Re: merging imread and imwrite

[Thomas Treichl]

John W. Eaton schrieb:
> On  1-Feb-2008, Thomas Treichl wrote:
>
> | However, I still think that all backends of ImageMagick are not needed for an 
> | standalone Octave pack and some compromise about which backend should be added 
> | and which one should be left away must be done (eg. ?jbig, lcms, lqr, openexr? - 
> | I actually have never heard of them before - are they needed?)...
>
> Probably not.  But isn't that up to the person building/packaging the
> Magick++ library?
>
> jwe

Yes sure. I misunderstood some things before -- until you wrote in another email 
'We can use Magick++ to read png and jpeg files, though I think it ultimately 
uses libpng and libjpeg to actually read them'.

What I was thinking about (without still having a closer look into the Magick++ 
codes) is that there exists eg. exactly one/two higher level C/C++ functions in 
the ImageMagick library which loads/saves images of any type and that Octave's 
'imread' and 'imwrite' functions then also should read/write images from any 
type (currently >100 file formats if ImageMagick is used and a huge number of 
external libs is needed to support all these types, starting somewhere with a 
needed ghostscript library and ending with openexr -- and I expect an equivalent 
number of file formats and dependencies for GraphicsMagick).

If so, I wouldn't solve all these dependencies for Octave.app because of an 
extreme increase of Octave.app's size but would have suggested that users better 
get a GNU Image Manipulation Program or an ImageMagick/GraphicsMagick binary 
which both are available even on nearly every other platform and then convert 
the image file into one of three/four file formats before it can be used with 
Octave's 'imread' and 'imwrite' functions.

I hope it's clearer now why I was asking these questions...

  Thomas