sapling/tests/test-url.py
Brodie Rao 186bc90ec4 url: provide url object
This adds a url object that re-implements urlsplit() and
unsplit(). The implementation splits out usernames, passwords, and
ports.

The implementation is based on the behavior specified by RFC
2396[1]. However, it is much more forgiving than the RFC's
specification; it places no specific restrictions on what characters
are allowed in each segment of the URL other than what is necessary to
split the URL into its constituent parts.

[1]: http://www.ietf.org/rfc/rfc2396.txt
2011-03-25 22:58:56 -07:00

191 lines
6.0 KiB
Python

import sys
def check(a, b):
if a != b:
print (a, b)
def cert(cn):
return dict(subject=((('commonName', cn),),))
from mercurial.url import _verifycert
# Test non-wildcard certificates
check(_verifycert(cert('example.com'), 'example.com'),
None)
check(_verifycert(cert('example.com'), 'www.example.com'),
'certificate is for example.com')
check(_verifycert(cert('www.example.com'), 'example.com'),
'certificate is for www.example.com')
# Test wildcard certificates
check(_verifycert(cert('*.example.com'), 'www.example.com'),
None)
check(_verifycert(cert('*.example.com'), 'example.com'),
'certificate is for *.example.com')
check(_verifycert(cert('*.example.com'), 'w.w.example.com'),
'certificate is for *.example.com')
# Test subjectAltName
san_cert = {'subject': ((('commonName', 'example.com'),),),
'subjectAltName': (('DNS', '*.example.net'),
('DNS', 'example.net'))}
check(_verifycert(san_cert, 'example.net'),
None)
check(_verifycert(san_cert, 'foo.example.net'),
None)
# subject is only checked when subjectAltName is empty
check(_verifycert(san_cert, 'example.com'),
'certificate is for *.example.net, example.net')
# Avoid some pitfalls
check(_verifycert(cert('*.foo'), 'foo'),
'certificate is for *.foo')
check(_verifycert(cert('*o'), 'foo'),
'certificate is for *o')
check(_verifycert({'subject': ()},
'example.com'),
'no commonName or subjectAltName found in certificate')
check(_verifycert(None, 'example.com'),
'no certificate received')
import doctest
def test_url():
"""
>>> from mercurial.url import url
This tests for edge cases in url.URL's parsing algorithm. Most of
these aren't useful for documentation purposes, so they aren't
part of the class's doc tests.
Query strings and fragments:
>>> url('http://host/a?b#c')
<url scheme: 'http', host: 'host', path: 'a', query: 'b', fragment: 'c'>
>>> url('http://host/a?')
<url scheme: 'http', host: 'host', path: 'a'>
>>> url('http://host/a#b#c')
<url scheme: 'http', host: 'host', path: 'a', fragment: 'b#c'>
>>> url('http://host/a#b?c')
<url scheme: 'http', host: 'host', path: 'a', fragment: 'b?c'>
>>> url('http://host/?a#b')
<url scheme: 'http', host: 'host', path: '', query: 'a', fragment: 'b'>
>>> url('http://host/?a#b', parse_query=False)
<url scheme: 'http', host: 'host', path: '?a', fragment: 'b'>
>>> url('http://host/?a#b', parse_fragment=False)
<url scheme: 'http', host: 'host', path: '', query: 'a#b'>
>>> url('http://host/?a#b', parse_query=False, parse_fragment=False)
