pcb segfaults when attempting postscript export

[ronald miloh alexander]

a pcb layout I'm working on segmentation faults when exporting to postscript.

I'm going to spend some time manually reverting changes to the layout and run a diff to find out what is causing the issue

[ronald miloh alexander (miloh)]

eps export works.

[Bert Timmerman (bert-timmerman)]

Hi Ronald,

I cannot reproduce this bug.

Could you please give instructions how to reproduce this segmentation fault.

Mentioning of the PCB version and Operating System (including version and patch level) you used would help.

A minimal pcb file giving the segfault would help too.

Kind regards,

Bert Timmerman.

[Steve Morphet (s-steve)]

I'm not the original reporter, but here's an example file that causes behaviour similar to that described. A small polygon is completely covered by the width/clearance of a line. Attempting to export to PS causes a segmentation fault.

PCB version is 20140316. OS is Ubuntu 12.04.

[stan katz (stan-katz-hk)]

As per request by miloh, I have attached my pcb file that segfaults in 1.9.1 and 20140316 when exporting to Postscript on my Debian Jessie amd64 system. This bug has remained on my system through at least 3 Jessie distribution updates.

[DJ Delorie (djdelorie)]

Reproduced. poly_callback() in draw.c is not checking for polygons "cleard out of existence" (and thus have Clipped==NULL). Not sure where the error is - trying to draw those polygons, or the fact that one of them has Clipped==NULL.

[Traumflug (mah-jump-ing)]

This was apparently fixed with the fix for bug 1339383:

http://git.geda-project.org/pcb/commit/?id=8e65f3bfa746d11e61b07aba8461a03c1f77d213

[Traumflug (mah-jump-ing)]

I mean, I can confirm the segfault before this commit and no longer experience it after.

[Chad Parker (parker-charles)]

marking fixed