Re: building from src, linking libpng

[email protected] Wed, 29 Jun 2011 23:18:55 +0000
Newsgroups gmane.comp.video.graphicsmagick.apis
Message-ID <[email protected]>
On Wed, Jun 29, 2011 at 11:09:03PM +0000, [email protected] wrote:
> On Wed, Jun 29, 2011 at 05:17:49PM -0500, Bob Friesenhahn wrote:
> > On Wed, 29 Jun 2011, [email protected] wrote:
> >
> >>
> >> I'm trying to build GraphicsMagick from src w/o X11 support, but with png support. The ./configure process tests for libpng library and headers (finds successfully), then does a link test trying to call png_get_io_ptr().
> >>
> >> I've run nm(1) on libpng.so, and it says there -is- a png_get_io_ptr(), but the test is still failing:
> >
> > What type of system are you using?
> >
> > What version of libpng is being used?
> >
> > What version of GraphicsMagick are you building?
> >
> > Please note that libpng 1.5 has a modified API.  Only the  
> > development/snapshot version of GraphicsMagick builds with libpng 1.5.
> 
> Can you explain what the sticking points are? I've got my own dev system
> running GM1.3.12 linked against libpng 1.5.2. I haven't worked .png
> support heavily, but I don't think I've run into any issues yet that
> weren't my own fault. If there's a serious problem that is possible,
> I'll work to get NetBSDs pkgsrc smartened-up, because this is our default
> build atm :P

Is this what you'd expect:
coders/png.c: In function 'PNGErrorHandler':
coders/png.c:1354: error: dereferencing pointer to incomplete type
coders/png.c: In function 'ReadOnePNGImage':
coders/png.c:1668: error: dereferencing pointer to incomplete type
coders/png.c:1752: error: dereferencing pointer to incomplete type
coders/png.c:1754: error: dereferencing pointer to incomplete type
coders/png.c:1764: error: dereferencing pointer to incomplete type
...


In that case, the pkgsrc installation has a modified copy of coders/png.c,
which explains what is otherwise an odd-looking installation...


> -bch
> 
> > Sometimes it helps to inspect the config.log file to see what error  
> > message was produced by the compiler when the configure test failed.  
> > Often the error message can be quite helpful to understand the failure.
> >
> > Bob
> > -- 
> > Bob Friesenhahn
> > [email protected], http://www.simplesystems.org/users/bfriesen/
> > GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/
> 
> -- 
> Brad Harder
> Method Logic Digital Consulting
> http://methodlogic.net/
> http://twitter.com/bcharder
> 
> 
> ------------------------------------------------------------------------------
> All of the data generated in your IT infrastructure is seriously valuable.
> Why? It contains a definitive record of application performance, security 
> threats, fraudulent activity, and more. Splunk takes this data and makes 
> sense of it. IT sense. And common sense.
> http://p.sf.net/sfu/splunk-d2d-c2
> _______________________________________________
> Graphicsmagick-apis mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/graphicsmagick-apis

-- 
Brad Harder
Method Logic Digital Consulting
http://methodlogic.net/
http://twitter.com/bcharder


------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security 
threats, fraudulent activity, and more. Splunk takes this data and makes 
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2d-c2