Patches item #670167, was opened at 2003-01-18 05:58 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=670167&group_id=103 Category: Pipermail Group: Mailman 2.1 Status: Open Resolution: Accepted
Priority: 7 Submitted By: Tokio Kikuchi (tkikuchi) Assigned to: Nobody/Anonymous (nobody) Summary: fix i18n attachment description in archive
Initial Comment: Bug-ID 669081 (Major attachment handling bug!) is caused inaccurate handling of charset in Scrubber.py. This is caused by misuse of email functions (or by bug in email package). In Scrubber.py, attempt is made to get message charset by part.get_charset() but it returns only None, as far as I experimented. In this patch, message charset is extracted from regular expression pattern in content-type header. This also set the charset of internally crafted message part. ---------------------------------------------------------------------- Comment By: Tokio Kikuchi (tkikuchi) Date: 2003-02-24 04:34 Message: Logged In: YES user_id=67709 Uploading revised patch to fix more error. Sorry for the people downloaded earlier. ---------------------------------------------------------------------- Comment By: Tokio Kikuchi (tkikuchi) Date: 2003-02-24 01:39 Message: Logged In: YES user_id=67709 update my patch to fix error reported by sumeet File "/list/Mailman/Handlers/Scrubber.py", line 320, in process if not t.endswith('\n'): AttributeError: 'NoneType' object has no attribute 'endswith' ---------------------------------------------------------------------- Comment By: Tokio Kikuchi (tkikuchi) Date: 2003-02-22 07:05 Message: Logged In: YES user_id=67709 This patch for Scrubber.py now include Bug-ID #691039. save_attachment() now returns (url, size), because in previous code, get_paylaod() is called before calling save_attachment() only to get the size. save_attachment() calls get_payload() inside. We might be able to discard the attachment if get_payload(decode=1) fail. (TBD: return discard message and size 0 ?) ---------------------------------------------------------------------- Comment By: Tokio Kikuchi (tkikuchi) Date: 2003-02-20 01:31 Message: Logged In: YES user_id=67709 Increment priority because this is now a bug fix. ---------------------------------------------------------------------- Comment By: Tokio Kikuchi (tkikuchi) Date: 2003-02-04 06:08 Message: Logged In: YES user_id=67709 I made a new patch to fix embedded multipart messages are stripped bug. this patch is for CVS. see http://mail.python.org/pipermail/mailman-users/2003-February/026071.html ---------------------------------------------------------------------- Comment By: Barry A. Warsaw (bwarsaw) Date: 2003-01-20 05:13 Message: Logged In: YES user_id=12800 Reopening so this issue doesn't get lost. ---------------------------------------------------------------------- Comment By: Tokio Kikuchi (tkikuchi) Date: 2003-01-20 04:03 Message: Logged In: YES user_id=67709 We, Japanese, need additional process of stringify and regenerate message instance.. Internal message is in 'euc-jp' but return value of get_content_charset() becomes 'iso-2022-jp' even though the get_payload() returns euc-jp text. The message and part body is converted to iso-2022-jp when it is output to SMTP (or whatever accepts in string format). ---------------------------------------------------------------------- Comment By: Barry A. Warsaw (bwarsaw) Date: 2003-01-20 02:18 Message: Logged In: YES user_id=12800 Accepted, with changes. ---------------------------------------------------------------------- Comment By: Tokio Kikuchi (tkikuchi) Date: 2003-01-19 00:52 Message: Logged In: YES user_id=67709 Update of patch. Please backout the first one if you have already applied. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=670167&group_id=103