Stats
stat = Stats()
stat['domain'] = self.domain
stat['url'] = response.url
stat['views'] = None
stat['likes'] = None
if comments is None:
stat['comments'] = 0
else:
stat['comments'] = len(comments)
yield stat