On Sun, Nov 12, 2006 at 12:13:02AM -0800, Steve Langasek wrote: > The corresponding bug in gcc-4.1 was #384036, which lists gcc/doc/gcc.texi > "and probably others". Matthias, could you please comment on whether there > are any "others" known that need to be removed in order to resolve this bug? > > Ron, is there any chance that mingw32 could be changed to build against the > current gcc-4.1-source package, eliminating the need to prune these sources > separately? Hmm, probably not a change to be making at this point in the > release cycle... :/
I talked to Danny early in the year about the prospects of a 4.x branch release (while looking at adding portable xlocale support) and he indicated there were a few issues that made that non trivial. Looks like the source we have now is still the current upstream release candidate, so I guess they haven't resolved them all yet. I'd love to have this merged with the mainline source, or easy to apply as a patch to it, but I fear we are still some way from that in practice, and I agree its definitely not something to be aiming directly for with a freeze this close. That said though, this package does rely upon the native gcc packages to provide the docs (they are identical and include the billware specific options anyway), so they are only in the source, and only because its tarball is pristine. I have plenty of less menial work than repacking that for each release to get done, but if we know precisely what files and/or dirs in the tree need to be purged to remove the offending material then it shouldn't be too hard to script something to preprocess the upstream source before it goes in the .deb. To do that though I mostly need an authoritative list of what we consider unacceptable source that would block the release. When we have an expert opinion on that I can swing the hatchet as required. Thanks! Ron -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
