[Anjuta-list] Re: [Anjuta-devel] [Help Wanted] Recent stability issues with Anjuta 1.x

[Jens Georg <mail jensgeorg de>]

On Thu, 24 Jan 2002 09:53:15 -0800 (PST)
"Naba Kumar" <kh_naba yahoo com> wrote:

> 
> --- Jens Georg <mail jensgeorg de> wrote:
> > On 23 Jan 2002 11:50:30 +0530
> > 
> > ok, solved the problem for this tarball, patch is attached.
> > 
> > it was an array-out-of-bounds-problem - biswa, check if i modified
> > something in a way that shouldn't be. 
> > 
> > Maybe gcc 2.96 and 3.0 are better at handling array out of bounds -
> > things
> 
> Thank you Jens. That was really great.
> 
> > 
> > it does not work with current cvs... i'll track that down now...
> 
> Does this mean we may still have similar bug around? Will Biswa's

no - as johannes said it's all the same error, in cvs and biswa's tarball.
in cvs, it only occurs more often, while biswa's patch will be much more
cleaner!

i suggest to include biswa's fixed patch (with right array bounds) - that
shold solve all problems.

biswa, you sould have a look at the statements like

root[sv_root_max_t-1] = NULL 

and so on. I'm not quite sure, but are they really neccessary after all?
I suspect they may overrite the last added element (there are some
for-loops that look suspiciosly).

-- 
 * Origin:                                        http://www.jensgeorg.de