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

Re: Bug#172158: ITP: rsxs -- Really Slick X Screensavers

On Sun, Dec 08, 2002 at 04:49:10AM -0500, Ari Pollak wrote:

 > I suppose it would be possible to just include the binaries without
 > touching the XScreenSaver config file but instead including
 > instructions on including it on a user-specific basis, but I don't
 > know how well the changes would transfer between XScreensaver
 > versions.

 Simplest hack I can think of: modify xscreensaver to be able to include
 configuration files from the main configuration file and use menu to
 register new screensavers.  Something like:

    ?package(rss):needs=xscreensaver \
            section="Come up with something for this field" \
            title="Sky Rocket" \
            visual="GL" \

 and there you go.  You write a menu-method, run menu and you have an
 up-to-date list of installed screensavers.

 The problem is basically xscreensaver's configuration file.  I have
 considered a number of options to solve this.  For example, you could
 modify xscreensaver to read "info files" for each screensaver.  In
 those info files it would say, for example, if the program requires a
 GL visual or not.  But they could be more flexible than that.  They
 could document options, which could in turn be used for the
 configuration part of xscreensaver.  xscreensaver would then store the
 list of "active" screensavers, instead of that ugly block in the config
 file which mixes state with options with visuals with ...

 It's dormant project of mine to take xscreensaver appart and write a
 library that people can use to write screensavers for xscreensaver (and
 whatever else -- but I'm only interested in xscreensaver really).
 xscreensaver has some requirements (specific command line options,
 specific behaviour regarding visuals, ...) and the code is _horribly_
 duplicated across screensavers.  The GL screensavers in particular
 reinvent a (non-trivial to some degree) wheel way too often.

 But this will remain dormant until ... well, I don't know for how long,
 really.  Perhaps someone will read this and go ahead with it :-)


Reply to: