a2c15e30658c7a203d9e7c5efdeaa32ea12bdb00
1 #
2 # Copyright (c) 2001 Bizar Software Pty Ltd (http://www.bizarsoftware.com.au/)
3 # This module is free software, and you may redistribute it and/or modify
4 # under the same terms as Python, so long as this copyright message and
5 # disclaimer are retained in their original form.
6 #
7 # IN NO EVENT SHALL BIZAR SOFTWARE PTY LTD BE LIABLE TO ANY PARTY FOR
8 # DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING
9 # OUT OF THE USE OF THIS CODE, EVEN IF THE AUTHOR HAS BEEN ADVISED OF THE
10 # POSSIBILITY OF SUCH DAMAGE.
11 #
12 # BIZAR SOFTWARE PTY LTD SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING,
13 # BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
14 # FOR A PARTICULAR PURPOSE. THE CODE PROVIDED HEREUNDER IS ON AN "AS IS"
15 # BASIS, AND THERE IS NO OBLIGATION WHATSOEVER TO PROVIDE MAINTENANCE,
16 # SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
17 #
18 '''
19 An e-mail gateway for Roundup.
21 Incoming messages are examined for multiple parts:
22 . In a multipart/mixed message or part, each subpart is extracted and
23 examined. The text/plain subparts are assembled to form the textual
24 body of the message, to be stored in the file associated with a "msg"
25 class node. Any parts of other types are each stored in separate files
26 and given "file" class nodes that are linked to the "msg" node.
27 . In a multipart/alternative message or part, we look for a text/plain
28 subpart and ignore the other parts.
30 Summary
31 -------
32 The "summary" property on message nodes is taken from the first non-quoting
33 section in the message body. The message body is divided into sections by
34 blank lines. Sections where the second and all subsequent lines begin with
35 a ">" or "|" character are considered "quoting sections". The first line of
36 the first non-quoting section becomes the summary of the message.
38 Addresses
39 ---------
40 All of the addresses in the To: and Cc: headers of the incoming message are
41 looked up among the user nodes, and the corresponding users are placed in
42 the "recipients" property on the new "msg" node. The address in the From:
43 header similarly determines the "author" property of the new "msg"
44 node. The default handling for addresses that don't have corresponding
45 users is to create new users with no passwords and a username equal to the
46 address. (The web interface does not permit logins for users with no
47 passwords.) If we prefer to reject mail from outside sources, we can simply
48 register an auditor on the "user" class that prevents the creation of user
49 nodes with no passwords.
51 Actions
52 -------
53 The subject line of the incoming message is examined to determine whether
54 the message is an attempt to create a new item or to discuss an existing
55 item. A designator enclosed in square brackets is sought as the first thing
56 on the subject line (after skipping any "Fwd:" or "Re:" prefixes).
58 If an item designator (class name and id number) is found there, the newly
59 created "msg" node is added to the "messages" property for that item, and
60 any new "file" nodes are added to the "files" property for the item.
62 If just an item class name is found there, we attempt to create a new item
63 of that class with its "messages" property initialized to contain the new
64 "msg" node and its "files" property initialized to contain any new "file"
65 nodes.
67 Triggers
68 --------
69 Both cases may trigger detectors (in the first case we are calling the
70 set() method to add the message to the item's spool; in the second case we
71 are calling the create() method to create a new node). If an auditor raises
72 an exception, the original message is bounced back to the sender with the
73 explanatory message given in the exception.
75 $Id: mailgw.py,v 1.10 2001-08-07 00:24:42 richard Exp $
76 '''
79 import string, re, os, mimetools, cStringIO, smtplib, socket, binascii, quopri
80 import traceback
81 import date
83 class Message(mimetools.Message):
84 ''' subclass mimetools.Message so we can retrieve the parts of the
85 message...
86 '''
87 def getPart(self):
88 ''' Get a single part of a multipart message and return it as a new
89 Message instance.
90 '''
91 boundary = self.getparam('boundary')
92 mid, end = '--'+boundary, '--'+boundary+'--'
93 s = cStringIO.StringIO()
94 while 1:
95 line = self.fp.readline()
96 if not line:
97 break
98 if line.strip() in (mid, end):
99 break
100 s.write(line)
101 if not s.getvalue().strip():
102 return None
103 s.seek(0)
104 return Message(s)
106 subject_re = re.compile(r'(\[?(fwd|re):\s*)*'
107 r'(\[(?P<classname>[^\d]+)(?P<nodeid>\d+)?\])'
108 r'(?P<title>[^\[]+)(\[(?P<args>.+?)\])?', re.I)
110 class MailGW:
111 def __init__(self, db):
112 self.db = db
114 def main(self, fp):
115 ''' fp - the file from which to read the Message.
