def parse_article_detail_html(self, response):
"""
文章详情解析 html 版
:param response:
:return:
"""
article_title = response.xpath(
'//div[@class="title"]/text()').extract_first(default='')
article_pub_time = response.xpath(
'//span[@class="time"]/text()').extract_first(default='')
article_content = response.xpath(
'//div[@class="WB_editor_iframe"]').extract_first(default='')
fetch_result_item = FetchResultItem()
fetch_result_item['task_id'] = response.meta['task_id']
fetch_result_item['platform_id'] = response.meta['platform_id']
fetch_result_item['platform_name'] = platform_name_map.get(
response.meta['platform_id'], '')
fetch_result_item['channel_id'] = response.meta['channel_id']
fetch_result_item['channel_name'] = channel_name_map.get(
response.meta['channel_id'], '')
fetch_result_item['article_id'] = response.meta['article_id']
fetch_result_item['article_title'] = article_title
fetch_result_item['article_author_id'] = response.meta['follow_id']
fetch_result_item['article_author_name'] = response.meta['follow_name']
fetch_result_item['article_pub_time'] = article_pub_time
fetch_result_item['article_url'] = response.url
fetch_result_item['article_tags'] = ''
fetch_result_item['article_abstract'] = response.meta[
'article_abstract']
fetch_result_item['article_content'] = article_content
yield fetch_result_item
def parse_article_detail_js(self, response):
"""
文章详情解析 js 版
:param response:
:return:
"""
article_detail_body = response.body_as_unicode()
article_detail_rule = r'<script>FM.view\({"ns":.*?"html":"(.*?)"}\)</script>'
article_detail_re_parse = re.compile(article_detail_rule,
re.S).findall(article_detail_body)
if not article_detail_re_parse:
return
article_detail_html = ''.join(article_detail_re_parse)
# 转义字符处理
article_detail_html = article_detail_html.replace('\\r', '')
article_detail_html = article_detail_html.replace('\\t', '')
article_detail_html = article_detail_html.replace('\\n', '')
article_detail_html = article_detail_html.replace('\\"', '"')
article_detail_html = article_detail_html.replace('\\/', '/')
article_detail_doc = fromstring(article_detail_html)
article_title_parse = article_detail_doc.xpath(
'//h1[@class="title"]/text()')
article_title = article_title_parse[0].strip(
) if article_title_parse else ''
article_pub_time_parse = article_detail_doc.xpath(
'//span[@class="time"]/text()')
article_pub_time = self.trans_time(article_pub_time_parse[0].strip(
)) if article_pub_time_parse else time.strftime('%Y-%m-%d %H:%M:%S')
article_content_parse = article_detail_doc.xpath(
'//div[@class="WBA_content"]')
article_content = tostring(
article_content_parse[0],
encoding='unicode').strip() if article_content_parse else ''
fetch_result_item = FetchResultItem()
fetch_result_item['task_id'] = response.meta['task_id']
fetch_result_item['platform_id'] = response.meta['platform_id']
fetch_result_item['platform_name'] = platform_name_map.get(
response.meta['platform_id'], '')
fetch_result_item['channel_id'] = response.meta['channel_id']
fetch_result_item['channel_name'] = channel_name_map.get(
response.meta['channel_id'], '')
fetch_result_item['article_id'] = response.meta['article_id']
fetch_result_item['article_title'] = article_title
fetch_result_item['article_author_id'] = response.meta['follow_id']
fetch_result_item['article_author_name'] = response.meta['follow_name']
fetch_result_item['article_pub_time'] = time_local_to_utc(
article_pub_time).strftime('%Y-%m-%d %H:%M:%S')
fetch_result_item['article_url'] = response.url
fetch_result_item['article_tags'] = ''
fetch_result_item['article_abstract'] = response.meta[
'article_abstract']
fetch_result_item['article_content'] = article_content
yield fetch_result_item
def parse_article_detail_js(self, response):
"""
文章详情解析 js 版
:param response:
:return:
"""
article_detail_body = response.body_as_unicode()
article_detail_rule = ur'<script>FM.view\({"ns":.*?"html":"(.*?)"}\)</script>'
article_detail_re_parse = re.compile(article_detail_rule,
re.S).findall(article_detail_body)
if not article_detail_re_parse:
return
article_detail_html = u''.join(article_detail_re_parse)
article_detail_doc = fromstring(article_detail_html.replace(
u'\\', u''))
article_title_parse = article_detail_doc.xpath(
'//h1[@class="title"]/text()')
article_title = article_title_parse[0].strip(
