aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorGregory M. Turner <gmt@be-evil.net>2017-11-27 15:01:40 -0800
committerZac Medico <zmedico@gentoo.org>2017-11-28 14:25:18 -0800
commit77d8abd747a15f90d7a45c334efecaf47261501a (patch)
tree9db4ccf7210743e80632eea541fe6802702f4976 /COPYING
parentUpdates for portage-2.3.16 release (diff)
downloadportage-77d8abd747a15f90d7a45c334efecaf47261501a.tar.gz
portage-77d8abd747a15f90d7a45c334efecaf47261501a.tar.bz2
portage-77d8abd747a15f90d7a45c334efecaf47261501a.zip
versions: Drop non-PMS "cvs." prefix in ${PV}
This feature was introduced 12 years ago in (the cvs commit corresponding to the git commit) 9f3a46665c. There are a lot of reasons not to keep it around: o PMS permits no such prefix in ${PV} o Apparently nobody uses it (perhaps nobody /ever/ used it) o It special-cases cvs, which nobody uses, either, in 2017 o Almost* causes ambiguity between ${PN} and ${PV} in ${P} if some future EAPI tried to support both this and dots in package names simultaneously. Therefore, remove it from from the "_v" regular expression, renumber hard-coded group indexes and nuke corresponding tests. *PMS would technically avoid abiguity as ยง3.1.2 requires ${PV} to "win" any such conflict over contested bits in the middle of ${P}. However, clearly it's prefereable for this rule to be as redundant as possible. Note: this is my 3rd spin of this commitmsg; it's coming along nicely I think :P Signed-off-by: Gregory M. Turner <gmt@be-evil.net>
Diffstat (limited to 'COPYING')
0 files changed, 0 insertions, 0 deletions