On Thu, 2011-12-22 at 13:35 +0000, Colin Guthrie wrote: > 'Twas brillig, and Maarten Bosmans at 21/12/11 21:50 did gyre and gimble: > > 2011/12/21 Arun Raghavan <arun.raghavan at collabora.co.uk>: > >> IMO EsounD is really quite irrelevant on most modern systems today, and more > >> so for embedded systems. > >> > >> Any objections to making it optional? > > > > No, seems fine. > > > > I'd go for HAVE_ESOUND though just to keep it consistent with the rest. > > > > As seen in Makefile.am as a conditional or in source files as a macro, > > the semantics of the symbol for esound are not different from those > > for e.g. solaris. It just determines whether a certain piece of code > > gets compiled or not. > > > > The only place where esound is different from the others is in > > configure.ac. But even here we have already lost a precise link > > between HAVE_SOLARIS and the availability of the solaris headers. As > > when --disable-solaris is passed to configure, HAVE_SOLARIS=0 even > > when the headers are available. > > > > So it's probably better to be consistently imprecise with our variable > > naming than to introduce a new name prefix. > > Yeah what Maarten said :D > > Looks good. Thanks for the review, guys -- pushed with an added post-configure message. Cheers, Arun
