1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
|
# Copyright (C) 2005 Canonical Ltd
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
"""Implementation of Transport over http.
"""
from bzrlib.transport import Transport, register_transport
from bzrlib.errors import (TransportNotPossible, NoSuchFile,
TransportError, ConnectionError)
import os, errno
from cStringIO import StringIO
import urllib, urllib2
import urlparse
from bzrlib.errors import BzrError, BzrCheckError
from bzrlib.branch import Branch
from bzrlib.trace import mutter
def extract_auth(url, password_manager):
"""
Extract auth parameters from am HTTP/HTTPS url and add them to the given
password manager. Return the url, minus those auth parameters (which
confuse urllib2).
"""
assert url.startswith('http://') or url.startswith('https://')
scheme, host = url.split('//', 1)
if '/' in host:
host, path = host.split('/', 1)
path = '/' + path
else:
path = ''
port = ''
if '@' in host:
auth, host = host.split('@', 1)
if ':' in auth:
username, password = auth.split(':', 1)
else:
username, password = auth, None
if ':' in host:
host, port = host.split(':', 1)
port = ':' + port
# FIXME: if password isn't given, should we ask for it?
if password is not None:
username = urllib.unquote(username)
password = urllib.unquote(password)
password_manager.add_password(None, host, username, password)
url = scheme + '//' + host + port + path
return url
def get_url(url):
import urllib2
mutter("get_url %s" % url)
manager = urllib2.HTTPPasswordMgrWithDefaultRealm()
url = extract_auth(url, manager)
auth_handler = urllib2.HTTPBasicAuthHandler(manager)
opener = urllib2.build_opener(auth_handler)
url_f = opener.open(url)
return url_f
class HttpTransport(Transport):
"""This is the transport agent for http:// access.
TODO: Implement pipelined versions of all of the *_multi() functions.
"""
def __init__(self, base):
"""Set the base path where files will be stored."""
assert base.startswith('http://') or base.startswith('https://')
super(HttpTransport, self).__init__(base)
# In the future we might actually connect to the remote host
# rather than using get_url
# self._connection = None
(self._proto, self._host,
self._path, self._parameters,
self._query, self._fragment) = urlparse.urlparse(self.base)
def should_cache(self):
"""Return True if the data pulled across should be cached locally.
"""
return True
def clone(self, offset=None):
"""Return a new HttpTransport with root at self.base + offset
For now HttpTransport does not actually connect, so just return
a new HttpTransport object.
"""
if offset is None:
return HttpTransport(self.base)
else:
return HttpTransport(self.abspath(offset))
def abspath(self, relpath):
"""Return the full url to the given relative path.
This can be supplied with a string or a list
"""
assert isinstance(relpath, basestring)
if isinstance(relpath, basestring):
relpath_parts = relpath.split('/')
else:
# TODO: Don't call this with an array - no magic interfaces
relpath_parts = relpath[:]
if len(relpath_parts) > 1:
if relpath_parts[0] == '':
raise ValueError("path %r within branch %r seems to be absolute"
% (relpath, self._path))
if relpath_parts[-1] == '':
raise ValueError("path %r within branch %r seems to be a directory"
% (relpath, self._path))
basepath = self._path.split('/')
if len(basepath) > 0 and basepath[-1] == '':
basepath = basepath[:-1]
for p in relpath_parts:
if p == '..':
if len(basepath) == 0:
# In most filesystems, a request for the parent
# of root, just returns root.
continue
basepath.pop()
elif p == '.' or p == '':
continue # No-op
else:
basepath.append(p)
# Possibly, we could use urlparse.urljoin() here, but
# I'm concerned about when it chooses to strip the last
# portion of the path, and when it doesn't.
path = '/'.join(basepath)
return urlparse.urlunparse((self._proto,
self._host, path, '', '', ''))
def has(self, relpath):
"""Does the target location exist?
TODO: HttpTransport.has() should use a HEAD request,
not a full GET request.
TODO: This should be changed so that we don't use
urllib2 and get an exception, the code path would be
cleaner if we just do an http HEAD request, and parse
the return code.
"""
path = relpath
try:
path = self.abspath(relpath)
f = get_url(path)
# Without the read and then close()
# we tend to have busy sockets.
f.read()
f.close()
return True
except urllib2.URLError, e:
mutter('url error code: %s for has url: %r', e.code, path)
if e.code == 404:
return False
raise
except IOError, e:
mutter('io error: %s %s for has url: %r',
e.errno, errno.errorcode.get(e.errno), path)
if e.errno == errno.ENOENT:
return False
raise TransportError(orig_error=e)
def get(self, relpath, decode=False):
"""Get the file at the given relative path.
:param relpath: The relative path to the file
"""
path = relpath
try:
path = self.abspath(relpath)
return get_url(path)
except urllib2.HTTPError, e:
mutter('url error code: %s for has url: %r', e.code, path)
if e.code == 404:
raise NoSuchFile(path, extra=e)
raise
except (BzrError, IOError), e:
if hasattr(e, 'errno'):
mutter('io error: %s %s for has url: %r',
e.errno, errno.errorcode.get(e.errno), path)
if e.errno == errno.ENOENT:
raise NoSuchFile(path, extra=e)
raise ConnectionError(msg = "Error retrieving %s: %s"
% (self.abspath(relpath), str(e)),
orig_error=e)
def put(self, relpath, f, mode=None):
"""Copy the file-like or string object into the location.
:param relpath: Location to put the contents, relative to base.
:param f: File-like or string object.
"""
raise TransportNotPossible('http PUT not supported')
def mkdir(self, relpath, mode=None):
"""Create a directory at the given path."""
raise TransportNotPossible('http does not support mkdir()')
def append(self, relpath, f):
"""Append the text in the file-like object into the final
location.
"""
raise TransportNotPossible('http does not support append()')
def copy(self, rel_from, rel_to):
"""Copy the item at rel_from to the location at rel_to"""
raise TransportNotPossible('http does not support copy()')
def copy_to(self, relpaths, other, mode=None, pb=None):
"""Copy a set of entries from self into another Transport.
:param relpaths: A list/generator of entries to be copied.
TODO: if other is LocalTransport, is it possible to
do better than put(get())?
"""
# At this point HttpTransport might be able to check and see if
# the remote location is the same, and rather than download, and
# then upload, it could just issue a remote copy_this command.
if isinstance(other, HttpTransport):
raise TransportNotPossible('http cannot be the target of copy_to()')
else:
return super(HttpTransport, self).copy_to(relpaths, other, mode=mode, pb=pb)
def move(self, rel_from, rel_to):
"""Move the item at rel_from to the location at rel_to"""
raise TransportNotPossible('http does not support move()')
def delete(self, relpath):
"""Delete the item at relpath"""
raise TransportNotPossible('http does not support delete()')
def listable(self):
"""See Transport.listable."""
return False
def stat(self, relpath):
"""Return the stat information for a file.
"""
raise TransportNotPossible('http does not support stat()')
def lock_read(self, relpath):
"""Lock the given file for shared (read) access.
:return: A lock object, which should be passed to Transport.unlock()
"""
# The old RemoteBranch ignore lock for reading, so we will
# continue that tradition and return a bogus lock object.
class BogusLock(object):
def __init__(self, path):
self.path = path
def unlock(self):
pass
return BogusLock(relpath)
def lock_write(self, relpath):
"""Lock the given file for exclusive (write) access.
WARNING: many transports do not support this, so trying avoid using it
:return: A lock object, which should be passed to Transport.unlock()
"""
raise TransportNotPossible('http does not support lock_write()')
|