[gl2ps] Memory Leak

Christophe Geuzaine cag32 at case.edu
Thu Nov 9 14:22:50 CET 2006


Micha Bieber wrote:
>> Hi Jason - Indeed, looks looks like a memory leak to me too. And I
>> cannot remember why we need to make a deep copy of the primitives in 
>> this pdfprimlist in the first place... Micha, any insights?
> 
> I haven't the newest version here, but older ones (cf. v1.1.2) still
> contain somewhat similar code (like the fixed example, provided from
> Jason). So, someone did premature optimizations along the way :->
> 

Looking back at the cvs history I think I'm the one to blame :-)

(The deallocation was never included, so I must have applied your PDF 
transparency patch by hand and missed that bit.)

In any case, why *do* we need this extra pdfprimlist? From what I see in 
the code, it's only use is to have a full copy of the original primitive 
list, minus the culled primitives... Is there something else I'm missing?





> Micha  


-- 
Christophe Geuzaine
Assistant Professor, Case Western Reserve University, Mathematics
http://www.case.edu/artsci/math/geuzaine