Created attachment 443506 [details, diff] media-sound/pms-0.42-r2 The submitted ebuild bumps EAPI to version 6 and fixes bugs #526530 and #351996. https://bugs.gentoo.org/show_bug.cgi?id=351996 https://bugs.gentoo.org/show_bug.cgi?id=351996 Fixing bug #351996 requires patching the source code, so a revbump is required.
Created attachment 443508 [details, diff] Remove old ebuilds from the tree Additionally, older revisions of media-sound/pms shall be dropped from the portage tree.
Excellent, thanks! The 0.42-r2 ebuild still points to SourceForge in both HOMEPAGE and SRC_URI. Would you like to change these to GitHub before your patches get committed? I feel it would make sense, given what the metadata file already says. By the way, this isn't directly related to the version bump because this issue was already there before you took over but it would be great if you set the correct sys-libs/ncurses slot in DEPEND, for both the release and the live ebuild.
Created attachment 444062 [details, diff] media-sound/pms-0.42-r2, drop old revisions (In reply to Marek Szuba from comment #2) > Excellent, thanks! > > The 0.42-r2 ebuild still points to SourceForge in both HOMEPAGE and SRC_URI. > Would you like to change these to GitHub before your patches get committed? > I feel it would make sense, given what the metadata file already says. > > By the way, this isn't directly related to the version bump because this > issue was already there before you took over but it would be great if you > set the correct sys-libs/ncurses slot in DEPEND, for both the release and > the live ebuild. Thanks a lot for noticing the old URLs, I've probably overlooked it. irt ncurses slots, I've spoken with the upstream some time ago: they said that it should work on either version (5 or 6). I've attached a new series of patches which update the URLs as well as properly deal with the manifest (and fix a typo).
(In reply to Marek Szuba from comment #2) > By the way, this isn't directly related to the version bump because this > issue was already there before you took over but it would be great if you > set the correct sys-libs/ncurses slot in DEPEND, for both the release and > the live ebuild. Or you mean that I shall specify the := slot, if it works with any version instead?
Yes, if any slot indeed does the trick := is perfectly acceptable.
Created attachment 444100 [details, diff] Match ncurses slot And here is the final commit, which clarifies the slot dependency in both ebuilds (live and versioned).
(In reply to Mykyta Holubakha from comment #6) > Created attachment 444100 [details, diff] [details, diff] > Match ncurses slot > > And here is the final commit, which clarifies the slot dependency in both > ebuilds (live and versioned). you should obsolete the current versions and provide a single patch diff
(In reply to Kristian Fiskerstrand from comment #7) > (In reply to Mykyta Holubakha from comment #6) > > Created attachment 444100 [details, diff] [details, diff] [details, diff] > > Match ncurses slot > > > > And here is the final commit, which clarifies the slot dependency in both > > ebuilds (live and versioned). > > you should obsolete the current versions and provide a single patch diff to be even a bit more precise, you should provide one patch for each atomic change, so addition of new revision would be separate from removal of old.
(In reply to Kristian Fiskerstrand from comment #8) > (In reply to Kristian Fiskerstrand from comment #7) > > (In reply to Mykyta Holubakha from comment #6) > > > Created attachment 444100 [details, diff] [details, diff] [details, diff] [details, diff] > > > Match ncurses slot > > > > > > And here is the final commit, which clarifies the slot dependency in both > > > ebuilds (live and versioned). > > > > you should obsolete the current versions and provide a single patch diff > > to be even a bit more precise, you should provide one patch for each atomic > change, so addition of new revision would be separate from removal of old. In fact, git am will apply the provided patch as three distinct atomic commits. Take a closer look.
(In reply to Mykyta Holubakha from comment #9) > (In reply to Kristian Fiskerstrand from comment #8) > > (In reply to Kristian Fiskerstrand from comment #7) > > > (In reply to Mykyta Holubakha from comment #6) > > > > Created attachment 444100 [details, diff] [details, diff] [details, diff] [details, diff] [details, diff] > > > > Match ncurses slot > > > > > > > > And here is the final commit, which clarifies the slot dependency in both > > > > ebuilds (live and versioned). > > > > > > you should obsolete the current versions and provide a single patch diff > > > > to be even a bit more precise, you should provide one patch for each atomic > > change, so addition of new revision would be separate from removal of old. > > In fact, git am will apply the provided patch as three distinct atomic > commits. > Take a closer look. The patch in comment #6 should be squashed into [PATCH 1/3] media-sound/pms: revbump to 0.42-r2
Created attachment 444120 [details, diff] media-sound/pms: revbump to 0.42-r2, drop old versions, match ncurses slots Here is another patch. It has 4 commits (in the following order): 1) revbump to 0.42-r2 (already with ncurses matching slot dep) 2) set ncurses dep slot to matching in the live ebuild 3) drop 0.42 4) drop 0.42-r1.
(In reply to Mykyta Holubakha from comment #11) > Created attachment 444120 [details, diff] [details, diff] > media-sound/pms: revbump to 0.42-r2, drop old versions, match ncurses slots > > Here is another patch. It has 4 commits (in the following order): > 1) revbump to 0.42-r2 (already with ncurses matching slot dep) > 2) set ncurses dep slot to matching in the live ebuild > 3) drop 0.42 > 4) drop 0.42-r1. Thanks, pushed https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=7bfbcb11dd61ac980b5380b29599d73071af9286 https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=f8c37ed6cb707f44a21df0c1561dba6b46dad064 https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=a82547f902dc0328316ef2617945d409058e5654 https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=1a0a1691f6f70cad55b538de0a7f947da297f822 Package not stable, so closing