Benjamin_Loison/waybackpy
1
0
Fork 0
Code Issues 1 Pull Requests Packages Projects Releases Wiki Activity

Update wrapper.py

Browse Source
This commit is contained in:
akamhy 2020-05-04 08:56:01 +05:30 committed by GitHub
parent d5fb9a5219
commit 53561ec9c4
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 5 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
waybackpy/wrapper.py
View File

@ -30,9 +30,12 @@ class InvalidUrlError(Exception):
Wayback machine respects these file, will not archive. Wayback machine respects these file, will not archive.
""" """
def clean_url(url):
return str(url).strip().replace(" ","_")
def save(url,UA=default_UA): def save(url,UA=default_UA):
base_save_url = "https://web.archive.org/save/" base_save_url = "https://web.archive.org/save/"
request_url = base_save_url + url request_url = base_save_url + clean_url(url)
hdr = { 'User-Agent' : '%s' % UA } hdr = { 'User-Agent' : '%s' % UA }
req = Request(request_url, headers=hdr) req = Request(request_url, headers=hdr)
if "." not in url: if "." not in url:
@ -62,7 +65,7 @@ def near(
UA=default_UA, UA=default_UA,
): ):
timestamp = str(year)+str(month)+str(day)+str(hour)+str(minute) timestamp = str(year)+str(month)+str(day)+str(hour)+str(minute)
request_url = "https://archive.org/wayback/available?url=%s&timestamp=%s" % (str(url).strip(), str(timestamp)) request_url = "https://archive.org/wayback/available?url=%s&timestamp=%s" % (clean_url(url), str(timestamp))
hdr = { 'User-Agent' : '%s' % UA } hdr = { 'User-Agent' : '%s' % UA }
req = Request(request_url, headers=hdr) req = Request(request_url, headers=hdr)
response = urlopen(req) #nosec response = urlopen(req) #nosec