Re: imagemmagick patch 5

[Stefan Monnier]

> The Imagemagick patch allows Emacs to use the Imagemagick libraries to
> load images. This support can be used in parallell with the existing
> image loading support. There is support for asking your imagemagick
> installation which image types it supports, and registering them in
> Emacs selectively.

Thanks.  Looks pretty good.  Feel free to install it in the `pending'
branch, but please see the comments below first, and don't forget to add
a good NEWS entry when you install this change.

> +HAVE_IMAGEMAGICK=no
> +if test "${HAVE_X11}" = "yes" ; then

Do I understand it right that this X11-only restriction could be lifted
at some point in the future?
 
> diff --git a/src/dbusbind.c b/src/dbusbind.c
> index 7c0be49..7a47730 100644
> --- a/src/dbusbind.c
> +++ b/src/dbusbind.c
> @@ -773,6 +773,7 @@ xd_add_watch (watch, data)
>        if (fd == -1)
>       return FALSE;
>  
> +
>        /* Add the file descriptor to input_wait_mask.  */
>        add_keyboard_wait_descriptor (fd);
>      }

Please drop this gratuitous change.

> +/***********************************************************************
> +                              imagemagick
> + ***********************************************************************/
> +#if defined (HAVE_IMAGEMAGICK)
> +Lisp_Object Vimagemagick_render_type;
> +/* Function prototypes.  */
> +
> +static int imagemagick_image_p P_ ((Lisp_Object object));
> +static int imagemagick_load P_ ((struct frame *f, struct image *img));
> +
> +static int imagemagick_load_image P_ ((struct frame *, struct image *,
> +                                       unsigned char *, unsigned int, 
> unsigned char *));

Please don't use the P_ macro in new code: just use prototypes.

> +//#include 
> "/home/joakim/current/unison/data/ImageMagick-6.5.4-7/magick/xwindow-private.h"

This should be removed as well.

> +    //try if magicexportimage is any faster than pixelpushing
[...]
> +    //oddly, the below code doesnt seem to work:

We currently only use /*...*/ comments, so I'd rather we don't introduce
// comments for now.


        Stefan