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 523541 - Usage e_cal_component_get_recurid_as_string () leads to leaks
Usage e_cal_component_get_recurid_as_string () leads to leaks
Status: RESOLVED FIXED
Product: evolution-data-server
Classification: Platform
Component: Calendar
unspecified
Other Linux
: Normal normal
: ---
Assigned To: Milan Crha
Evolution QA team
Depends on:
Blocks: 523533
 
 
Reported: 2008-03-20 13:03 UTC by Christian Kellner
Modified: 2008-05-14 07:31 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
proposed evo patch (955 bytes, patch)
2008-03-20 17:18 UTC, Milan Crha
none Details | Review
proposed eds patch (13.76 KB, patch)
2008-03-20 17:21 UTC, Milan Crha
committed Details | Review

Description Christian Kellner 2008-03-20 13:03:35 UTC
The patch of bug #523533 fixes a behavior the to correctly return newly allocated memory in one case. This clearly shows that we are not handling the return value of e_cal_component_get_recurid_as_string () correctly otherwise that would have lead to a lot of crashes. We should check all invocations of e_cal_component_get_recurid_as_string () and see if we handle the memory correctly.
Comment 1 Milan Crha 2008-03-20 17:18:49 UTC
Created attachment 107690 [details] [review]
proposed evo patch

for evolution;

Only one, not bad.
Comment 2 Milan Crha 2008-03-20 17:21:17 UTC
Created attachment 107691 [details] [review]
proposed eds patch

for evolution-data-server;

here it's more. One thing, I removed call in backends/google/e-cal-backend-google.c: (send_object):, it wasn't used there, someone whom knows the code should look and think whether it was intended or not to not use that value.
Comment 3 Srinivasa Ragavan 2008-03-27 06:10:35 UTC
Chen/Gicmo: Can you review/Approve this?
Comment 4 Christian Kellner 2008-03-27 11:10:13 UTC
The first patch (the one for evo) is uneccessary since we are passing the rid variable to the ItipFormatterData struct.
The other one looks good. Go ahead and commit that one.
Comment 5 Milan Crha 2008-03-27 13:13:04 UTC
evo part committed to trunk (slightly modified). Committed revision 35275.
eds part committed to trunk. Committed revision 8592.

We decided on IRC to change the evo patch a bit, even I agree with gicmo, it's kinda for nothing there, just to be sure (I marked patch as obsolete, because not committed exactly that one).
Comment 6 Chenthill P 2008-05-05 14:03:10 UTC
Please commit the patch to stable branch too.
Comment 7 Matthew Barnes 2008-05-06 20:15:40 UTC
Reopening until this is committed to stable.

Chen, you might want to just commit this yourself.  Milan is on vacation for the next couple weeks.
Comment 8 Chenthill P 2008-05-07 04:53:49 UTC
sure, will get it done.
Comment 9 Chenthill P 2008-05-14 07:31:54 UTC
Committed to stable.