You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@openoffice.apache.org by bu...@apache.org on 2023/12/31 18:05:40 UTC
[Issue 128579] 32-bit editengine regression causes a crash when opening PPT files
https://bz.apache.org/ooo/show_bug.cgi?id=128579
--- Comment #2 from damjan@apache.org ---
So I tried building and testing various commits in that range of 32 commits
that caused this regression:
Commit Builds? Works?
36a464621660bc9c55b210922d14dc3735a9077e
9ec87cd3b34edb40cfe3e776689fb643e6988567
8ee1071cd11ad3f8718884f8c00e3b804d0af3b1
2cdd1ab0954d1a9988859888d6f6959a67d65693
c2eaa082a297a157b60dd87990a41542135872c8
9d9bf804fbab7089311c26afdf81e1d8d28d853b
0f74efc0ec8e3d3afd0cecccb02fbb5c870cb9f2
5707f3c20a9e1a1c755e034eba60a84635e66cc1 FALSE
79496c0e51bec39643863c32709ffb17c141f62f
7a5aa0144b33f34e6026854c21acbf8d7489983a
ced59291f7a8e2ed44ad73b26bcdef75827bc7b2
86af277237f2917ff5a5d13dc3647298c7d998f6
fbcf52f4d5facbb0884558ee718f6746df6bc62f
87936a440797b7a985fef974d60f4a48c39b855c
961599b3ba1767fc0f5ca79ba244c90113c3aa24
8f081814dede3b5e8825d4cd6e5c9c363b19f21e
67866f572d8a86e153be747862ec69d5c4477d82 FALSE
d8fd122573a3a07dedb0b01271fbbe9e47077b7b FALSE
3974c0663f06bb0a4e51c69a5d49c61022b8fd5b
209925efc4daf4fbf9be8d0ac9cf3fe7fc360b24
1e6c8db2b969e0b438d749b8ed79f6a08fe75d2c
434f02cad7370eb311370dc8dcc01b377a85f9e8 FALSE
bb99baf43803e8317e9f40ec033cd2f7a36b87e9 FALSE
597c409aade8823e17f55ff8711dc896ad293690 FALSE
20e487bda952ee879baf3ee086928fd08706c564 FALSE
fffd1d5de18ed84b4db5b15a57540d804700e38c FALSE
3155dbc2d6387a8f19e0cf04cd12c3d4501eaef4 FALSE
a38ef00207b4524282a09fe5cd78defe45caf613 FALSE
8dc46c982409f2e0e073dd55d6b2e4241173c266 TRUE FALSE
3228dc3342001440d11c7fb59e0ed697f17dc10f TRUE FALSE
a890f778ee3da0f9a88d833ee3aa73bbd9d6406e TRUE FALSE
765ae79b138b5cc2543282200466b2ef4be7be8e TRUE TRUE
Most don't build, but by some miracle, out of the first 4 that did, it was
definitely commit a890f778ee3da0f9a88d833ee3aa73bbd9d6406e that caused this
regression.
Probably the changed return values from ParaPortionList::FindParagraph()
(0xFFFF to SAL_MAX_UINT32) or FastGetPos() (USHRT_MAX to SAL_MAX_UINT32) are
not checked properly by some caller, resulting in it not detecting the element
does not exist, and then trying to access an invalid element (possibly NULL?)
leads to the crash.
--
You are receiving this mail because:
You are the assignee for the issue.