3 # Patchwork - automated patch tracking system
4 # Copyright (C) 2008 Jeremy Kerr <jk@ozlabs.org>
6 # This file is part of the Patchwork package.
8 # Patchwork is free software; you can redistribute it and/or modify
9 # it under the terms of the GNU General Public License as published by
10 # the Free Software Foundation; either version 2 of the License, or
11 # (at your option) any later version.
13 # Patchwork is distributed in the hope that it will be useful,
14 # but WITHOUT ANY WARRANTY; without even the implied warranty of
15 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 # GNU General Public License for more details.
18 # You should have received a copy of the GNU General Public License
19 # along with Patchwork; if not, write to the Free Software
20 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
27 from email import message_from_file
29 from email.header import Header
30 from email.utils import parsedate_tz, mktime_tz
32 # Python 2.4 compatibility
33 from email.Header import Header
34 from email.Utils import parsedate_tz, mktime_tz
36 from patchwork.parser import parse_patch
37 from patchwork.models import Patch, Project, Person, Comment
39 list_id_headers = ['List-ID', 'X-Mailing-List']
41 def find_project(mail):
43 listid_re = re.compile('.*<([^>]+)>.*', re.S)
45 for header in list_id_headers:
47 match = listid_re.match(mail.get(header))
51 listid = match.group(1)
54 project = Project.objects.get(listid = listid)
61 def find_author(mail):
63 from_header = mail.get('From').strip()
64 (name, email) = (None, None)
66 # tuple of (regex, fn)
67 # - where fn returns a (name, email) tuple from the match groups resulting
68 # from re.match().groups()
70 # for "Firstname Lastname" <example@example.com> style addresses
71 (re.compile('"?(.*?)"?\s*<([^>]+)>'), (lambda g: (g[0], g[1]))),
73 # for example@example.com (Firstname Lastname) style addresses
74 (re.compile('"?(.*?)"?\s*\(([^\)]+)\)'), (lambda g: (g[1], g[0]))),
77 (re.compile('(.*)'), (lambda g: (None, g[0]))),
80 for regex, fn in from_res:
81 match = regex.match(from_header)
83 (name, email) = fn(match.groups())
87 raise Exception("Could not parse From: header")
96 person = Person.objects.get(email = email)
97 except Person.DoesNotExist:
98 person = Person(name = name, email = email)
101 return (person, new_person)
104 t = parsedate_tz(mail.get('Date', ''))
106 return datetime.datetime.utcnow()
107 return datetime.datetime.utcfromtimestamp(mktime_tz(t))
109 def mail_headers(mail):
110 return reduce(operator.__concat__,
111 ['%s: %s\n' % (k, Header(v, header_name = k, \
112 continuation_ws = '\t').encode()) \
113 for (k, v) in mail.items()])
115 def find_content(project, mail):
119 for part in mail.walk():
120 if part.get_content_maintype() != 'text':
123 #print "\t%s, %s" % \
124 # (part.get_content_subtype(), part.get_content_charset())
126 charset = part.get_content_charset()
128 charset = mail.get_charset()
132 payload = unicode(part.get_payload(decode=True), charset, "replace")
134 if part.get_content_subtype() == 'x-patch':
137 if part.get_content_subtype() == 'plain':
139 (patchbuf, c) = parse_patch(payload)
144 commentbuf += c.strip() + '\n'
151 name = clean_subject(mail.get('Subject'), [project.linkname])
152 patch = Patch(name = name, content = patchbuf,
153 date = mail_date(mail), headers = mail_headers(mail))
159 cpatch = find_patch_for_comment(mail)
162 comment = Comment(patch = cpatch, date = mail_date(mail),
163 content = clean_content(commentbuf),
164 headers = mail_headers(mail))
166 return (patch, comment)
168 def find_patch_for_comment(mail):
169 # construct a list of possible reply message ids
171 if 'In-Reply-To' in mail:
172 refs.append(mail.get('In-Reply-To'))
174 if 'References' in mail:
175 rs = mail.get('References').split()
184 # first, check for a direct reply
186 patch = Patch.objects.get(msgid = ref)
188 except Patch.DoesNotExist:
191 # see if we have comments that refer to a patch
193 comment = Comment.objects.get(msgid = ref)
195 except Comment.DoesNotExist:
201 split_re = re.compile('[,\s]+')
203 def split_prefixes(prefix):
204 """ Turn a prefix string into a list of prefix tokens
206 >>> split_prefixes('PATCH')
208 >>> split_prefixes('PATCH,RFC')
210 >>> split_prefixes('')
212 >>> split_prefixes('PATCH,')
214 >>> split_prefixes('PATCH ')
216 >>> split_prefixes('PATCH,RFC')
218 >>> split_prefixes('PATCH 1/2')
221 matches = split_re.split(prefix)
222 return [ s for s in matches if s != '' ]
224 re_re = re.compile('^(re|fwd?)[:\s]\s*', re.I)
225 prefix_re = re.compile('^\[([^\]]*)\]\s*(.*)$')
226 whitespace_re = re.compile('\s+')
228 def clean_subject(subject, drop_prefixes = None):
229 """ Clean a Subject: header from an incoming patch.
