[parser] Fix spacing for encoded headers

[patchwork] / apps / patchwork / bin / parsemail.py

diff --git a/apps/patchwork/bin/parsemail.py b/apps/patchwork/bin/parsemail.py
index 2310ae8bc187fa52e73b9ffb7378feee0d38193f..e2beeae1b752d8ce03d64ee77d839caf294cd09d 100755 (executable)
--- a/apps/patchwork/bin/parsemail.py
+++ b/apps/patchwork/bin/parsemail.py
@@ -26,11 +26,11 @@ import time
 import operator
 from email import message_from_file
 try:
-    from email.header import Header
+    from email.header import Header, decode_header
     from email.utils import parsedate_tz, mktime_tz
 except ImportError:
     # Python 2.4 compatibility
-    from email.Header import Header
+    from email.Header import Header, decode_header
     from email.Utils import parsedate_tz, mktime_tz
 
 from patchwork.parser import parse_patch
@@ -38,6 +38,23 @@ from patchwork.models import Patch, Project, Person, Comment
 
 list_id_headers = ['List-ID', 'X-Mailing-List']
 
+whitespace_re = re.compile('\s+')
+def normalise_space(str):
+    return whitespace_re.sub(' ', str).strip()
+
+def clean_header(header):
+    """ Decode (possibly non-ascii) headers """
+
+    def decode(fragment):
+        (frag_str, frag_encoding) = fragment
+        if frag_encoding:
+            return frag_str.decode(frag_encoding)
+        return frag_str.decode()
+
+    fragments = map(decode, decode_header(header))
+
+    return normalise_space(u' '.join(fragments))
+
 def find_project(mail):
     project = None
     listid_re = re.compile('.*<([^>]+)>.*', re.S)
@@ -60,7 +77,7 @@ def find_project(mail):
 
 def find_author(mail):
 
-    from_header = mail.get('From').strip()
+    from_header = clean_header(mail.get('From'))
     (name, email) = (None, None)
 
     # tuple of (regex, fn)
@@ -93,7 +110,7 @@ def find_author(mail):
     new_person = False
 
     try:
-        person = Person.objects.get(email = email)
+        person = Person.objects.get(email__iexact = email)
     except Person.DoesNotExist:
         person = Person(name = name, email = email)
         new_person = True
@@ -103,7 +120,6 @@ def find_author(mail):
 def mail_date(mail):
     t = parsedate_tz(mail.get('Date', ''))
     if not t:
-        print "using now()"
         return datetime.datetime.utcnow()
     return datetime.datetime.utcfromtimestamp(mktime_tz(t))
 
@@ -149,9 +165,9 @@ def find_content(project, mail):
 
     if patchbuf:
         mail_headers(mail)
-        patch = Patch(name = clean_subject(mail.get('Subject')),
-                content = patchbuf, date = mail_date(mail),
-                headers = mail_headers(mail))
+       name = clean_subject(mail.get('Subject'), [project.linkname])
+        patch = Patch(name = name, content = patchbuf,
+                    date = mail_date(mail), headers = mail_headers(mail))
 
     if commentbuf:
         if patch:
@@ -224,7 +240,6 @@ def split_prefixes(prefix):
 
 re_re = re.compile('^(re|fwd?)[:\s]\s*', re.I)
 prefix_re = re.compile('^\[([^\]]*)\]\s*(.*)$')
-whitespace_re = re.compile('\s+')
 
 def clean_subject(subject, drop_prefixes = None):
     """ Clean a Subject: header from an incoming patch.
@@ -240,6 +255,8 @@ def clean_subject(subject, drop_prefixes = None):
     'meep'
     >>> clean_subject('[PATCH] meep')
     'meep'
+    >>> clean_subject('[PATCH] meep \\n meep')
+    'meep meep'
     >>> clean_subject('[PATCH RFC] meep')
     '[RFC] meep'
     >>> clean_subject('[PATCH,RFC] meep')
@@ -274,6 +291,8 @@ def clean_subject(subject, drop_prefixes = None):
     # remove Re:, Fwd:, etc
     subject = re_re.sub(' ', subject)
 
+    subject = normalise_space(subject)
+
     prefixes = []
 
     match = prefix_re.match(subject)
@@ -286,7 +305,7 @@ def clean_subject(subject, drop_prefixes = None):
         subject = match.group(2)
         match = prefix_re.match(subject)
 
-    subject = whitespace_re.sub(' ', subject)
+    subject = normalise_space(subject)
 
     subject = subject.strip()
     if prefixes:
@@ -294,8 +313,9 @@ def clean_subject(subject, drop_prefixes = None):
 
     return subject
 
-sig_re = re.compile('^(-{2,3} ?|_+)\n.*', re.S | re.M)
+sig_re = re.compile('^(-- |_+)\n.*', re.S | re.M)
 def clean_content(str):
+    """ Try to remove signature (-- ) and list footer (_____) cruft """
     str = sig_re.sub('', str)
     return str.strip()