Tom & Group, Having just gotten lost crawling through the code, I concur that pitching IM in favor of GM is the way to go.
73 Dave KB3EFS FN24BI81GP On Fri, Jan 1, 2010 at 12:33 PM, Tom Russo <ru...@bogodyn.org> wrote: > After a recent upgrade of ImageMagick, many new "deprecated" warnings have > been showing up. Even the most basic function, "InitializeMagick," is now > a deprecated part of the API. > > Maintaining support in Xastir for all the various versions of ImageMagick that > have broken API time and again has lead to code that is unbelievably ugly. > Throughout map_geo.c there are tests of the preprocessor symbol > MagickLibVersion > where we check to see which broken API we have to deal with, and then do > different things depending on how old or new ImageMagick is. > > Now, some distros are starting to ship ImageMagick with the "HDRI" > (High Dynamic Range Image) option turned on. Even the ImageMagick > documentation > calls this an "experimental" feature. It causes the data type of the Quantum > (pixel value) to be a double instead of an integer, breaking our bit-fiddling > and requiring that we hack in additional special cases --- special cases > that will almost certainly be broken again when the "experiment" calls for > a change in API. > > I say it's time that we declare ImageMagick too unstable an external library > for us to support. GraphicsMagick, the stable fork of ImageMagick, has been > much better and provides all the features we actually use. Now that the > 8-bit quantum bug in Xastir is fixed, GraphicsMagick is a viable alternative > for all platforms. > > I propose that following our next stable release (which should happen very > soon), support for ImageMagick be dropped completely. The code for > maintaining > backward compatibility with all old versions of ImageMagick would be removed. > Configure macros for probing ImageMagick would be removed. All features > currently requiring either ImageMagick or GraphicsMagick would change to > requiring GraphicsMagick. > > I think it is high time that we did this, as ImageMagick has been a thorn in > our sides ever since it was introduced as a dependency for Xastir. > > Ideally, even GraphicsMagick should not be required --- what we use of that > library is so small a set that there should be something less elaborate that > would serve the purpose. But for now, GraphicsMagick is at least stable and > functional, and doesn't require bizarre coding circumlocutions just to keep > it going. > > Discussion? > -- > Tom Russo KM5VY SAR502 DM64ux http://www.swcp.com/~russo/ > Tijeras, NM QRPL#1592 K2#398 SOC#236 http://kevan.org/brain.cgi?DDTNM > In some cultures what I do would be considered normal. > -- Ineffective daily affirmation > > _______________________________________________ > Xastir mailing list > Xastir@lists.xastir.org > http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir > _______________________________________________ Xastir mailing list Xastir@lists.xastir.org http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir
