Fix the importing of multipart emails
[stgit] / stgit / commands / imprt.py
1 __copyright__ = """
2 Copyright (C) 2005, Catalin Marinas <catalin.marinas@gmail.com>
3
4 This program is free software; you can redistribute it and/or modify
5 it under the terms of the GNU General Public License version 2 as
6 published by the Free Software Foundation.
7
8 This program is distributed in the hope that it will be useful,
9 but WITHOUT ANY WARRANTY; without even the implied warranty of
10 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11 GNU General Public License for more details.
12
13 You should have received a copy of the GNU General Public License
14 along with this program; if not, write to the Free Software
15 Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
16 """
17
18 import sys, os, re, email
19 from email.Header import decode_header, make_header
20 from mailbox import UnixMailbox
21 from StringIO import StringIO
22 from optparse import OptionParser, make_option
23
24 from stgit.commands.common import *
25 from stgit.utils import *
26 from stgit import stack, git
27
28
29 help = 'import a GNU diff file as a new patch'
30 usage = """%prog [options] [<file>|<url>]
31
32 Create a new patch and apply the given GNU diff file (or the standard
33 input). By default, the file name is used as the patch name but this
34 can be overridden with the '--name' option. The patch can either be a
35 normal file with the description at the top or it can have standard
36 mail format, the Subject, From and Date headers being used for
37 generating the patch information. The command can also read series and
38 mbox files.
39
40 If a patch does not apply cleanly, the failed diff is written to the
41 .stgit-failed.patch file and an empty StGIT patch is added to the
42 stack.
43
44 The patch description has to be separated from the data with a '---'
45 line."""
46
47 options = [make_option('-m', '--mail',
48 help = 'import the patch from a standard e-mail file',
49 action = 'store_true'),
50 make_option('-M', '--mbox',
51 help = 'import a series of patches from an mbox file',
52 action = 'store_true'),
53 make_option('-s', '--series',
54 help = 'import a series of patches',
55 action = 'store_true'),
56 make_option('-u', '--url',
57 help = 'import a patch from a URL',
58 action = 'store_true'),
59 make_option('-n', '--name',
60 help = 'use NAME as the patch name'),
61 make_option('-t', '--strip',
62 help = 'strip numbering and extension from patch name',
63 action = 'store_true'),
64 make_option('-i', '--ignore',
65 help = 'ignore the applied patches in the series',
66 action = 'store_true'),
67 make_option('--replace',
68 help = 'replace the unapplied patches in the series',
69 action = 'store_true'),
70 make_option('-b', '--base',
71 help = 'use BASE instead of HEAD for file importing'),
72 make_option('-e', '--edit',
73 help = 'invoke an editor for the patch description',
74 action = 'store_true'),
75 make_option('-p', '--showpatch',
76 help = 'show the patch content in the editor buffer',
77 action = 'store_true'),
78 make_option('-a', '--author', metavar = '"NAME <EMAIL>"',
79 help = 'use "NAME <EMAIL>" as the author details'),
80 make_option('--authname',
81 help = 'use AUTHNAME as the author name'),
82 make_option('--authemail',
83 help = 'use AUTHEMAIL as the author e-mail'),
84 make_option('--authdate',
85 help = 'use AUTHDATE as the author date'),
86 make_option('--commname',
87 help = 'use COMMNAME as the committer name'),
88 make_option('--commemail',
89 help = 'use COMMEMAIL as the committer e-mail')]
90
91
92 def __end_descr(line):
93 return re.match('---\s*$', line) or re.match('diff -', line) or \
94 re.match('Index: ', line)
95
96 def __strip_patch_name(name):
97 stripped = re.sub('^[0-9]+-(.*)$', '\g<1>', name)
98 stripped = re.sub('^(.*)\.(diff|patch)$', '\g<1>', stripped)
99
100 return stripped
101
102 def __replace_slashes_with_dashes(name):
103 stripped = name.replace('/', '-')
104
105 return stripped
106
107 def __split_descr_diff(string):
108 """Return the description and the diff from the given string
109 """
110 descr = diff = ''
111 top = True
112
113 for line in string.split('\n'):
114 if top:
115 if not __end_descr(line):
116 descr += line + '\n'
117 continue
118 else:
119 top = False
120 diff += line + '\n'
121
122 return (descr.rstrip(), diff)
123
124 def __parse_description(descr):
125 """Parse the patch description and return the new description and
126 author information (if any).
