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, decode_header
30 from email.utils import parsedate_tz, mktime_tz
32 # Python 2.4 compatibility
33 from email.Header import Header, decode_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 clean_header(header):
42 """ Decode (possibly non-ascii) headers """
44 def decode(str, fragment):
45 (frag_str, frag_encoding) = fragment
47 return str + frag_str.decode(frag_encoding)
48 return str + frag_str.decode()
50 return reduce(decode, decode_header(header), u'').strip()
52 def find_project(mail):
54 listid_re = re.compile('.*<([^>]+)>.*', re.S)
56 for header in list_id_headers:
58 match = listid_re.match(mail.get(header))
62 listid = match.group(1)
65 project = Project.objects.get(listid = listid)
72 def find_author(mail):
74 from_header = clean_header(mail.get('From'))
75 (name, email) = (None, None)
77 # tuple of (regex, fn)
78 # - where fn returns a (name, email) tuple from the match groups resulting
79 # from re.match().groups()
81 # for "Firstname Lastname" <example@example.com> style addresses
82 (re.compile('"?(.*?)"?\s*<([^>]+)>'), (lambda g: (g[0], g[1]))),
84 # for example@example.com (Firstname Lastname) style addresses
85 (re.compile('"?(.*?)"?\s*\(([^\)]+)\)'), (lambda g: (g[1], g[0]))),
88 (re.compile('(.*)'), (lambda g: (None, g[0]))),
91 for regex, fn in from_res:
92 match = regex.match(from_header)
94 (name, email) = fn(match.groups())
98 raise Exception("Could not parse From: header")
100 email = email.strip()
107 person = Person.objects.get(email = email)
108 except Person.DoesNotExist:
109 person = Person(name = name, email = email)
112 return (person, new_person)
115 t = parsedate_tz(mail.get('Date', ''))
117 return datetime.datetime.utcnow()
118 return datetime.datetime.utcfromtimestamp(mktime_tz(t))
120 def mail_headers(mail):
121 return reduce(operator.__concat__,
122 ['%s: %s\n' % (k, Header(v, header_name = k, \
123 continuation_ws = '\t').encode()) \
124 for (k, v) in mail.items()])
126 def find_content(project, mail):
130 for part in mail.walk():
131 if part.get_content_maintype() != 'text':
134 #print "\t%s, %s" % \
135 # (part.get_content_subtype(), part.get_content_charset())
137 charset = part.get_content_charset()
139 charset = mail.get_charset()
143 payload = unicode(part.get_payload(decode=True), charset, "replace")
145 if part.get_content_subtype() == 'x-patch':
148 if part.get_content_subtype() == 'plain':
150 (patchbuf, c) = parse_patch(payload)
155 commentbuf += c.strip() + '\n'
162 name = clean_subject(mail.get('Subject'), [project.linkname])
163 patch = Patch(name = name, content = patchbuf,
164 date = mail_date(mail), headers = mail_headers(mail))
170 cpatch = find_patch_for_comment(mail)
173 comment = Comment(patch = cpatch, date = mail_date(mail),
174 content = clean_content(commentbuf),
175 headers = mail_headers(mail))
177 return (patch, comment)
179 def find_patch_for_comment(mail):
180 # construct a list of possible reply message ids
182 if 'In-Reply-To' in mail:
183 refs.append(mail.get('In-Reply-To'))
185 if 'References' in mail:
186 rs = mail.get('References').split()
195 # first, check for a direct reply
197 patch = Patch.objects.get(msgid = ref)
199 except Patch.DoesNotExist:
202 # see if we have comments that refer to a patch
204 comment = Comment.objects.get(msgid = ref)
206 except Comment.DoesNotExist:
212 split_re = re.compile('[,\s]+')
214 def split_prefixes(prefix):
215 """ Turn a prefix string into a list of prefix tokens
217 >>> split_prefixes('PATCH')
219 >>> split_prefixes('PATCH,RFC')
221 >>> split_prefixes('')
223 >>> split_prefixes('PATCH,')
225 >>> split_prefixes('PATCH ')
227 >>> split_prefixes('PATCH,RFC')
229 >>> split_prefixes('PATCH 1/2')
232 matches = split_re.split(prefix)
233 return [ s for s in matches if s != '' ]
235 re_re = re.compile('^(re|fwd?)[:\s]\s*', re.I)
236 prefix_re = re.compile('^\[([^\]]*)\]\s*(.*)$')
237 whitespace_re = re.compile('\s+')
239 def clean_subject(subject, drop_prefixes = None):
240 """ Clean a Subject: header from an incoming patch.
