http://sourceforge.net/project/showfiles.php?group_id=235&package_id=253 Gaim 2.0 Beta 3.1 is available for download now on Sourceforge.
The good news is a version bump ebuild works. The bad news is that I have no idea what to call it since gaim-2.0.0_beta3.1 is non-canonical. # ebuild gaim-2.0.0_beta3.1.ebuild digest !!! net-im/gaim-2.0.0_beta3.1 does not follow correct package syntax. Any ideas?
(In reply to comment #1) > The good news is a version bump ebuild works. The bad news is that I have no > idea what to call it since gaim-2.0.0_beta3.1 is non-canonical. > > # ebuild gaim-2.0.0_beta3.1.ebuild digest > !!! net-im/gaim-2.0.0_beta3.1 does not follow correct package syntax. > > Any ideas? > What about net-im/gaim-2.0.0_beta31 ? beta 4 would be net-im/gaim-2.0.0_beta40
(In reply to comment #2) > (In reply to comment #1) > > The good news is a version bump ebuild works. The bad news is that I have no > > idea what to call it since gaim-2.0.0_beta3.1 is non-canonical. > > > > # ebuild gaim-2.0.0_beta3.1.ebuild digest > > !!! net-im/gaim-2.0.0_beta3.1 does not follow correct package syntax. > > > > Any ideas? > > > > What about net-im/gaim-2.0.0_beta31 ? > beta 4 would be net-im/gaim-2.0.0_beta40 > i would prefere net-im/gaim-2.0.0_beta3a if a beta 4 will be released, than there would be no need to call the package beta40
(In reply to comment #3) > i would prefere net-im/gaim-2.0.0_beta3a As would I, but... # ebuild gaim-2.0.0_beta3a.ebuild digest !!! net-im/gaim-2.0.0_beta3a does not follow correct package syntax.
Why not just call the build gaim-2.0.0_beta3-r2? You could add a note with the changelog saying that it is beta3.1. Anyone clever enough to unmask the betas should know enough to read the changelogs too.
We could also ask ourself if we should change the package syntax or not ? Will we have other packages called *-beat#.# coming ? want we go this way ? If not, I think -r2 would goes well. Even if we find that changing the syntax is the way, it would take time to be accepted. I suggest going the -r2 way. What's your opinion guys ?
Created attachment 95279 [details] version bump ebuild The Gaim devs have always taken a bit of a renegade approach to naming; I don't see why this should be any different. Perhaps a little more freedom in naming is necessary. Here's the -r2 ebuild.
> is necessary. Here's the -r2 ebuild. Thank you for that ebuild. Unfortunately emerging this results in following error: "!!! ERROR: net-im/gaim-2.0.0_beta3-r2 failed. Call stack: ebuild.sh, line 1555: Called dyn_setup ebuild.sh, line 668: Called pkg_setup gaim-2.0.0_beta3-r2.ebuild, line 130: Called die !!! Configure failed" Regards, Lupo
Could we see more info than just that? Also, what version of Portage are you using?
(In reply to comment #8) > > is necessary. Here's the -r2 ebuild. > > Thank you for that ebuild. Unfortunately emerging this results in following > error: > "!!! ERROR: net-im/gaim-2.0.0_beta3-r2 failed. > Call stack: > ebuild.sh, line 1555: Called dyn_setup > ebuild.sh, line 668: Called pkg_setup > gaim-2.0.0_beta3-r2.ebuild, line 130: Called die > > !!! Configure failed" > > Regards, > Lupo > I compared this ebuild with my update of the gaim beta 3 ebuild (which is working for me, btw.), and these are the differences: the -r2 Version in the bugzilla: src_unpack() { unpack ${A} cd ${S} epatch ${FILESDIR}/${P}-as-needed.patch epatch ${FILESDIR}/${P}-avahi-compat.patch eautomake || die "Failed running eautomake" } my updated version: src_unpack() { unpack ${A} cd ${S} epatch ${FILESDIR}/${PN}-2.0.0_beta3-as-needed.patch epatch ${FILESDIR}/${PN}-2.0.0_beta3-avahi-compat.patch eautomake || die "Failed running eautomake" } i don't know if this is good or bad what i've done, because my knowledge about ebuilds isn't very great...
The way we both have it defined works, because the variable expansions are identical. That is, "${PN}-2.0.0_beta3" == "${P}" as long as the ebuild is of the form gaim-2.0.0_beta3-r*.ebuild. Which one is better is beyond me, though.
(In reply to comment #11) > The way we both have it defined works, because the variable expansions are > identical. That is, "${PN}-2.0.0_beta3" == "${P}" as long as the ebuild is of > the form gaim-2.0.0_beta3-r*.ebuild. > > Which one is better is beyond me, though. > Scratch that, my way (${P}) is more future-proof, because your way will require a rewrite if and when beta4 comes out, or so I think... perhaps an expert can confirm that.
(In reply to comment #12) > (In reply to comment #11) > > The way we both have it defined works, because the variable expansions are > > identical. That is, "${PN}-2.0.0_beta3" == "${P}" as long as the ebuild is of > > the form gaim-2.0.0_beta3-r*.ebuild. > > > > Which one is better is beyond me, though. > > > > Scratch that, my way (${P}) is more future-proof, because your way will require > a rewrite if and when beta4 comes out, or so I think... perhaps an expert can > confirm that. > my ebuild was has a _beta31 ending, so it wouldn't work with ${P} - so i used ${P}, i thought this was the fault when he emerged is, because the ebuilds are similar except this part and my ebuild works on my machine. my fault ;)
(In reply to comment #7) > Created an attachment (id=95279) [edit] > version bump ebuild > > The Gaim devs have always taken a bit of a renegade approach to naming; I don't > see why this should be any different. Perhaps a little more freedom in naming > is necessary. Here's the -r2 ebuild. > I tried this ebuild and I only got an armless warning during compilation : configure: WARNING: *** doxygen not found, docs will not be available All fine for me
Created attachment 95365 [details] patch to include IUSE=doc Here's a simple ebuild patch I threw together to enable support for the "doc" USE flag and squelch the warning if doc is not enabled. Try it and let me know (I've only done preliminary testing).
(In reply to comment #15) > Created an attachment (id=95365) [edit] > patch to include IUSE=doc > > Here's a simple ebuild patch I threw together to enable support for the "doc" > USE flag and squelch the warning if doc is not enabled. Try it and let me know > (I've only done preliminary testing). > No more warnings about the doc. Everything seems fine.
Sorry for the delay, I've added the doc use flag, and added beta 3.1 as -r2. And thanks for doing all the work for me!
(In reply to comment #17) > Sorry for the delay, I've added the doc use flag, and added beta 3.1 as -r2. > And thanks for doing all the work for me! > Thanks! Should -r1 be removed from Portage?