127 """
128 subject = body = ''
129 authname = authemail = authdate = None
130
131 descr_lines = [line.rstrip() for line in descr.split('\n')]
132 if not descr_lines:
133 raise CmdException, "Empty patch description"
134
135 lasthdr = 0
136 end = len(descr_lines)
137
138 # Parse the patch header
139 for pos in range(0, end):
140 if not descr_lines[pos]:
141 continue
142 # check for a "From|Author:" line
143 if re.match('\s*(?:from|author):\s+', descr_lines[pos], re.I):
144 auth = re.findall('^.*?:\s+(.*)$', descr_lines[pos])[0]
145 authname, authemail = name_email(auth)
146 lasthdr = pos + 1
147 continue
148 # check for a "Date:" line
149 if re.match('\s*date:\s+', descr_lines[pos], re.I):
150 authdate = re.findall('^.*?:\s+(.*)$', descr_lines[pos])[0]
151 lasthdr = pos + 1
152 continue
153 if subject:
154 break
155 # get the subject
156 subject = descr_lines[pos]
157 lasthdr = pos + 1
158
159 # get the body
160 if lasthdr < end:
161 body = reduce(lambda x, y: x + '\n' + y, descr_lines[lasthdr:], '')
162
163 return (subject + body, authname, authemail, authdate)
164
165 def __parse_mail(msg):
166 """Parse the message object and return (description, authname,
167 authemail, authdate, diff)
168 """
169 def __decode_header(header):
170 """Decode a qp-encoded e-mail header as per rfc2047"""
171 try:
172 words_enc = decode_header(header)
173 hobj = make_header(words_enc)
174 except Exception, ex:
175 raise CmdException, 'header decoding error: %s' % str(ex)
176 return unicode(hobj).encode('utf-8')
177
178 # parse the headers
179 if msg.has_key('from'):
180 authname, authemail = name_email(__decode_header(msg['from']))
181 else:
182 authname = authemail = None
183
184 # '\n\t' can be found on multi-line headers
185 descr = __decode_header(msg['subject']).replace('\n\t', ' ')
186 authdate = msg['date']
187
188 # remove the '[*PATCH*]' expression in the subject
189 if descr:
190 descr = re.findall('^(\[.*?[Pp][Aa][Tt][Cc][Hh].*?\])?\s*(.*)$',
191 descr)[0][1]
192 else:
193 raise CmdException, 'Subject: line not found'
194
195 # the rest of the message
196 msg_text = ''
197 for part in msg.walk():
198 if part.get_content_type() == 'text/plain':
199 msg_text += part.get_payload(decode = True)
200
201 rem_descr, diff = __split_descr_diff(msg_text)
202 if rem_descr:
203 descr += '\n\n' + rem_descr
204 if not diff:
205 out.warn('Message does not contain any diff')
206
207 # parse the description for author information
208 descr, descr_authname, descr_authemail, descr_authdate = \
209 __parse_description(descr)
210 if descr_authname:
211 authname = descr_authname
212 if descr_authemail:
213 authemail = descr_authemail
214 if descr_authdate:
215 authdate = descr_authdate
216
217 return (descr, authname, authemail, authdate, diff)
218
219 def __parse_patch(fobj):
220 """Parse the input file and return (description, authname,
221 authemail, authdate, diff)
222 """
223 descr, diff = __split_descr_diff(fobj.read())
224 descr, authname, authemail, authdate = __parse_description(descr)
225
226 # we don't yet have an agreed place for the creation date.
