def update_all_mpwx_cron():
"""
更新微信公众号
"""
sites = Site.objects.filter(status='active',
creator='wemp').order_by('-star')
for site in sites:
host, action = get_host_name(site.rss), None
if settings.QNMLGB_HOST in host or settings.ANYV_HOST in host or settings.ERSHICIMI_HOST in host:
if settings.ERSHICIMI_HOST in host:
action = 11
elif settings.QNMLGB_HOST in host:
action = 10
elif settings.WEMP_HOST in host:
action = 12
elif settings.CHUANSONGME_HOST in host:
action = 13
elif settings.ANYV_HOST in host:
action = 14
else:
logger.warning(f"未知的公众号域名：`{host}`{site.cname}")
if action is not None:
make_mpwx_job(site, action)
return True
def submit_a_feed(request):
"""
用户添加一个自定义的订阅源
"""
feed_url = request.POST.get('url', '').strip()[:1024]
user = get_login_user(request)
if feed_url:
host = get_host_name(feed_url)
if 'ershicimi.com' in host:
rsp = parse_wemp_ershicimi(feed_url)
elif host in settings.ALLOWED_HOSTS:
rsp = parse_self_atom(feed_url)
else:
rsp = parse_atom(feed_url)
if rsp:
logger.warning(f"有新订阅源被提交：`{feed_url}")
# 已登录用户，自动订阅
if user:
add_user_sub_feeds(user.oauth_id, [
rsp['name'],
])
return JsonResponse(rsp)
else:
logger.warning(f"RSS 解析失败：`{feed_url}")
return HttpResponseNotFound("Param error")
def update_sites_async(sites, force_update=False):
"""
异步更新某一批订阅源，只支持普通源和公众号的更新
"""
for site_id in sites:
try:
site = Site.objects.get(status='active', pk=site_id)
except:
continue
# 最近已经更新过了，跳过
if not force_update and is_updated_site(site_id):
continue
if site.creator != 'system':
logger.info(f"开始异步更新：{site_id}")
host = get_host_name(site.rss)
if 'ershicimi.com' in host:
parse_wemp_ershicimi(site.rss, update=True)
else:
atom_spider(site)
return True
def submit_a_feed(request):
"""
用户添加一个自定义的订阅源
"""
feed_url = request.POST.get('url', '').strip()[:1024]
user = get_login_user(request)
if feed_url:
host = get_host_name(feed_url)
if settings.ERSHICIMI_HOST in host:
feed_url = feed_url.replace('/user/analysis?bid=', '/a/')
rsp = add_ershicimi_feed(feed_url)
elif host in settings.ALLOWED_HOSTS:
rsp = add_self_feed(feed_url)
elif settings.QNMLGB_HOST in host:
rsp = add_qnmlgb_feed(feed_url)
elif settings.WEMP_HOST in host:
rsp = add_wemp_feed(feed_url)
elif settings.CHUANSONGME_HOST in host:
rsp = add_chuansongme_feed(feed_url)
elif settings.ANYV_HOST in host:
rsp = add_anyv_feed(feed_url)
else:
# 区分播客还是普通 RSS
feed_obj = feedparser.parse(feed_url)
if is_podcast_feed(feed_obj):
rsp = add_postcast_feed(feed_obj)
else:
rsp = add_atom_feed(feed_obj)
if rsp:
logger.warning(f"有新订阅源被提交：`{feed_url}")
set_active_site(rsp['site'])
# 已登录用户，自动订阅
if user:
add_user_sub_feeds(user.oauth_id, [
rsp['site'],
])
if rsp.get('creator') == 'user':
# 新增的普通 RSS 才触发异步更新任务
django_rq.enqueue(update_sites_async, [
rsp['site'],
],
result_ttl=1,
ttl=3600,
failure_ttl=3600)
return JsonResponse(rsp)
else:
logger.warning(f"RSS 解析失败：`{feed_url}")
return HttpResponseNotFound("Param Error")
def wemp_spider(url, site):
"""
抓取微信内容
:param url:
:param site:
:return:
"""
if is_crawled_url(url):
return
try:
rsp = requests.get(url, timeout=10)
if rsp.ok:
try:
if get_host_name(rsp.url) == 'mp.weixin.qq.com':
title, author, content = parse_weixin_page(rsp)
elif 'ershicimi.com' in get_host_name(rsp.url):
title, author, content = parse_ershicimi_page(rsp)
else:
logger.warning(f'公众号域名解析异常：`{rsp.url}')
