python mapscript segfault

[Umberto Nicoletti]

Howard,
I have been a bit imprecise.
Creating and manipulating maps dinamically is mostly safe, with the
exception of some objects and/or usage patterns. I dinamically add
some simple layers myself in an application and it works.

Sorry for the imprecision.

Umberto


On 11/9/06, Howard Butler <hobu at iastate.edu> wrote:
>
> On Nov 9, 2006, at 2:01 AM, Umberto Nicoletti wrote:
>
> > On 11/9/06, John Cartwright <John.C.Cartwright at noaa.gov> wrote:
> >> Hello All,
> >>
> >> I'm trying to use python mapscript (4.10.0) on MacOS 10.4 to create a
> >> map dynamically and output the image.  Code seems to segfault at the
> >> termination of the program. Can anyone see what I'm doing wrong or
> >> suggest a method of tracing the problem?
> >
> > John,
> > you are not doing anything wrong, simply mapsscript doesn't have
> > properly implemented memory management so creating maps dinamically is
> > not supported (even though it could work).
> > I think in your case you should get the image correctly drawn anyway,
> > so you could ignore the (annoying) segfault at program exit.
>
> Umberto,
>
> Creating maps dynamically is the entire point of MapScript. It very
> much supports the concept of creating maps dynamically.   IIRC, the
> code John has posted is right out of an example that Sean and I did
> for a workshop at OSG '05.  It may be that this example is no longer
> valid in 4.10 because of how the memory management has changed, but I
> wouldn't go so far as to say MapScript's memory management is
> outright implemented incorrectly.  There may be some issues, and it
> may be a little rough in areas, but it has been improving.
>
> I suspect that the colorObjs of the styleObjs are not properly
> referencing the styleObj.  Running this script through gdb to get a
> traceback to find out which destructor is being called is what
> probably needs to be done.
>
> Howard
>
>
>
>