> I'll chime in with support for Scott's view here. (And yes, I > apologize that this is off-topic to the issue of build documentation).
Sorry, topic renamed. > As one of those responsible for getting Xerces 3.0 to the point it is > now, (and it's been at that point for several years now), I can say > that it works well for me. But we need community leadership to get it > out the door, and I don't have the time for that. There seems to be > auto* issues on a few platforms for 3.0, but those aren't platforms I > have access to nor know anything about: it really is up to committers > who can support those platforms to get things working there, or to > suggest changes that will allow them to work, or (if needed) to come > up with an alternate non auto* build system for those systems. Anything that's not Windows should stay with autoconf, anything else is nuts. As far as fixing the scripts is concerned, the thing to keep in mind is that it's not a stable target. Supporting a platform usually means keeping up with it and patching the scripts as time goes on, so that each roll up release eventually pulls in all the known patching that has gone on to keep the scripts working. I can tell you that for Fedora, for example, gcc is churning at an astounding rate. It *will* break a lot, there just isn't a lot you can do about it unless you drop support for Fedora (that's just an example, I don't know if Fedora is a supported OS here). So I would suggest that specific people are tasked (on the basis of them volunteering) to deal with the ongoing maintenance of specific platform/compiler combos, and if nobody steps up, you cut it. For those dropped platforms, packagers will likely step up outside the mainline project tree and patch things, and if they care to donate back, that's fine. Otherwise it just doesn't happen. > Boris, you offered earlier to take the lead in pushing 3.0 out the > door. Are you still able to do that? (perhaps on a new thread...) I'm still willing to help on platforms I have access to (and need), but the last time a set of the open issues was posted, somebody indicated that it was out of date, so I didn't take it very seriously as an accurate snapshot. -- Scott --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