) if article_title_parse else u''
article_pub_time_parse = article_detail_doc.xpath(
'//span[@class="time"]/text()')
article_pub_time = self.trans_time(article_pub_time_parse[0].strip())
article_content_parse = article_detail_doc.xpath(
'//div[@class="WBA_content"]/text()')
article_content = article_content_parse[0].strip(
) if article_content_parse else u''
fetch_result_item = FetchResultItem()
fetch_result_item['task_id'] = response.meta['task_id']
fetch_result_item['platform_id'] = response.meta['platform_id']
fetch_result_item['platform_name'] = platform_name_map.get(
response.meta['platform_id'], u'')
fetch_result_item['channel_id'] = response.meta['channel_id']
fetch_result_item['channel_name'] = channel_name_map.get(
response.meta['channel_id'], u'')
fetch_result_item['article_id'] = response.meta['article_id']
fetch_result_item['article_title'] = article_title
fetch_result_item['article_author_id'] = response.meta['follow_id']
fetch_result_item['article_author_name'] = response.meta['follow_name']
fetch_result_item['article_pub_time'] = article_pub_time
fetch_result_item['article_url'] = response.url
fetch_result_item['article_tags'] = u''
fetch_result_item['article_abstract'] = response.meta[
'article_abstract']
fetch_result_item['article_content'] = article_content
yield fetch_result_item
def parse_detail(self, response):
"""
详细页面
:param response:
:return:
"""
print(response.meta)
article_content = ''.join([
i.strip()
for i in response.xpath('//div[@id="js_content"]/*').extract()
])
# 原创内容处理（处理内容为空）
if not article_content:
share_source_url = response.xpath(
'//a[@id="js_share_source"]/@href').extract_first()
yield scrapy.Request(url=share_source_url,
callback=self.parse_detail,
meta=response.meta)
return
fetch_result_item = FetchResultItem()
fetch_result_item['task_id'] = response.meta['task_id']
fetch_result_item['platform_id'] = response.meta['platform_id']
fetch_result_item['platform_name'] = platform_name_map.get(
response.meta['platform_id'], u'')
fetch_result_item['channel_id'] = response.meta['channel_id']
fetch_result_item['channel_name'] = channel_name_map.get(
response.meta['channel_id'], u'')
fetch_result_item['article_id'] = response.meta['article_id']
fetch_result_item['article_title'] = response.meta['article_title']
fetch_result_item['article_author_id'] = response.meta['follow_id']
fetch_result_item['article_author_name'] = response.meta['follow_name']
fetch_result_item['article_pub_time'] = response.meta[
'article_pub_time']
fetch_result_item['article_url'] = response.meta['article_url']
fetch_result_item['article_tags'] = u''
fetch_result_item['article_abstract'] = response.meta[
'article_abstract']
fetch_result_item['article_content'] = article_content
yield fetch_result_item
def parse_article_detail(self, response):
"""
文章详情
:param response:
:return:
"""
toutiao_body = response.body_as_unicode()
js_body = parse_toutiao_js_body(toutiao_body,
response.meta['detail_url'])
if not js_body:
return
pj = ParseJsTt(js_body=js_body)
article_id = pj.parse_js_item_id()
article_title = pj.parse_js_title()
article_abstract = pj.parse_js_abstract()
article_content = pj.parse_js_content()
article_pub_time = pj.parse_js_pub_time()
article_tags = pj.parse_js_tags()
fetch_result_item = FetchResultItem()
fetch_result_item['task_id'] = response.meta['task_id']
fetch_result_item['platform_id'] = response.meta['platform_id']
fetch_result_item['platform_name'] = platform_name_map.get(
response.meta['platform_id'], '')
fetch_result_item['channel_id'] = response.meta['channel_id']
fetch_result_item['channel_name'] = channel_name_map.get(
response.meta['channel_id'], '')
fetch_result_item['article_id'] = article_id
fetch_result_item['article_title'] = article_title
fetch_result_item['article_author_id'] = response.meta['follow_id']
fetch_result_item['article_author_name'] = response.meta['follow_name']
fetch_result_item['article_pub_time'] = time_local_to_utc(
article_pub_time).strftime('%Y-%m-%d %H:%M:%S')
fetch_result_item[
'article_url'] = response.url or response.meta['detail_url']
fetch_result_item['article_tags'] = article_tags
fetch_result_item['article_abstract'] = article_abstract
fetch_result_item['article_content'] = article_content
yield fetch_result_item