return
except:
logger.info(f'公众号内容解析异常：`{rsp.url}')
return
article = Article(title=title,
author=author,
site=site,
uindex=current_ts(),
content=content,
src_url=url)
article.save()
mark_crawled_url(url)
except (ConnectTimeout, HTTPError, ReadTimeout, Timeout, ConnectionError):
logger.warning(f'公众号爬取出现网络异常：`{url}')
except:
logger.warning(f'公众号爬取出现未知异常：`{url}')
def wemp_spider(url, site):
"""
抓取微信内容，支持直接微信域名或者 ershicimi 域名
:param url:
:param site:
:return:
"""
if is_crawled_url(url):
return
rsp = get_with_proxy(url)
if rsp is None:
return
if rsp.ok:
try:
if get_host_name(rsp.url) == 'mp.weixin.qq.com':
title, author, content = parse_weixin_page(rsp)
elif 'ershicimi.com' in get_host_name(rsp.url):
title, author, content = parse_ershicimi_page(rsp)
else:
logger.warning(f'公众号域名解析异常：`{rsp.url}')
return
except:
logger.info(f'公众号内容解析异常：`{rsp.url}')
return
article = Article(title=title,
author=author,
site=site,
uindex=current_ts(),
content=content,
src_url=url)
article.save()
mark_crawled_url(url)
def parse_detail_page(job):
response = HtmlResponse(url=job.url, body=job.rsp, encoding='utf8')
title, author, content = None, None, None
# 判断跳转后的域名
host = get_host_name(job.rsp_url)
if job.action == 20 or settings.MPWX_HOST in host:
try:
if response.selector.xpath("//div[@class='weui-msg__text-area']").extract_first():
mark_crawled_url(job.url, job.rsp_url)
logger.info(f"内容违规或删除：`{job.url}")
return 6
except:
pass
title, author, content = parse_mpwx_detail_page(response)
if job.action != 20 and settings.MPWX_HOST in host:
logger.info(f"跳转到微信原文：`{job.url}`{job.rsp_url}`{title}")
elif job.action == 21:
title, author, content = parse_ershicimi_detail_page(response)
elif job.action == 22:
title, author, content = parse_wemp_detail_page(response)
elif job.action == 23:
title, author, content = parse_chuansongme_detail_page(response)
elif job.action == 24:
title, author, content = parse_anyv_detail_page(response)
mark_crawled_url(job.url, job.rsp_url)
if title is None:
logger.warning(f"页面解析失败：`{title}`{job.url}")
return 4
else:
try:
uindex = current_ts()
article = Article(title=title, author=author, site=job.site, uindex=uindex, src_url=job.url)
article.save()
write_dat2_file(uindex, job.site_id, content)
except:
logger.warning(f"插入文章异常：`{title}`{job.site}`{job.url}")
return 7
return 2
def submit_a_feed(request):
"""
用户添加一个自定义的订阅源
"""
feed_url = request.POST.get('url', '').strip()[:1024]
user = get_login_user(request)
if feed_url:
host = get_host_name(feed_url)
if 'ershicimi.com' in host:
feed_url = feed_url.replace('/user/analysis?bid=', '/a/')
rsp = parse_wemp_ershicimi(feed_url)
elif host in settings.ALLOWED_HOSTS:
rsp = parse_self_atom(feed_url)
elif 'qnmlgb.tech' in host:
rsp = parse_qnmlgb_atom(feed_url)
else:
rsp = parse_atom(feed_url)
if rsp:
logger.warning(f"有新订阅源被提交：`{feed_url}")
# 已登录用户，自动订阅
if user:
add_user_sub_feeds(user.oauth_id, [
rsp['site'],
])
# 异步更新任务
django_rq.enqueue(update_sites_async, [
rsp['site'],
])
return JsonResponse(rsp)
else:
logger.warning(f"RSS 解析失败：`{feed_url}")
return HttpResponseNotFound("Param Error")
def update_all_wemp_cron():
"""
更新微信公众号，每天 1～2 次
"""
sites = Site.objects.filter(status='active',
creator='wemp').order_by('-star')
for site in sites:
# 无人订阅的源且不推荐的源不更新
if not is_active_rss(site.pk) and site.star < 9:
