def calc_item_content(doc, spider):
pages = []
for attach_id in doc['attachments']:
if isinstance(attach_id, ObjectId):
attach = spider.db_adapter.find_attachment_by_id(attach_id)
else:
attach = spider.db_adapter.find_attachment_by_url(attach_id)
if attach:
mime_type = attach.get('mime_type', '')
if mime_type == 'text/html':
pages.append(attach.get('data', ''))
content = ''.join([get_content(page) for page in pages])
return content
def process_item(self, spider):
now = datetime.datetime.utcnow()
doc = self['doc']
doc.pop('detail_url', None)
#计算部分
site_url_hash = url_digest(doc['url'])
content = ''.join([get_content(page) for page in doc.get('detail_pages', [])])
calc_doc = {
'url_hash': site_url_hash,
'content': content,
}
#提取数据的默认填充部分
default_doc1 = {
'spider_name': spider.name,
'data_type': '新闻',
'crawl_time': now,
'query': self['query'],
}
all_doc = chain(calc_doc.iteritems(), default_doc1.iteritems())
for k, v in all_doc:
doc.setdefault(k, v)