231 Removes Re: and Fwd: strings, as well as [PATCH]-style prefixes. By
232 default, only [PATCH] is removed, and we keep any other bracketed data
233 in the subject. If drop_prefixes is provided, remove those too,
234 comparing case-insensitively.
236 >>> clean_subject('meep')
238 >>> clean_subject('Re: meep')
240 >>> clean_subject('[PATCH] meep')
242 >>> clean_subject('[PATCH] meep \\n meep')
244 >>> clean_subject('[PATCH RFC] meep')
246 >>> clean_subject('[PATCH,RFC] meep')
248 >>> clean_subject('[PATCH,1/2] meep')
250 >>> clean_subject('[PATCH RFC 1/2] meep')
252 >>> clean_subject('[PATCH] [RFC] meep')
254 >>> clean_subject('[PATCH] [RFC,1/2] meep')
256 >>> clean_subject('[PATCH] [RFC] [1/2] meep')
258 >>> clean_subject('[PATCH] rewrite [a-z] regexes')
259 'rewrite [a-z] regexes'
260 >>> clean_subject('[PATCH] [RFC] rewrite [a-z] regexes')
261 '[RFC] rewrite [a-z] regexes'
262 >>> clean_subject('[foo] [bar] meep', ['foo'])
264 >>> clean_subject('[FOO] [bar] meep', ['foo'])
268 if drop_prefixes is None:
271 drop_prefixes = [ s.lower() for s in drop_prefixes ]
273 drop_prefixes.append('patch')
275 # remove Re:, Fwd:, etc
276 subject = re_re.sub(' ', subject)
278 # normalise whitespace
279 subject = whitespace_re.sub(' ', subject)
283 match = prefix_re.match(subject)
286 prefix_str = match.group(1)
287 prefixes += [ p for p in split_prefixes(prefix_str) \
288 if p.lower() not in drop_prefixes]
290 subject = match.group(2)
291 match = prefix_re.match(subject)
293 subject = whitespace_re.sub(' ', subject)
295 subject = subject.strip()
297 subject = '[%s] %s' % (','.join(prefixes), subject)
301 sig_re = re.compile('^(-{2,3} ?|_+)\n.*', re.S | re.M)
302 def clean_content(str):
303 str = sig_re.sub('', str)
307 mail = message_from_file(sys.stdin)
309 # some basic sanity checks
310 if 'From' not in mail:
313 if 'Subject' not in mail:
316 if 'Message-Id' not in mail:
319 hint = mail.get('X-Patchwork-Hint', '').lower()
323 project = find_project(mail)
325 print "no project found"
328 msgid = mail.get('Message-Id').strip()
330 (author, save_required) = find_author(mail)
332 (patch, comment) = find_content(project, mail)
335 # we delay the saving until we know we have a patch.
338 save_required = False
339 patch.submitter = author
341 patch.project = project
344 except Exception, ex:
350 # looks like the original constructor for Comment takes the pk
351 # when the Comment is created. reset it here.
353 comment.patch = patch
354 comment.submitter = author
355 comment.msgid = msgid
358 except Exception, ex:
363 if __name__ == '__main__':
364 sys.exit(main(sys.argv))