cvs commit: ports/print/ghostscript9 Makefile
ports/print/ghostscript9/files patch-base-gdevl256.c patch-base-gdevvglb.c
Hiroki Sato
hrs at FreeBSD.org
Thu Jul 21 06:13:57 UTC 2011
Alexey Dokuchaev <danfe at FreeBSD.ORG> wrote
in <20110721054015.GA72702 at FreeBSD.org>:
da> On Sun, Jul 17, 2011 at 03:22:48PM +0200, Matthias Andree wrote:
da> > Am 17.07.2011 13:18, schrieb Hiroki Sato:
da> > > hrs 2011-07-17 11:18:47 UTC
da> > >
da> > > Modified files:
da> > > print/ghostscript9 Makefile
da> > > Added files:
da> > > print/ghostscript9/files patch-base-gdevl256.c
da> > > patch-base-gdevvglb.c
da> > > Log:
da> > > Fix a build failure when vaglib and/or lvga256 is specified.
da> >
da> > It is very unfortunate that the PORTREVISION was bumped, and wasteful.
da> > This was not necessary, let me explain:
da> >
da> > - if the build fails given specific (apparently non-default) options,
da> > there is no prior port/package where files change
da> >
da> > - we bump PORTREVISION only if installed files or dependencies change.
da>
da> Even for deps/plist changes, it is not always necessary to bump revision.
da> If, for example, port behavior is changed so it installs few extra files,
da> PORTREVISION can be left intact, as long as previous version of the port has
da> correct plist. On contrast, if pkg-plist was updated to include files that
da> were installed but didn't have plist entry before, PORTREVISION should be
da> bumped so users can get "fixed" package.
da>
da> In general, I totally agree with Matthias on this issue: please give enough
da> thinking before bumping PORTREVISION.
Sorry, I mistakenly thought one of the options was enabled by default
for some reason.
-- Hiroki
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/cvs-all/attachments/20110721/91885da2/attachment.pgp
More information about the cvs-all
mailing list