This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.

Bug 37850 - Mailing list archives are not Unicode-aware
Summary: Mailing list archives are not Unicode-aware
Status: RESOLVED INVALID
Alias: None
Product: obsolete
Classification: Unclassified
Component: collabnet (show other bugs)
Version: 3.x
Hardware: All All
: P3 blocker (vote)
Assignee: support
URL: http://www.netbeans.org/servlets/Brow...
Keywords: I18N
Depends on:
Blocks:
 
Reported: 2003-12-08 17:18 UTC by Jesse Glick
Modified: 2009-11-08 02:32 UTC (History)
1 user (show)

See Also:
Issue Type: DEFECT
Exception Reporter:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Jesse Glick 2003-12-08 17:18:50 UTC
Check out the above link, showing archives of the
Chinese discussion list. You can see that the
subject lines are garbled, displaying the raw
ASCII of the Subject header. If you open the
message, e.g.

http://www.netbeans.org/servlets/ReadMsg?msgId=636212&listName=nbdiscuss_zh

then the contents are garbled too. Here the
character encoding GB2312 was specified for both
the subject line (note the RFC-compliant
"=?GB2312?Q?"...."?=" marker for this purpose) and
the body text (note Content-Type header). Yet the
archive page shows garbage.

The reader servlet should set UTF-8 encoding on
the page it returns, interpret encoding markers in
both the message body(ies) using Content-Type as
well as the message headers using the RFC designed
for this purpose (don't have RFC# offhand),
internally load everything in Unicode (it is a
Java servlet after all! this should not be hard)
and print it all in UTF-8 for correct display in
any Unicode-enabled browser.
Comment 1 Unknown 2003-12-09 06:26:53 UTC
Verified the issue and filed an internal issue 24605 for the 
engineers to look at. Will update here as soon as i get the update 
from them.

Thanks,
priya
Comment 2 Unknown 2004-01-06 14:07:42 UTC
Engineer updated that this is fixed in 2.6.x -- that version is  
properly unicode aware.  Once the upgrade is done , new messages will 
be handled correctly. 

- Priya

Comment 3 Unknown 2004-02-05 07:30:56 UTC
As this is fixed in 2.6.x, an upgrade version for NB, closing this as 
resolved and review again once the upgrade is done.
Comment 4 padmar 2006-10-17 18:11:54 UTC
The subject is still garbled. The body looks fine. 

Reopening this issue
Comment 5 Unknown 2006-11-06 10:13:51 UTC
Followin up this internally. 
Comment 6 Unknown 2006-11-15 12:38:19 UTC
The message in the attachment is dated in 2003. Existing messages will require 
re-indexing or re-archiving which is a pretty expensive task. New messages 
will archive and index properly. 

If you can find a recent message with an improperly handled subject, that 
should help us to check whether this is still an issue. If you need the old 
messages to be indexed to find a time when we can reindex their mailing lists. 
But to re-archive the entire site, which might take quite some time in NB and 
it is an expensive task. Let us know your feedback on it. 

-Priya
Comment 7 Unknown 2006-11-16 06:47:02 UTC
We have an option to index a few old messages by setting up indexer queue. 
This will index the message in the queue during the incremental rebuild. Still 
i think there will be quite a lot of messages existing. As mentioned above, re-
archiving is an expensive task. 
Comment 8 Unknown 2006-11-21 09:09:15 UTC
jglick: Hi, any updates on how do you want to go furhter on this? 
Comment 9 Jesse Glick 2006-11-21 18:14:03 UTC
I don't think we care much about existing messages, if it's fixed for new messages.
Comment 10 Unknown 2006-11-22 06:51:04 UTC
Yup, The new messages are fixed, if you happen to see any message with a 
garbled characters it could be a old one. If you happen to see any new 
messages with this kind please reopen the issue. 
Comment 11 Unknown 2007-04-03 15:26:57 UTC
closing..
Comment 12 Marian Mirilovic 2009-11-08 02:32:59 UTC
We recently moved out from Collabnet's infrastructure