242 Removes Re: and Fwd: strings, as well as [PATCH]-style prefixes. By
243 default, only [PATCH] is removed, and we keep any other bracketed data
244 in the subject. If drop_prefixes is provided, remove those too,
245 comparing case-insensitively.
247 >>> clean_subject('meep')
249 >>> clean_subject('Re: meep')
251 >>> clean_subject('[PATCH] meep')
253 >>> clean_subject('[PATCH] meep \\n meep')
255 >>> clean_subject('[PATCH RFC] meep')
257 >>> clean_subject('[PATCH,RFC] meep')
259 >>> clean_subject('[PATCH,1/2] meep')
261 >>> clean_subject('[PATCH RFC 1/2] meep')
263 >>> clean_subject('[PATCH] [RFC] meep')
265 >>> clean_subject('[PATCH] [RFC,1/2] meep')
267 >>> clean_subject('[PATCH] [RFC] [1/2] meep')
269 >>> clean_subject('[PATCH] rewrite [a-z] regexes')
270 'rewrite [a-z] regexes'
271 >>> clean_subject('[PATCH] [RFC] rewrite [a-z] regexes')
272 '[RFC] rewrite [a-z] regexes'
273 >>> clean_subject('[foo] [bar] meep', ['foo'])
275 >>> clean_subject('[FOO] [bar] meep', ['foo'])
279 if drop_prefixes is None:
282 drop_prefixes = [ s.lower() for s in drop_prefixes ]
284 drop_prefixes.append('patch')
286 # remove Re:, Fwd:, etc
287 subject = re_re.sub(' ', subject)
289 # normalise whitespace
290 subject = whitespace_re.sub(' ', subject)
294 match = prefix_re.match(subject)
297 prefix_str = match.group(1)
298 prefixes += [ p for p in split_prefixes(prefix_str) \
299 if p.lower() not in drop_prefixes]
301 subject = match.group(2)
302 match = prefix_re.match(subject)
304 subject = whitespace_re.sub(' ', subject)
306 subject = subject.strip()
308 subject = '[%s] %s' % (','.join(prefixes), subject)
312 sig_re = re.compile('^(-- |_+)\n.*', re.S | re.M)
313 def clean_content(str):
314 """ Try to remove signature (-- ) and list footer (_____) cruft """
315 str = sig_re.sub('', str)
319 mail = message_from_file(sys.stdin)
321 # some basic sanity checks
322 if 'From' not in mail:
325 if 'Subject' not in mail:
328 if 'Message-Id' not in mail:
331 hint = mail.get('X-Patchwork-Hint', '').lower()
335 project = find_project(mail)
337 print "no project found"
340 msgid = mail.get('Message-Id').strip()
342 (author, save_required) = find_author(mail)
344 (patch, comment) = find_content(project, mail)
347 # we delay the saving until we know we have a patch.
350 save_required = False
351 patch.submitter = author
353 patch.project = project
356 except Exception, ex:
362 # looks like the original constructor for Comment takes the pk
363 # when the Comment is created. reset it here.
365 comment.patch = patch
366 comment.submitter = author
367 comment.msgid = msgid
370 except Exception, ex:
375 if __name__ == '__main__':
376 sys.exit(main(sys.argv))