2017-03-17 15:16:20 -03:00
|
|
|
"""Convert a NT pathname to a file URL and vice versa.
|
|
|
|
|
|
|
|
This module only exists to provide OS-specific code
|
|
|
|
for urllib.requests, thus do not use directly.
|
|
|
|
"""
|
|
|
|
# Testing is done through test_urllib.
|
1996-06-26 16:47:56 -03:00
|
|
|
|
|
|
|
def url2pathname(url):
|
2005-12-26 18:53:56 -04:00
|
|
|
"""OS-specific conversion from a relative URL of the 'file' scheme
|
|
|
|
to a file system path; not recommended for general use."""
|
|
|
|
# e.g.
|
2015-10-24 11:39:36 -03:00
|
|
|
# ///C|/foo/bar/spam.foo
|
|
|
|
# and
|
|
|
|
# ///C:/foo/bar/spam.foo
|
|
|
|
# become
|
|
|
|
# C:\foo\bar\spam.foo
|
2008-06-18 19:38:24 -03:00
|
|
|
import string, urllib.parse
|
2005-12-15 17:59:00 -04:00
|
|
|
# Windows itself uses ":" even in URLs.
|
|
|
|
url = url.replace(':', '|')
|
2001-01-14 20:50:52 -04:00
|
|
|
if not '|' in url:
|
|
|
|
# No drive specifier, just convert slashes
|
2024-11-14 16:22:14 -04:00
|
|
|
if url[:3] == '///':
|
|
|
|
# URL has an empty authority section, so the path begins on the
|
|
|
|
# third character.
|
2001-01-14 20:50:52 -04:00
|
|
|
url = url[2:]
|
|
|
|
# make sure not to convert quoted slashes :-)
|
2024-11-08 12:47:51 -04:00
|
|
|
return urllib.parse.unquote(url.replace('/', '\\'))
|
2001-02-09 07:10:16 -04:00
|
|
|
comp = url.split('|')
|
2001-07-20 15:52:02 -03:00
|
|
|
if len(comp) != 2 or comp[0][-1] not in string.ascii_letters:
|
2001-01-14 20:50:52 -04:00
|
|
|
error = 'Bad URL: ' + url
|
2012-12-25 10:47:37 -04:00
|
|
|
raise OSError(error)
|
2001-02-09 07:10:16 -04:00
|
|
|
drive = comp[0][-1].upper()
|
2024-11-08 12:47:51 -04:00
|
|
|
tail = urllib.parse.unquote(comp[1].replace('/', '\\'))
|
|
|
|
return drive + ':' + tail
|
1996-06-26 16:47:56 -03:00
|
|
|
|
|
|
|
def pathname2url(p):
|
2005-12-26 18:53:56 -04:00
|
|
|
"""OS-specific conversion from a file system path to a relative URL
|
|
|
|
of the 'file' scheme; not recommended for general use."""
|
|
|
|
# e.g.
|
2015-10-24 11:39:36 -03:00
|
|
|
# C:\foo\bar\spam.foo
|
2005-12-26 18:53:56 -04:00
|
|
|
# becomes
|
2015-10-24 11:39:36 -03:00
|
|
|
# ///C:/foo/bar/spam.foo
|
2008-06-18 19:38:24 -03:00
|
|
|
import urllib.parse
|
2021-04-23 14:02:47 -03:00
|
|
|
# First, clean up some special forms. We are going to sacrifice
|
|
|
|
# the additional information anyway
|
2024-11-12 15:52:30 -04:00
|
|
|
p = p.replace('\\', '/')
|
|
|
|
if p[:4] == '//?/':
|
2021-04-23 14:02:47 -03:00
|
|
|
p = p[4:]
|
2024-11-12 15:52:30 -04:00
|
|
|
if p[:4].upper() == 'UNC/':
|
|
|
|
p = '//' + p[4:]
|
2021-04-23 14:02:47 -03:00
|
|
|
elif p[1:2] != ':':
|
|
|
|
raise OSError('Bad path: ' + p)
|
2001-01-14 20:50:52 -04:00
|
|
|
if not ':' in p:
|
2024-11-12 15:52:30 -04:00
|
|
|
# No DOS drive specified, just quote the pathname
|
|
|
|
return urllib.parse.quote(p)
|
2021-04-23 14:02:47 -03:00
|
|
|
comp = p.split(':', maxsplit=2)
|
2001-01-14 20:50:52 -04:00
|
|
|
if len(comp) != 2 or len(comp[0]) > 1:
|
|
|
|
error = 'Bad path: ' + p
|
2012-12-25 10:47:37 -04:00
|
|
|
raise OSError(error)
|
1996-06-26 16:47:56 -03:00
|
|
|
|
2008-06-18 19:38:24 -03:00
|
|
|
drive = urllib.parse.quote(comp[0].upper())
|
2024-11-12 15:52:30 -04:00
|
|
|
tail = urllib.parse.quote(comp[1])
|
2024-11-08 12:47:51 -04:00
|
|
|
return '///' + drive + ':' + tail
|