+
+#
+# Patch description/e-mail/diff parsing
+#
+def __end_descr(line):
+ return re.match('---\s*$', line) or re.match('diff -', line) or \
+ re.match('Index: ', line)
+
+def __split_descr_diff(string):
+ """Return the description and the diff from the given string
+ """
+ descr = diff = ''
+ top = True
+
+ for line in string.split('\n'):
+ if top:
+ if not __end_descr(line):
+ descr += line + '\n'
+ continue
+ else:
+ top = False
+ diff += line + '\n'
+
+ return (descr.rstrip(), diff)
+
+def __parse_description(descr):
+ """Parse the patch description and return the new description and
+ author information (if any).
+ """
+ subject = body = ''
+ authname = authemail = authdate = None
+
+ descr_lines = [line.rstrip() for line in descr.split('\n')]
+ if not descr_lines:
+ raise CmdException, "Empty patch description"
+
+ lasthdr = 0
+ end = len(descr_lines)
+
+ # Parse the patch header
+ for pos in range(0, end):
+ if not descr_lines[pos]:
+ continue
+ # check for a "From|Author:" line
+ if re.match('\s*(?:from|author):\s+', descr_lines[pos], re.I):
+ auth = re.findall('^.*?:\s+(.*)$', descr_lines[pos])[0]
+ authname, authemail = name_email(auth)
+ lasthdr = pos + 1
+ continue
+ # check for a "Date:" line
+ if re.match('\s*date:\s+', descr_lines[pos], re.I):
+ authdate = re.findall('^.*?:\s+(.*)$', descr_lines[pos])[0]
+ lasthdr = pos + 1
+ continue
+ if subject:
+ break
+ # get the subject
+ subject = descr_lines[pos]
+ lasthdr = pos + 1
+
+ # get the body
+ if lasthdr < end:
+ body = reduce(lambda x, y: x + '\n' + y, descr_lines[lasthdr:], '')
+
+ return (subject + body, authname, authemail, authdate)
+
+def parse_mail(msg):
+ """Parse the message object and return (description, authname,
+ authemail, authdate, diff)
+ """
+ from email.Header import decode_header, make_header
+
+ def __decode_header(header):
+ """Decode a qp-encoded e-mail header as per rfc2047"""
+ try:
+ words_enc = decode_header(header)
+ hobj = make_header(words_enc)
+ except Exception, ex:
+ raise CmdException, 'header decoding error: %s' % str(ex)
+ return unicode(hobj).encode('utf-8')
+
+ # parse the headers
+ if msg.has_key('from'):
+ authname, authemail = name_email(__decode_header(msg['from']))
+ else:
+ authname = authemail = None
+
+ # '\n\t' can be found on multi-line headers
+ descr = __decode_header(msg['subject']).replace('\n\t', ' ')
+ authdate = msg['date']
+
+ # remove the '[*PATCH*]' expression in the subject
+ if descr:
+ descr = re.findall('^(\[.*?[Pp][Aa][Tt][Cc][Hh].*?\])?\s*(.*)$',
+ descr)[0][1]
+ else:
+ raise CmdException, 'Subject: line not found'
+
+ # the rest of the message
+ msg_text = ''
+ for part in msg.walk():
+ if part.get_content_type() == 'text/plain':
+ msg_text += part.get_payload(decode = True)
+
+ rem_descr, diff = __split_descr_diff(msg_text)
+ if rem_descr:
+ descr += '\n\n' + rem_descr
+
+ # parse the description for author information
+ descr, descr_authname, descr_authemail, descr_authdate = \
+ __parse_description(descr)
+ if descr_authname:
+ authname = descr_authname
+ if descr_authemail:
+ authemail = descr_authemail
+ if descr_authdate:
+ authdate = descr_authdate
+
+ return (descr, authname, authemail, authdate, diff)
+
+def parse_patch(fobj):
+ """Parse the input file and return (description, authname,
+ authemail, authdate, diff)
+ """
+ descr, diff = __split_descr_diff(fobj.read())
+ descr, authname, authemail, authdate = __parse_description(descr)
+
+ # we don't yet have an agreed place for the creation date.
+ # Just return None
+ return (descr, authname, authemail, authdate, diff)