svn commit: r254224 - head/share/man/man7

[Hiroki Sato]

Glen Barber <gjb at FreeBSD.org> wrote
  in <20130813195257.GA2241 at glenbarber.us>:

gj> On Wed, Aug 14, 2013 at 04:25:57AM +0900, Hiroki Sato wrote:
gj> > gj> +.It Va SRCREVISION
gj> > gj> +The revision of the
gj> > gj> +.Li src/
gj> > gj> +tree to use.
gj> > gj> +Defaults to the current top of tree revision.
gj> > gj> +.It Va DOCREVISION
gj> > gj> +The revision of the
gj> > gj> +.Li doc/
gj> > gj> +tree to use.
gj> > gj> +Defaults to the current top of tree revision.
gj> > gj> +.It Va PORTREVISION
gj> >
gj> >  Why separation between revision and branch is needed?  I
gj> >  intentionally dropped this part from your old patch to
gj> >  generate-release.sh because branch and revision number can be
gj> >  specified like releng/9.2 at NNN in a single variable.  An incorrect
gj> >  configuration of the two variables do not always cause a fatal error,
gj> >  so specification in fewer number of variables is more foolproof.
gj> >
gj>
gj> I personally do not like using branch at rNNNNNN, but will update to remove
gj> the {SRC,DOC,PORT}REVISION variables.

 Thank you.  I don't like it either, honestly...

gj> > gj> +.It Va TARGET
gj> > gj> +The target machine type for the release.
gj> > gj> +Defaults to the current machine type.
gj> > gj> +.It Va TARGET_ARCH
gj> > gj> +The target machine architecture for the release.
gj> > gj> +Defaults to the value of
gj> > gj> +.Va TARGET .
gj> > gj> +.Pp
gj> >
gj> >  Please remove default configuration of these variables from
gj> >  release.sh.  The reasonable defaults are already set in src/Makefile
gj> >  and setting TARGET_ARCH=$TARGET by default is simply wrong.  Also,
gj> >  get_rev_branch() is redundant.
gj> >
gj>
gj> Will be changed shortly.
gj>
gj> Why is get_rev_branch() redundant?

 Because variables in get_rev_branch() are already defined in
 src/release/Makefile, too.  If one wants to use them in release.sh
 for some reason, "make -VBRANCH" is preferable instead of duplicating
 the same logic in Makefile to extract them from newvers.sh.  This
 script should not do the same job which Makefile does.

 IMO, the default value of OSRELEASE is not needed here.  It seems
 used in RELSTRING, but it does not work (and even if it works, it
 just generates the same default value).  If we always need to
 override the default value, Makefile should be fixed instead.

-- 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/svn-src-head/attachments/20130814/7bd1372f/attachment.sig>