117 Read a message from fp and then call handle_message() with the
118 result. This method's job is to make that call and handle any
119 errors in a sane manner. It should be replaced if you wish to
120 handle errors in a different manner.
121 '''
122 # ok, figure the subject, author, recipients and content-type
123 message = Message(fp)
124 try:
125 self.handle_message(message)
126 except:
127 # bounce the message back to the sender with the error message
128 sendto = [message.getaddrlist('from')[0][1]]
129 m = ['Subject: failed issue tracker submission']
130 m.append('')
131 # TODO as attachments?
132 m.append('---- traceback of failure ----')
133 s = cStringIO.StringIO()
134 import traceback
135 traceback.print_exc(None, s)
136 m.append(s.getvalue())
137 m.append('---- failed message follows ----')
138 try:
139 fp.seek(0)
140 except:
141 pass
142 m.append(fp.read())
143 try:
144 smtp = smtplib.SMTP(self.MAILHOST)
145 smtp.sendmail(self.ADMIN_EMAIL, sendto, '\n'.join(m))
146 except socket.error, value:
147 return "Couldn't send confirmation email: mailhost %s"%value
148 except smtplib.SMTPException, value:
149 return "Couldn't send confirmation email: %s"%value
151 def handle_message(self, message):
152 ''' message - a Message instance
154 Parse the message as per the module docstring.
155 '''
156 # handle the subject line
157 m = subject_re.match(message.getheader('subject'))
158 if not m:
159 raise ValueError, 'No [designator] found in subject "%s"'
160 classname = m.group('classname')
161 nodeid = m.group('nodeid')
162 title = m.group('title').strip()
163 subject_args = m.group('args')
164 cl = self.db.getclass(classname)
165 properties = cl.getprops()
166 props = {}
167 args = m.group('args')
168 if args:
169 for prop in string.split(m.group('args'), ';'):
170 try:
171 key, value = prop.split('=')
172 except ValueError, message:
173 raise ValueError, 'Args list not of form [arg=value,value,...;arg=value,value,value..] (specific exception message was "%s")'%message
174 type = properties[key]
175 if type.isStringType:
176 props[key] = value
177 elif type.isDateType:
178 props[key] = date.Date(value)
179 elif type.isIntervalType:
180 props[key] = date.Interval(value)
181 elif type.isLinkType:
182 props[key] = value
183 elif type.isMultilinkType:
184 props[key] = value.split(',')
186 # handle the users
187 author = self.db.uidFromAddress(message.getaddrlist('from')[0])
188 recipients = []
189 for recipient in message.getaddrlist('to') + message.getaddrlist('cc'):
190 if recipient[1].strip().lower() == self.ISSUE_TRACKER_EMAIL:
191 continue
192 recipients.append(self.db.uidFromAddress(recipient))
194 # now handle the body - find the message
195 content_type = message.gettype()
196 attachments = []
197 if content_type == 'multipart/mixed':
198 # skip over the intro to the first boundary
199 part = message.getPart()
200 content = None
201 while 1:
202 # get the next part
203 part = message.getPart()
204 if part is None:
205 break
206 # parse it
207 subtype = part.gettype()
208 if subtype == 'text/plain' and not content:
209 # add all text/plain parts to the message content
210 if content is None:
211 content = part.fp.read()
212 else:
213 content = content + part.fp.read()
215 elif subtype == 'message/rfc822':
216 # handle message/rfc822 specially - the name should be
217 # the subject of the actual e-mail embedded here
218 i = part.fp.tell()
219 mailmess = Message(part.fp)
220 name = mailmess.getheader('subject')
221 part.fp.seek(i)
222 attachments.append((name, 'message/rfc822', part.fp.read()))
224 else:
225 # try name on Content-Type
226 name = part.getparam('name')
227 # this is just an attachment
228 data = part.fp.read()
229 encoding = part.getencoding()
230 if encoding == 'base64':
231 data = binascii.a2b_base64(data)
232 elif encoding == 'quoted-printable':
233 data = quopri.decode(data)
234 elif encoding == 'uuencoded':
235 data = binascii.a2b_uu(data)
236 attachments.append((name, part.gettype(), data))
238 if content is None:
239 raise ValueError, 'No text/plain part found'
241 elif content_type[:10] == 'multipart/':
242 # skip over the intro to the first boundary
243 message.getPart()
244 content = None
245 while 1:
246 # get the next part
247 part = message.getPart()
248 if part is None:
249 break
250 # parse it
251 if part.gettype() == 'text/plain' and not content:
252 # this one's our content
253 content = part.fp.read()
254 if content is None:
255 raise ValueError, 'No text/plain part found'
257 elif content_type != 'text/plain':
258 raise ValueError, 'No text/plain part found'
260 else:
261 content = message.fp.read()
263 summary, content = parseContent(content)
265 # handle the files
266 files = []
267 for (name, type, data) in attachments:
268 files.append(self.db.file.create(type=type, name=name,
269 content=data))
271 # now handle the db stuff
272 if nodeid:
273 # If an item designator (class name and id number) is found there,
274 # the newly created "msg" node is added to the "messages" property
275 # for that item, and any new "file" nodes are added to the "files"
276 # property for the item.
