Adobe Reader plugin crashes on IEEE Xplore pdf
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Pipelight |
Fix Released
|
Medium
|
Unassigned |
Bug Description
The adobe reader plugin is working great in every single website I've used it, except for IEEE Xplore. The website is known for handling pdf in a weird way.
Every time I try to access an article (pdf) I see the pdf being loaded, a progress bar, and when it's done I get a crash message and nothing works. I'm uploading the tracelog I got from the crash.
I know that this is going to be a hard bug to replicate if you don't have access to IEEE Xplore. If you have access to a university library or a university network, you'll be able to access IEEE Xplore, and just doing any search an opening any pdf will give you the error.
I can see under plugins on Firefox that the Adobe Acrobat is working and this is what i get:
File: libpipelight-
Path: /usr/lib/
Version: Adobe PDF Plug-In For Firefox and Netscape 11.0.06
State: Enabled
Adobe PDF Plug-In For Firefox and Netscape 11.0.06
I'm using Ubuntu 13.10 64bit with Firefox 28.0 from the official repositories.
I'm also using Pipelight 0.2.6~ubuntu13.
Changed in pipelight: | |
status: | Triaged → Incomplete |
Hi,
I didn't had time yet to do a in depth analysis, but i would guess that the crash is related to the richedit warnings which are shown before the crash:
fixme:richedit: ITextDocument_ fnRange stub 0x8538a8 ITextDocument_ fnFreeze stub 0x8538a8 ITextDocument_ fnRange stub 0x8538a8 ITextSelection_ fnGetFont not implemented ITextSelection_ fnGetFont not implemented
fixme:richedit:
fixme:richedit:
fixme:richedit:
fixme:richedit:
wine: Unhandled page fault on read access to 0x00000000 at address 0x601a8466 (thread 0024), starting debugger...
The function ITextSelection_ fnGetFont is supposed to return a pointer to an Interface and it is only a stub in Wine ( http:// source. winehq. org/source/ dlls/riched20/ richole. c#L893 ) and does not set any value for this parameter. If the Adobe Reader tries to dereference the pointer without checking the return value of the function this would result in an invalid read access. I need to verify this theory later, but I fear that there is no easy way to fix this since neither the ITextDocument nor the ITextSelection interface is correctly implemented for the richedit control in Wine.
Michael