continue
if not is_updated_site(site.pk):
host = get_host_name(site.rss)
if 'ershicimi.com' in host:
parse_wemp_ershicimi(site.rss, update=True)
elif 'qnmlgb.tech' in host:
atom_spider(site)
else:
pass
return True
def atom_spider(site):
"""
更新源内容
"""
resp = get_with_retry(site.rss)
if resp is None:
if site.star > 9:
guard_log(f"RSS 源可能失效了`{site.rss}")
else:
logger.info(f"RSS源可能失效了`{site.rss}")
return None
content = BytesIO(resp.content)
feed_obj = feedparser.parse(content)
for entry in feed_obj.entries[:12]:
# 有些是空的
if not entry:
continue
try:
title = entry.title
link = entry.link
except AttributeError:
logger.warning(f'必要属性获取失败：`{site.rss}')
continue
if is_crawled_url(link):
continue
try:
author = entry['author'][:20]
except:
author = ''
try:
value = entry.content[0].value
except:
value = entry.get('description') or entry.link
# to absolute image url
try:
content_soup = BeautifulSoup(value, "html.parser")
for img in content_soup.find_all('img'):
rel_src = img.attrs.get('src')
abs_src = urllib.parse.urljoin(link, rel_src)
img.attrs['src'] = abs_src
value = str(content_soup)
except:
logger.warning(f'修复图片路径异常：`{title}`{link}')
# 公众号 RSS 二次抓取
if get_host_name(site.rss) in ('qnmlgb.tech', ):
if get_host_name(link) in ('mp.weixin.qq.com', ):
rsp = get_with_proxy(link)
if rsp is not None and rsp.ok:
try:
title, author, value = parse_weixin_page(rsp)
except:
pass
try:
article = Article(site=site,
title=title,
author=author,
src_url=link,
uindex=current_ts(),
content=value)
article.save()
mark_crawled_url(link)
except django.db.utils.IntegrityError:
logger.info(f'数据重复插入：`{title}`{link}')
except:
logger.warning(f'数据插入异常：`{title}`{link}')
set_updated_site(site.pk)
return True
def atom_spider(site):
"""
更新源内容
"""
try:
resp = requests.get(site.rss, timeout=30, verify=False)
except:
if site.star > 9:
logger.warning(f"RSS源可能失效了`{site.rss}")
else:
logger.info(f"RSS源可能失效了`{site.rss}")
return None
content = BytesIO(resp.content)
feed_obj = feedparser.parse(content)
for entry in feed_obj.entries[:12]:
try:
title = entry.title
link = entry.link
except AttributeError:
logger.warning(f'必要属性获取失败：`{site.rss}')
continue
if is_crawled_url(link):
continue
try:
author = entry['author'][:20]
except:
author = None
try:
value = entry.content[0].value
except:
value = entry.get('description') or entry.link
# to absolute image url
try:
content_soup = BeautifulSoup(value, "html.parser")
for img in content_soup.find_all('img'):
rel_src = img.attrs.get('src')
abs_src = urllib.parse.urljoin(link, rel_src)
img.attrs['src'] = abs_src
value = str(content_soup)
except:
logger.warning(f'修复图片路径异常：`{title}`{link}')
# 公众号 RSS 二次抓取
if get_host_name(site.rss) in ('qnmlgb.tech', ):
if get_host_name(link) in ('mp.weixin.qq.com', ):
try:
rsp = requests.get(link, timeout=10)
title, author, value = parse_weixin_page(rsp)
except (ConnectTimeout, HTTPError, ReadTimeout, Timeout, ConnectionError):
logger.warning(f"公众号二次爬取出现网络异常：`{link}")
except:
logger.warning(f"公众号二次爬取出现未知异常：`{link}")
try:
article = Article(site=site, title=title, author=author, src_url=link, uindex=current_ts(), content=value)
article.save()
mark_crawled_url(link)
except django.db.utils.IntegrityError:
logger.info(f'数据重复插入：`{title}`{link}')
except:
logger.warning(f'数据插入异常：`{title}`{link}')
