Re: [Lasi-devel] Report Cygwin, MingW, Visual C++ 2005

[Alan W. I. <ir...@be...>]

On 2007-02-15 09:34-0800 Alan W. Irwin wrote:

> The remaining segfaults you are getting on Cygwin are disappointing, but
> probably expected.  My experience tells me that severe memory management
> problems (as discovered by valgrind for all our examples for modern
> libraries) tend to be associated with segfaults.  Because of these memory
> management problems I think we are just lucky not to experience segfaults on
> other platforms as well.  Also, I believe you had similar luck with example0
> on Cygwin.  Note, the severe memory management problems reported by valgrind
> occur regardless of whether glyphs are missing or not.

I have looked a bit more at the memory management issues for systems with
modern pango libraries as reported by valgrind.  _Every single one_ of these
issues is caused by the single call to pango_itemize in psDoc.cpp.  I think
that is a pretty important clue so I have been looking for some
incompatibility between the argument list for this function in psDoc.cpp and
the modern library documentation of pango_itemize in
http://developer.gnome.org/doc/API/2.0/pango/pango-Text-Processing.html#id2524163.

My C++ skills are weak so could somebody look carefully at pangoContext() to
make absolutely sure it sets up the first argument of pango_itemize to be
consistent with what is expected by the modern pango library?

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________