277 message_id = self.db.msg.create(author=author,
278 recipients=recipients, date=date.Date('.'), summary=summary,
279 content=content, files=files)
280 messages = cl.get(nodeid, 'messages')
281 messages.append(message_id)
282 props['messages'] = messages
283 cl.set(nodeid, **props)
284 else:
285 # If just an item class name is found there, we attempt to create a
286 # new item of that class with its "messages" property initialized to
287 # contain the new "msg" node and its "files" property initialized to
288 # contain any new "file" nodes.
289 message_id = self.db.msg.create(author=author,
290 recipients=recipients, date=date.Date('.'), summary=summary,
291 content=content, files=files)
292 # fill out the properties with defaults where required
293 if properties.has_key('assignedto') and \
294 not props.has_key('assignedto'):
295 props['assignedto'] = '1' # "admin"
296 if properties.has_key('status') and not props.has_key('status'):
297 props['status'] = '1' # "unread"
298 if properties.has_key('title') and not props.has_key('title'):
299 props['title'] = title
300 props['messages'] = [message_id]
301 props['nosy'] = recipients[:]
302 props['nosy'].append(author)
303 props['nosy'].sort()
304 nodeid = cl.create(**props)
306 def parseContent(content, blank_line=re.compile(r'[\r\n]+\s*[\r\n]+'),
307 eol=re.compile(r'[\r\n]+'), signature=re.compile(r'^[>|\s]*[-_]+\s*$')):
308 ''' The message body is divided into sections by blank lines.
309 Sections where the second and all subsequent lines begin with a ">" or "|"
310 character are considered "quoting sections". The first line of the first
311 non-quoting section becomes the summary of the message.
312 '''
313 sections = blank_line.split(content)
314 # extract out the summary from the message
315 summary = ''
316 l = []
317 for section in sections:
318 section = section.strip()
319 if not section:
320 continue
321 lines = eol.split(section)
322 if lines[0] and lines[0][0] in '>|':
323 continue
324 if len(lines) > 1 and lines[1] and lines[1][0] in '>|':
325 continue
326 if not summary:
327 summary = lines[0]
328 l.append(section)
329 continue
330 if signature.match(lines[0]):
331 break
332 l.append(section)
333 return summary, '\n'.join(l)
335 #
336 # $Log: not supported by cvs2svn $
337 # Revision 1.9 2001/08/07 00:15:51 richard
338 # Added the copyright/license notice to (nearly) all files at request of
339 # Bizar Software.
340 #
341 # Revision 1.8 2001/08/05 07:06:07 richard
342 # removed some print statements
343 #
344 # Revision 1.7 2001/08/03 07:18:22 richard
345 # Implemented correct mail splitting (was taking a shortcut). Added unit
346 # tests. Also snips signatures now too.
347 #
348 # Revision 1.6 2001/08/01 04:24:21 richard
349 # mailgw was assuming certain properties existed on the issues being created.
350 #
351 # Revision 1.5 2001/07/29 07:01:39 richard
352 # Added vim command to all source so that we don't get no steenkin' tabs :)
353 #
354 # Revision 1.4 2001/07/28 06:43:02 richard
355 # Multipart message class has the getPart method now. Added some tests for it.
356 #
357 # Revision 1.3 2001/07/28 00:34:34 richard
358 # Fixed some non-string node ids.
359 #
360 # Revision 1.2 2001/07/22 12:09:32 richard
361 # Final commit of Grande Splite
362 #
363 #
364 # vim: set filetype=python ts=4 sw=4 et si