From 9bdb4029f331ea6230c668e408c2f694491ec6e4 Mon Sep 17 00:00:00 2001
From: tkikuchi <>
Date: Thu, 14 Oct 2004 08:05:06 +0000
Subject: fix bugs when using ARCHIVE_HTML_SANITIZER = 1

---
 Mailman/Archiver/pipermail.py | 3 ++-
 Mailman/Handlers/Scrubber.py  | 5 +++--
 Mailman/Handlers/ToDigest.py  | 3 ++-
 3 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/Mailman/Archiver/pipermail.py b/Mailman/Archiver/pipermail.py
index ec8e46f6..44dce52c 100644
--- a/Mailman/Archiver/pipermail.py
+++ b/Mailman/Archiver/pipermail.py
@@ -217,7 +217,8 @@ class Article:
                 self.headers[i] = message[i]
 
         # Read the message body
-        s = StringIO(message.get_payload())
+        s = StringIO(message.get_payload(decode=1)\
+                     or message.as_string().split('\n\n',1)[1])
         self.body = s.readlines()
 
     def _set_date(self, message):
diff --git a/Mailman/Handlers/Scrubber.py b/Mailman/Handlers/Scrubber.py
index 7429c0b4..6116806b 100644
--- a/Mailman/Handlers/Scrubber.py
+++ b/Mailman/Handlers/Scrubber.py
@@ -309,7 +309,7 @@ Url : %(url)s
     # We still have to sanitize multipart messages to flat text because
     # Pipermail can't handle messages with list payloads.  This is a kludge;
     # def (n) clever hack ;).
-    if msg.is_multipart():
+    if msg.is_multipart() and sanitize <> 2:
         # By default we take the charset of the first text/plain part in the
         # message, but if there was none, we'll use the list's preferred
         # language's charset.
@@ -402,7 +402,8 @@ def save_attachment(mlist, msg, dir, filter_html=True):
     # For safety, we should confirm this is valid ext for content-type
     # but we can use fnext if we introduce fnext filtering
     if mm_cfg.SCRUBBER_USE_ATTACHMENT_FILENAME_EXTENSION:
-        ext = fnext
+        # HTML message doesn't have filename :-(
+        ext = fnext or guess_extension(ctype, fnext)
     else:
         ext = guess_extension(ctype, fnext)
     if not ext:
diff --git a/Mailman/Handlers/ToDigest.py b/Mailman/Handlers/ToDigest.py
index f350768e..2f64b369 100644
--- a/Mailman/Handlers/ToDigest.py
+++ b/Mailman/Handlers/ToDigest.py
@@ -314,7 +314,8 @@ def send_i18n_digests(mlist, mboxfp):
                 uh = '\n\t'.join(uh.split('\n'))
                 print >> plainmsg, uh
         print >> plainmsg
-        payload = msg.get_payload(decode=True)
+        payload = msg.get_payload(decode=True)\
+                  or msg.as_string().split('\n\n',1)[1]
         print >> plainmsg, payload
         if not payload.endswith('\n'):
             print >> plainmsg
-- 
cgit v1.2.3