After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 337317 - Non-ascii characters are not always handled gracefully
Non-ascii characters are not always handled gracefully
Status: RESOLVED FIXED
Product: Pan
Classification: Other
Component: general
pre-1.0 betas
Other Linux
: Normal normal
: 1.0
Assigned To: Charles Kerr
Pan QA Team
Depends on:
Blocks:
 
 
Reported: 2006-04-05 07:23 UTC by Søren Boll Overgaard
Modified: 2006-07-06 21:47 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
Image showing how a portion of the message body gets lost (81.05 KB, image/jpeg)
2006-04-05 07:25 UTC, Søren Boll Overgaard
  Details
possible patch to fix the encoding bug (2.85 KB, patch)
2006-04-05 18:08 UTC, Charles Kerr
committed Details | Review
patched version of utf8-utils.cc (8.37 KB, text/plain)
2006-04-05 19:39 UTC, Charles Kerr
  Details
cached copy of the message cited above (921 bytes, text/plain)
2006-07-06 21:07 UTC, Charles Kerr
  Details
patch against 0.102 that fixes the test case. (1.33 KB, patch)
2006-07-06 21:43 UTC, Charles Kerr
none Details | Review

Description Søren Boll Overgaard 2006-04-05 07:23:56 UTC
On the server news.daimi.au.dk (which has global read access), the following message ids are available:

1) Message-ID: <dv8gfk$oeo$1@news.net.uni-c.dk>
2) Message-ID: <wvxirqhxi5n.fsf@hadrian.daimi.au.dk>

In the case of 1), non-ascii characters are displayed correctly in the body pane, whereas in 2) the first non-ascii character in the body causes the rest of the body to get lost.

This is possibly due to the fact that 2) does not have a Content-Type header, but I can't be sure.

This is illustrated in http://gallery.overgaard.org/misc/pan-0.91-error-1.jpg
Comment 1 Søren Boll Overgaard 2006-04-05 07:25:11 UTC
Created attachment 62778 [details]
Image showing how a portion of the message body gets lost
Comment 2 Charles Kerr 2006-04-05 18:08:22 UTC
Created attachment 62810 [details] [review]
possible patch to fix the encoding bug
Comment 3 Charles Kerr 2006-04-05 18:09:31 UTC
Søren, could you test out the patch in comment #2?
Comment 4 Søren Boll Overgaard 2006-04-05 18:32:46 UTC
Hi Charles,

I'm having trouble with the patch:

boll@andariel [sid]:~/sandbox/pan/pan-0.91/pan/usenet-utils$ patch < ~/sandbox/pan/encoding-patch.txt 
patching file utf8-utils.cc
Hunk #3 FAILED at 252.
1 out of 3 hunks FAILED -- saving rejects to file utf8-utils.cc.rej
boll@andariel [sid]:~/sandbox/pan/pan-0.91/pan/usenet-utils$ 

Was it possibly not made against 0.91, or did I do something completely wrong?
Comment 5 Charles Kerr 2006-04-05 19:37:41 UTC
You're probably right.  Here's the whole file.
Comment 6 Charles Kerr 2006-04-05 19:39:02 UTC
Created attachment 62820 [details]
patched version of utf8-utils.cc
Comment 7 Søren Boll Overgaard 2006-04-05 20:26:12 UTC
Ok, i've recompiled it with your patch.
It no longer chokes on non-ascii characters, although it doesn't actually display them. Nothing is displayed in their place, the text simply continues.
If you want, I will be happy to provide a screenshot.
Comment 8 Charles Kerr 2006-04-05 20:31:09 UTC
grumble. :)

What group is <wvxirqhxi5n.fsf@hadrian.daimi.au.dk> in?
Comment 9 Søren Boll Overgaard 2006-04-05 20:51:44 UTC
daimi.test (an internal group) on news.daimi.au.dk.
You may not have read access to that server.

Another message which triggers the error is <44314881$0$47005$edfadb0f@dread15.news.tele.dk> which is in the group dk.edb.programmering. That should be available on most news servers.
Comment 10 Charles Kerr 2006-07-06 21:07:36 UTC
Created attachment 68497 [details]
cached copy of the message cited above

This is a copy of the message mentioned above. As of 0.102, Pan is
trimming "nævnt" to "nvnt".
Comment 11 Charles Kerr 2006-07-06 21:43:52 UTC
Created attachment 68501 [details] [review]
patch against 0.102 that fixes the test case.

This patch makes the test case work right and will be in 0.103.
If you find any problems with this, feel free to reopen this
ticket and let me know.  Thank you!