<url scheme: 'http', host: 'host', path: '?a#b'>
IPv6 addresses:
>>> url('ldap://[2001:db8::7]/c=GB?objectClass?one')
<url scheme: 'ldap', host: '[2001:db8::7]', path: 'c=GB',
query: 'objectClass?one'>
>>> url('ldap://joe:xxx@[2001:db8::7]:80/c=GB?objectClass?one')
<url scheme: 'ldap', user: 'joe', passwd: 'xxx', host: '[2001:db8::7]',
port: '80', path: 'c=GB', query: 'objectClass?one'>
Missing scheme, host, etc.:
>>> url('://192.0.2.16:80/')
<url path: '://192.0.2.16:80/'>
>>> url('http://mercurial.selenic.com')
<url scheme: 'http', host: 'mercurial.selenic.com'>
>>> url('/foo')
<url path: '/foo'>
>>> url('bundle:/foo')
<url scheme: 'bundle', path: '/foo'>
>>> url('a?b#c')
<url path: 'a?b', fragment: 'c'>
>>> url('http://x.com?arg=/foo')
<url scheme: 'http', host: 'x.com', query: 'arg=/foo'>
>>> url('http://joe:xxx@/foo')
<url scheme: 'http', user: 'joe', passwd: 'xxx', path: 'foo'>
Just a scheme and a path:
>>> url('mailto:John.Doe@example.com')
<url scheme: 'mailto', path: 'John.Doe@example.com'>
>>> url('a:b:c:d')
<url scheme: 'a', path: 'b:c:d'>
SSH examples:
>>> url('ssh://joe@host//home/joe')
<url scheme: 'ssh', user: 'joe', host: 'host', path: '/home/joe'>
>>> url('ssh://joe:xxx@host/src')
<url scheme: 'ssh', user: 'joe', passwd: 'xxx', host: 'host', path: 'src'>
>>> url('ssh://joe:xxx@host')
<url scheme: 'ssh', user: 'joe', passwd: 'xxx', host: 'host'>
>>> url('ssh://joe@host')
<url scheme: 'ssh', user: 'joe', host: 'host'>
>>> url('ssh://host')
<url scheme: 'ssh', host: 'host'>
>>> url('ssh://')
<url scheme: 'ssh'>
>>> url('ssh:')
<url scheme: 'ssh'>
Non-numeric port:
>>> url('http://example.com:dd')
<url scheme: 'http', host: 'example.com', port: 'dd'>
>>> url('ssh://joe:xxx@host:ssh/foo')
<url scheme: 'ssh', user: 'joe', passwd: 'xxx', host: 'host', port: 'ssh',
path: 'foo'>
Bad authentication credentials:
>>> url('http://joe@joeville:123@4:@host/a?b#c')
<url scheme: 'http', user: 'joe@joeville', passwd: '123@4:',
host: 'host', path: 'a', query: 'b', fragment: 'c'>
>>> url('http://!*#?/@!*#?/:@host/a?b#c')
<url scheme: 'http', host: '!*', fragment: '?/@!*#?/:@host/a?b#c'>
>>> url('http://!*#?@!*#?:@host/a?b#c')
<url scheme: 'http', host: '!*', fragment: '?@!*#?:@host/a?b#c'>
>>> url('http://!*@:!*@@host/a?b#c')
<url scheme: 'http', user: '!*@', passwd: '!*@', host: 'host',
path: 'a', query: 'b', fragment: 'c'>
File paths:
>>> url('a/b/c/d.g.f')
<url path: 'a/b/c/d.g.f'>
>>> url('/x///z/y/')
<url path: '/x///z/y/'>
Empty URL:
>>> u = url('')
>>> u
<url path: ''>
>>> str(u)
''
Empty path with query string:
>>> str(url('http://foo/?bar'))
'http://foo/?bar'
Invalid path:
>>> u = url('http://foo/bar')
>>> u.path = 'bar'
>>> str(u)
'http://foo/bar'
>>> u = url('file:///foo/bar/baz')
>>> u
<url scheme: 'file', path: '/foo/bar/baz'>
>>> str(u)
'file:/foo/bar/baz'
"""
doctest.testmod(optionflags=doctest.NORMALIZE_WHITESPACE)
# Unicode (IDN) certname isn't supported
check(_verifycert(cert(u'\u4f8b.jp'), 'example.jp'),
'IDN in certificate not supported')