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 323888 - gnumeric crashes with attached file
gnumeric crashes with attached file
Status: RESOLVED FIXED
Product: Gnumeric
Classification: Applications
Component: General
git master
Other All
: Normal normal
: ---
Assigned To: Jody Goldberg
Jody Goldberg
Depends on:
Blocks:
 
 
Reported: 2005-12-12 15:53 UTC by Daniel Holbach
Modified: 2005-12-12 18:20 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Daniel Holbach 2005-12-12 15:53:11 UTC
Version details: 1.6.1
Distribution/Version: Ubuntu Dapper

Forwarded from: http://bugzilla.ubuntu.com/show_bug.cgi?id=20860

in dapper, gnumeric 1.6.1-2ubuntu1 crashes while loading the attached file.
(http://bugzilla.ubuntu.com/attachment.cgi?id=5266)

--

Thanks for your bug report. Do you remember, how you did obtain that .gnumeric
file? It's broken XML.

daniel@bert:~$ file latas.gnumeric
latas.gnumeric: gzip compressed data, from Unix
daniel@bert:~$ mv latas.gnumeric latas.gz
daniel@bert:~$ gunzip latas.gz
daniel@bert:~$ less latas
daniel@bert:~$ xmllint latas
latas:2336: parser error : Premature end of data in tag Sheets line 50

^
latas:2336: parser error : Premature end of data in tag Workbook line 2

^
daniel@bert:~$


--


Sure. I created a very simple spreadsheet. And then ran the Solver from the
Tools menu. After the solver finished successfuly it reported several
statistical reports and then I saved the file and quit gnumeric. The file is
unchanged since then (confirmed by the modification date).

----

This bug report actually reveals two bugs. 1) The solver seems to meddle with
the file and produce improper XML. 2) files could be checked for consistency
before. - If you want, I'll file another bug report.
Comment 1 Morten Welinder 2005-12-12 16:08:47 UTC
I can confirm the crash on load.  While that can be fixed, recovery of the
data is going to require manual editing of the xml.  Doesn't look too hard,
for what it is worth.

My guess is that some operation -- perhaps the solver -- is corrupting memory.
Obviously not good.
Comment 2 Morten Welinder 2005-12-12 18:12:03 UTC
cvs head has the load crash fixed and the sheet loads with a few warnings
on stderr.

If we can somehow get something concrete information as to what made the mess
in the first place, please open a new bug for that.
Comment 3 Daniel Holbach 2005-12-12 18:20:08 UTC
Thanks. I told the reporter.