Make parsing IANA mail for comments more flexible. Commit ready for merge.

- Legacy-Id: 9641

ietf/sync/iana.py

@@ -215,19 +215,29 @@ def update_history_with_changes(changes, send_email=True):
     return added_events, warnings
 
 
+def find_document_name(text):
+    prefixes = ['draft','conflict-review','status-change','charter']
+    leading_delimiter_re = '(?<![-a-zA-Z0-9])'
+    prefix_re = '(%s)' % '|'.join(prefixes)
+    tail_re = '(-[a-z0-9]+)+?(-\d\d\.txt)?'
+    trailing_delimiter_re = '((?![-a-zA-Z0-9])|$)'
+    name_re = '%s(%s%s)%s' % (leading_delimiter_re, prefix_re, tail_re, trailing_delimiter_re)
+    m = re.search(name_re,text)
+    return m and m.group(0).lower()
+
+def strip_version_extension(text):
+    if re.search(r"\.\w{3}$", text): # strip off extension
+        text = text[:-4]
+    if re.search(r"-\d{2}$", text): # strip off revision
+        text = text[:-3]
+    return text
+
 def parse_review_email(text):
     msg = email.message_from_string(text)
 
     # doc
-    doc_name = ""
-    m = re.search(r"<([^>]+)>", msg["Subject"])
-    if m:
-        doc_name = m.group(1).lower()
-        if re.search(r"\.\w{3}$", doc_name): # strip off extension
-            doc_name = doc_name[:-4]
-
-        if re.search(r"-\d{2}$", doc_name): # strip off revision
-            doc_name = doc_name[:-3]
+    doc_name = find_document_name(msg["Subject"]) or ""
+    doc_name = strip_version_extension(doc_name)
 
     # date
     review_time = datetime.datetime.now()
@@ -252,14 +262,18 @@ def parse_review_email(text):
 
     # comment
     body = msg.get_payload().decode('quoted-printable').replace("\r", "")
-    if "BEGIN IANA LAST CALL COMMENTS" in body:
-        b = body.find("(BEGIN IANA LAST CALL COMMENTS)")
-        e = body.find("(END IANA LAST CALL COMMENTS)")
-        comment = body[b + len("(BEGIN IANA LAST CALL COMMENTS)"):e].strip()
-    elif "BEGIN IANA COMMENTS" in body:
-        b = body.find("(BEGIN IANA COMMENTS)")
-        e = body.find("(END IANA COMMENTS)")
-        comment = body[b + len("(BEGIN IANA COMMENTS)"):e].strip()
+
+    begin_search = re.search('\(BEGIN\s+IANA\s+(LAST\s+CALL\s+)?COMMENTS?(\s*:\s*[a-zA-Z0-9-\.]*)?\s*\)',body)
+    end_search = re.search('\(END\s+IANA\s+(LAST\s+CALL\s+)?COMMENTS?\)',body)
+    if begin_search and end_search:
+        begin_string = begin_search.group(0)
+        end_string = end_search.group(0)
+        b = body.find(begin_string)
+        e = body.find(end_string)
+        comment = body[b + len(begin_string):e].strip()
+        embedded_name = strip_version_extension(find_document_name(begin_string) or "")
+        if embedded_name:
+            doc_name = embedded_name
     else:
         comment = ""
 

ietf/sync/tests.py

@@ -131,11 +131,12 @@ class IANASyncTests(TestCase):
     def test_iana_review_mail(self):
         draft = make_test_data()
 
+        subject_template = u'Subject: [IANA #12345] Last Call: <%(draft)s-%(rev)s.txt> (Long text) to Informational RFC'
         msg_template = u"""From: "%(person)s via RT" <drafts-lastcall@iana.org>
 Date: Thu, 10 May 2012 12:00:0%(rtime)d +0000
-Subject: [IANA #12345] Last Call: <%(draft)s-%(rev)s.txt> (Long text) to Informational RFC
+%(subject)s
 
-(BEGIN IANA %(tag)s)
+(BEGIN IANA %(tag)s%(embedded_name)s)
 
 IESG:
 
@@ -154,18 +155,30 @@ ICANN
 (END IANA %(tag)s)
 """
 
-        tags = ("LAST CALL COMMENTS","COMMENTS")
+        subjects =  ( subject_template % dict(draft=draft.name,rev=draft.rev) , 'Subject: Vacuous Subject' )
+
+        tags = ('LAST CALL COMMENTS', 'COMMENTS')
+
+        embedded_names = (': %s-%s.txt'%(draft.name,draft.rev), '')
+
+        for subject in subjects:
             for tag in tags:
+                for embedded_name in embedded_names:
+                    if embedded_name or not 'Vacuous' in subject: 
+                    
+                        rtime = 7*subjects.index(subject) + 5*tags.index(tag) + embedded_names.index(embedded_name)
                         msg = msg_template % dict(person=Person.objects.get(user__username="iana").name,
                                                   draft=draft.name,
                                                   rev=draft.rev,
                                                   tag=tag,
-                                      rtime=tags.index(tag))
+                                                  rtime=rtime,
+                                                  subject=subject,
+                                                  embedded_name=embedded_name,)
     
                         doc_name, review_time, by, comment = iana.parse_review_email(msg.encode('utf-8'))
     
                         self.assertEqual(doc_name, draft.name)
-#            self.assertEqual(review_time, datetime.datetime(2012, 5, 10, 5, 0, 0))
+                        self.assertEqual(review_time, datetime.datetime(2012, 5, 10, 5, 0, rtime))
                         self.assertEqual(by, Person.objects.get(user__username="iana"))
                         self.assertTrue("there are no IANA Actions" in comment.replace("\n", ""))