227 # Just return None
228 return (descr, authname, authemail, authdate, diff)
229
230 def __create_patch(filename, message, author_name, author_email,
231 author_date, diff, options):
232 """Create a new patch on the stack
233 """
234 if options.name:
235 patch = options.name
236 elif filename:
237 patch = os.path.basename(filename)
238 else:
239 patch = ''
240 if options.strip:
241 patch = __strip_patch_name(patch)
242
243 if not patch:
244 if options.ignore or options.replace:
245 unacceptable_name = lambda name: False
246 else:
247 unacceptable_name = crt_series.patch_exists
248 patch = make_patch_name(message, unacceptable_name)
249 else:
250 # fix possible invalid characters in the patch name
251 patch = re.sub('[^\w.]+', '-', patch).strip('-')
252
253 if not diff:
254 raise CmdException, 'No diff found inside the patch'
255
256 if options.ignore and patch in crt_series.get_applied():
257 out.info('Ignoring already applied patch "%s"' % patch)
258 return
259 if options.replace and patch in crt_series.get_unapplied():
260 crt_series.delete_patch(patch)
261
262 # refresh_patch() will invoke the editor in this case, with correct
263 # patch content
264 if not message:
265 can_edit = False
266
267 committer_name = committer_email = None
268
269 if options.author:
270 options.authname, options.authemail = name_email(options.author)
271
272 # override the automatically parsed settings
273 if options.authname:
274 author_name = options.authname
275 if options.authemail:
276 author_email = options.authemail
277 if options.authdate:
278 author_date = options.authdate
279 if options.commname:
280 committer_name = options.commname
281 if options.commemail:
282 committer_email = options.commemail
283
284 crt_series.new_patch(patch, message = message, can_edit = False,
285 author_name = author_name,
286 author_email = author_email,
287 author_date = author_date,
288 committer_name = committer_name,
289 committer_email = committer_email)
290
291 out.start('Importing patch "%s"' % patch)
292 if options.base:
293 git.apply_patch(diff = diff, base = git_id(options.base))
294 else:
295 git.apply_patch(diff = diff)
296 crt_series.refresh_patch(edit = options.edit,
297 show_patch = options.showpatch)
298 out.done()
299
300 def __import_file(filename, options, patch = None):
301 """Import a patch from a file or standard input
302 """
303 if filename:
304 f = file(filename)
305 else:
306 f = sys.stdin
307
308 if options.mail:
309 try:
310 msg = email.message_from_file(f)
311 except Exception, ex:
312 raise CmdException, 'error parsing the e-mail file: %s' % str(ex)
313 message, author_name, author_email, author_date, diff = \
314 __parse_mail(msg)
315 else:
316 message, author_name, author_email, author_date, diff = \
317 __parse_patch(f)
318
319 if filename:
320 f.close()
321
322 if patch:
323 pname = patch
324 else:
325 pname = filename
326
327 __create_patch(pname, message, author_name, author_email,
328 author_date, diff, options)
329
330 def __import_series(filename, options):
331 """Import a series of patches
332 """
333 applied = crt_series.get_applied()
334
335 if filename:
336 f = file(filename)
337 patchdir = os.path.dirname(filename)
338 else:
339 f = sys.stdin
340 patchdir = ''
341
342 for line in f:
343 patch = re.sub('#.*$', '', line).strip()
344 if not patch:
345 continue
346 patchfile = os.path.join(patchdir, patch)
347 patch = __replace_slashes_with_dashes(patch);
348
349 __import_file(patchfile, options, patch)
350
351 if filename:
352 f.close()
353
354 def __import_mbox(filename, options):
355 """Import a series from an mbox file
356 """
357 if filename:
358 f = file(filename, 'rb')
359 else:
360 f = StringIO(sys.stdin.read())
361
362 try:
363 mbox = UnixMailbox(f, email.message_from_file)
364 except Exception, ex:
365 raise CmdException, 'error parsing the mbox file: %s' % str(ex)
366
367 for msg in mbox:
368 message, author_name, author_email, author_date, diff = \
369 __parse_mail(msg)
370 __create_patch(None, message, author_name, author_email,
371 author_date, diff, options)
372
373 f.close()
374
375 def __import_url(url, options):
376 """Import a patch from a URL
377 """
378 import urllib
379 import tempfile
380
381 if not url:
382 parser.error('URL argument required')
383
384 patch = os.path.basename(urllib.unquote(url))
385 filename = os.path.join(tempfile.gettempdir(), patch)
386 urllib.urlretrieve(url, filename)
387 __import_file(filename, options)
388
389 def func(parser, options, args):
390 """Import a GNU diff file as a new patch
391 """
392 if len(args) > 1:
393 parser.error('incorrect number of arguments')
394
395 check_local_changes()
396 check_conflicts()
397 check_head_top_equal()
398
399 if len(args) == 1:
400 filename = args[0]
401 else:
402 filename = None
403
404 if options.series:
405 __import_series(filename, options)
406 elif options.mbox:
407 __import_mbox(filename, options)
408 elif options.url:
409 __import_url(filename, options)
410 else:
411 __import_file(filename, options)
412
413 print_crt_patch()