octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Bugs involving imread, imwrite, and GraphicsMagick


From: John Swensen
Subject: Re: Bugs involving imread, imwrite, and GraphicsMagick
Date: Tue, 21 Sep 2010 07:32:47 -0400

On Sep 21, 2010, at 7:25 AM, John Swensen wrote:

> 
> On Sep 20, 2010, at 11:48 PM, John W. Eaton wrote:
> 
>> On 20-Sep-2010, John Swensen wrote:
>> 
>> | I guess that was kindof my point.  We, as the Octave community,
>> | don't have much control over what the GraphicsMagick maintainers do.
>> | We can ask nicely and sit and wait for them to change a configure
>> | flag in the GraphicsMagick package builds.  On the other hand, we
>> | can easily ask Thomas Weber to add a "--with-magick=ImageMagick" to
>> | the Debian/Ubuntu package builds and have a great likelihood of it
>> | happening pretty quickly.  Even one of us could make a patch to the
>> | Debian package files so all he has to do is make it effective for
>> | the next package release.
>> 
>> If it is just a simple matter of different library/header file names
>> and no other code changes to enable using either GraphicsMagick or
>> ImageMagick, then let's do that part first.  If you have diffs (even
>> incomplete) at this point, send them to me and I'll do the configure
>> script hacking.
>> 
>> jwe
> 
> Included is a patchset that modifies configure.ac to add the functionality.  
> I was able to configure Octave with
> 
> ./configure <OSX specific configure options>
> ./configure --with-magick=GraphicsMagick <OSX specific configure options>
> ./configure --with-magick=ImageMagick <OSX specific configure options>
> 
> and have it compile and work fine.  The first two configure lines above used 
> GraphicsMagick and the third used ImageMagick, as desired.  That being said, 
> I have only tested it on OSX, but don't see any hangups for other platforms 
> as it uses pkg-config to find libraries and retrieve compiler and linker 
> flags.
> 
> John Swensen
> 
> <MagickChoice.diff>

Sorry I didn't think of this before submitting this patch, but maybe we should 
also add a check for the bit depth when using GraphicsMagick by parsing the 
results of 'gm -version' and displaying  a warning to the user as the end of 
the configure process if it is 8 bits, like is done when FLTK and other 
libraries are missing.

John Swensen


reply via email to

[Prev in Thread] Current Thread [Next in Thread]