python/sitemap
1
0
Fork 0
Code Issues Pull Requests Releases 1 Wiki Activity

修改get_url

Browse Source
This commit is contained in:
zeek 2020-03-22 14:13:23 +08:00
parent 2967957ab8
commit 63d48cab02
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
get_url.py
View File

@ -80,8 +80,10 @@ def filterlinks(links):
continue continue
if is_robot(url): if is_robot(url):
continue continue
if not ishttp: if not ishttp and url.startswith("/"):
url = url_root + url url = url_root + url
elif not ishttp:
url = url_root + '/' + url
tmplinks.append(url) tmplinks.append(url)
reslinks = list(set(tmplinks)) reslinks = list(set(tmplinks))
return reslinks return reslinks