4
# - UTF-8 filenames are now allowed (Eli Golovinsky)<br/>
5
# - File object is no more mandatory, Object only needs to have seek() read() attributes (Eli Golovinsky)<br/>
8
# - upload is now done with chunks (Adam Ambrose)
12
# bug fix: kosh @T aesaeion.com
13
# HTTPS support : Ryan Grow <ryangrow @T yahoo.com>
15
# Copyright (C) 2004,2005,2006 Fabien SEISEN
17
# This library is free software; you can redistribute it and/or
18
# modify it under the terms of the GNU Lesser General Public
19
# License as published by the Free Software Foundation; either
20
# version 2.1 of the License, or (at your option) any later version.
22
# This library is distributed in the hope that it will be useful,
23
# but WITHOUT ANY WARRANTY; without even the implied warranty of
24
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
25
# Lesser General Public License for more details.
27
# You should have received a copy of the GNU Lesser General Public
28
# License along with this library; if not, write to the Free Software
29
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
31
# you can contact me at: <fabien@seisen.org>
32
# http://fabien.seisen.org/python/
34
# Also modified by Adam Ambrose (aambrose @T pacbell.net) to write data in
35
# chunks (hardcoded to CHUNK_SIZE for now), so the entire contents of the file
36
# don't need to be kept in memory.
39
enable to upload files using multipart/form-data
42
upload files in python:
43
http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/146306
45
timeoutsocket.py: overriding Python socket API:
46
http://www.timo-tasi.org/python/timeoutsocket.py
47
http://mail.python.org/pipermail/python-announce-list/2001-December/001095.html
51
u = urllib2.urlopen('http://site.com/path' [, data])
53
data can be a mapping object or a sequence of two-elements tuples
54
(like in original urllib2.urlopen())
55
varname still need to be a string and
56
value can be string of a file object
80
def get_content_type(filename):
81
return mimetypes.guess_type(filename)[0] or 'application/octet-stream'
83
# if sock is None, juste return the estimate size
84
def send_data(v_vars, v_files, boundary, sock=None):
88
buffer += '--%s\r\n' % boundary
89
buffer += 'Content-Disposition: form-data; name="%s"\r\n' % k
95
for (k, v) in v_files:
97
if hasattr(fd, 'size'):
100
file_size = os.fstat(fd.fileno())[stat.ST_SIZE]
101
name = fd.name.split('/')[-1]
102
if isinstance(name, unicode):
103
name = name.encode('UTF-8')
105
buffer += '--%s\r\n' % boundary
106
buffer += 'Content-Disposition: form-data; name="%s"; filename="%s"\r\n' \
108
buffer += 'Content-Type: %s\r\n' % get_content_type(name)
109
buffer += 'Content-Length: %s\r\n' % file_size
115
if hasattr(fd, 'seek'):
118
chunk = fd.read(CHUNK_SIZE)
124
buffer += '--%s--\r\n' % boundary
131
# mainly a copy of HTTPHandler from urllib2
132
class newHTTPHandler(urllib2.BaseHandler):
133
def http_open(self, req):
134
return self.do_open(httplib.HTTP, req)
136
def do_open(self, http_class, req):
137
data = req.get_data()
140
# mapping object (dict)
141
if req.has_data() and type(data) != str:
142
if hasattr(data, 'items'):
146
if len(data) and not isinstance(data[0], tuple):
149
ty, va, tb = sys.exc_info()
150
raise TypeError, "not a valid non-string sequence or mapping object", tb
153
if hasattr(v, 'read'):
154
v_files.append((k, v))
156
v_vars.append( (k, v) )
157
# no file ? convert to string
158
if len(v_vars) > 0 and len(v_files) == 0:
159
data = urllib.urlencode(v_vars)
162
host = req.get_host()
164
raise urllib2.URLError('no host given')
166
h = http_class(host) # will parse host:port
168
h.putrequest('POST', req.get_selector())
169
if not 'Content-type' in req.headers:
171
boundary = mimetools.choose_boundary()
172
l = send_data(v_vars, v_files, boundary)
173
h.putheader('Content-Type',
174
'multipart/form-data; boundary=%s' % boundary)
175
h.putheader('Content-length', str(l))
177
h.putheader('Content-type',
178
'application/x-www-form-urlencoded')
179
if not 'Content-length' in req.headers:
180
h.putheader('Content-length', '%d' % len(data))
182
h.putrequest('GET', req.get_selector())
184
scheme, sel = urllib.splittype(req.get_selector())
185
sel_host, sel_path = urllib.splithost(sel)
186
h.putheader('Host', sel_host or host)
187
for name, value in self.parent.addheaders:
188
name = name.capitalize()
189
if name not in req.headers:
190
h.putheader(name, value)
191
for k, v in req.headers.items():
193
# httplib will attempt to connect() here. be prepared
194
# to convert a socket error to a URLError.
197
except socket.error, err:
198
raise urllib2.URLError(err)
202
l = send_data(v_vars, v_files, boundary, h)
203
elif len(v_vars) > 0:
204
# if data is passed as dict ...
205
data = urllib.urlencode(v_vars)
208
# "normal" urllib2.urlopen()
211
code, msg, hdrs = h.getreply()
214
resp = urllib.addinfourl(fp, hdrs, req.get_full_url())
219
return self.parent.error('http', req, fp, code, msg, hdrs)
221
urllib2._old_HTTPHandler = urllib2.HTTPHandler
222
urllib2.HTTPHandler = newHTTPHandler
224
class newHTTPSHandler(newHTTPHandler):
225
def https_open(self, req):
226
return self.do_open(httplib.HTTPS, req)
228
urllib2.HTTPSHandler = newHTTPSHandler
230
if __name__ == '__main__':
239
SYNTAX: %s -u url -f file [-v]
243
opts, args = getopt.getopt(sys.argv[1:], 'hvu:f:')
244
except getopt.GetoptError, errmsg:
245
print "ERROR:", errmsg
252
for name, value in opts:
256
elif name in ('-v',):
258
elif name in ('-u',):
260
elif name in ('-f',):
263
print "invalid argument:", name
277
fd = open(v_file, 'r')
281
# u = urllib2.urlopen(v_url, data)
282
req = urllib2.Request(v_url, data, {})
284
u = urllib2.urlopen(req)
285
except urllib2.HTTPError, errobj:
286
print "HTTPError:", errobj.code