Subject: Re: patch -- fix for Bug 760 [and ChangeLog question]
From: Martin Sevior (msevior@mccubbin.ph.unimelb.edu.au)
Date: Mon May 22 2000 - 01:37:04 CDT
On Mon, 22 May 2000, sam th wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Mon, 22 May 2000, Jesper Skov wrote:
>
> > >>>>> "sam" == sam th <sam@bur-jud-118-039.rh.uchicago.edu> writes:
> >
> > sam> open a new document. turn on overline (ctrl-t) type a type enter
> > sam> type ctrl-enter (page break) you get 11 of these asserts
> >
> > sam> **** (1) Assert **** **** (1) UT_SHOULD_NOT_HAPPEN at
> > sam> fl_BlockLayout.cpp:993 **** **** (1) Continue ? (y/n) [y] : y
> >
> > These happen because fl_BlockLayout::findPointCoords is called while
> > the only Run on page 2 is a pageBreak which cannot contain the IP.
> >
> > Another way to reproduce the problem (I think it's the same anyway) is
> > to do:
> >
> > 1234 + 2xleft + ctrl-return (+ up)
>
> I don't think this is the same problem. 2 reasons.
>
Hmm I got bored typing "y" to the asserts so I built a non-debug version.
After applying your patch Jesper I get no segfaults trying either method
of generating the bug.
However I had an "old" non-debug binary lying around and I couldn't get
that to segfault either.
I think we should concentrate on eliminating the asserts that arise after
a page break. If a patch removes those we have a good chance of having
solved the problem.
Cheers
Martin
This archive was generated by hypermail 2b25 : Mon May 22 2000 - 01:37:35 CDT