2009/8/30 Jesús Guerrero <[email protected]>: > Hi, > > The fact is that rplay doesn't seem to compile against newer gcc > versions. I never used rplay for anything, and I think I remember > some conversation (maybe in this same list or in workers) about > getting rid of it in fvwm, or maybe it was just a comment, I don't > really remember.
It was I who suggested something along those lines -- but perhaps more with an idea of *replacing* it with something like pulseaudio which has an API anyway, and is currently being championed by more than a few distros at the moment. Sounds on events (which is indeed the *only* reason we link against librplay -- back when it was just FvwmAudio) is akin to the 1995 coolness of "under construction" for webpages; it seems to me a thing of the past --- and whilst that isn't a reason for getting rid of functionality, it was always something to bear in mind -- and I *did* say as much when (IMO, more work than was necessary) FvwmEvent was changed a lot to support different rplay stuff behind the scenes. > There's a bug open at bugs.gentoo.org[1] and they are thinking > about getting rid of rplay. Someone else pointed out the fact that > fvwm needs it. So I looked at this bug, and looks simple enough to fix; and I would be willing to do so since: 1. Getting it fixed is worthwhile. 2. Someone should find out if it has an upstream maintainer. 3. If it turns out this ""bug"" is nothing more than a compiler-specific idiom for older code, let Gentoo and all the other distros do what they do best, unofficially patch the hell out of it. :P Another thing to bear in mind -- FVWM (unlike most other FOSS projects these days) *isn't* just running on Linux, it also runs on *BSD, Solaris, etc., where versions of librplay are still going to be maintained. So ripping it out isn't as straight-forward as you might think. > My question is this: as far as I know, rplay is used *only* to play > system sounds via FvwmEvent, is that correct? See above. > And second: that same functionality can be reproduced (also using > FvwmEvent) with virtually any command line player, like play, mpg123 > or whatever. Is there any loss of functionality that I am not aware > of if we do so instead of using rplay? The whole point of librplay was so that you didn't need to use external apps like this, where they were less common on older installations of UN*X. > If 1) rplay is only used for system sounds and 2) we can use virtually > any other player to do the same work in conjunction with FvwmEvent > then rplay is pointless and fixing it to work with newer gcc versions > is a waste of time. Please, kindly let me know what do you think about > this reasoning. See above. -- Thomas Adam
