Hopiu/linkchecker
1
0
Fork 0
mirror of https://github.com/Hopiu/linkchecker.git synced 2026-03-25 02:10:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

pathnoquote chars adjusted

Browse source 
git-svn-id: https://linkchecker.svn.sourceforge.net/svnroot/linkchecker/trunk/linkchecker@1548 e7d03fd6-7b0d-0410-9947-9c21f3af8025
This commit is contained in:
calvin 2004-08-27 17:57:34 +00:00
parent 7adc342ec5
commit bcffed778e
2 changed files with 7 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
linkcheck/tests/test_url.py
View file

@ -70,6 +70,10 @@ class TestUrl (unittest.TestCase):
url = "http://example.com/?q=1%2a2"
nurl = "http://example.com/?q=1%2A2"
self.assertEqual(linkcheck.url.url_norm(url), nurl)
# the no-quote chars
url = "http://example.com/a*+-();b"
nurl = url
self.assertEqual(linkcheck.url.url_norm(url), nurl)
def test_norm_case_sensitivity (self):
"""test url norm case sensitivity"""
@ -243,8 +247,8 @@ class TestUrl (unittest.TestCase):
url = 'nntp:'
nurl = 'nntp:///'
self.assertEqual(linkcheck.url.url_norm(url), nurl)
url = "news:§$%&/´`(§%"
nurl = 'news:%A7%24%25%26/%B4%60%28%A7%25'
url = "news:§$%&/´`§%"
nurl = 'news:%A7%24%25%26/%B4%60%A7%25'
self.assertEqual(linkcheck.url.url_norm(url), nurl)
# javascript url
url = "javascript:loadthis()"

5
linkcheck/url.py
View file

@ -194,10 +194,7 @@ def url_norm (url):
# quote parts again
urlparts[0] = urllib.quote(urlparts[0]) # scheme
urlparts[1] = urllib.quote(urlparts[1], ':@') # host
nopathquote = ';/=,~'
# note: the list of chars not to quote is different for javascript urls
if urlparts[0]=='javascript':
nopathquote += '()+-'
nopathquote = ';/=,~*-+()'
urlparts[2] = urllib.quote(urlparts[2], nopathquote) # path
res = urlparse.urlunsplit(urlparts)
if url.endswith('#') and not urlparts[4]: