svn commit: r476592 - in head/lang: . io io-devel

[Mathieu Arnold]

On Tue, Aug 07, 2018 at 07:42:48PM +0200, Pietro Cerutti wrote:
> 
> > On 7 Aug 2018, at 18:44, Mathieu Arnold <mat at FreeBSD.org> wrote:
> > 
> >> On Tue, Aug 07, 2018 at 02:29:06PM +0000, Pietro Cerutti wrote:
> >>> On Aug 07 2018, 14:24 UTC, Jochen Neumeister <joneum at FreeBSD.org> wrote:
> >>> [-- Type: text/plain; charset=utf-8, Encoding: 7bit, Size: 1.2K --]
> >>> 
> >>>> On 8/7/18 2:41 PM, Pietro Cerutti wrote:
> >>>> Author: gahr
> >>>> Date: Tue Aug  7 12:41:26 2018
> >>>> New Revision: 476592
> >>>> URL: https://svnweb.freebsd.org/changeset/ports/476592
> >>>> 
> >>>> Log:
> >>>>  lang/io-devel: new port, development snapshot of lang/io
> >>>> 
> >>>> Added:
> >>>>  head/lang/io-devel/
> >>>>  head/lang/io-devel/Makefile   (contents, props changed)
> >>>>  head/lang/io-devel/distinfo   (contents, props changed)
> >>>>  head/lang/io-devel/pkg-descr   (contents, props changed)
> >>>>  head/lang/io-devel/pkg-message   (contents, props changed)
> >>>>  head/lang/io-devel/pkg-plist   (contents, props changed)
> >>>> Modified:
> >>>>  head/lang/Makefile
> >>>>  head/lang/io/Makefile
> >>> 
> >>> 
> >>> Heya Pietro,
> >>> 
> >>> would not it be better to use a copy of "io" here?
> >>> So "svn cp lang/io lang/io-devel - and then continue editing?
> >> 
> >> Hi - yeah in principle why not, but the two ports are different enough that
> >> I chose to go the clean way and create this new one from scratch.
> >> 
> >> The principal reason for the big difference is that AddOns are handled as
> >> OPTIONS in the old port, and not handled at all in the new one.
> > 
> > The only reason to not use svn cp is if the software is not the same.
> > Here, it is clearly the same with a different version.  Please re-add
> > the port correctly.
> 
> svn cp + edit would lead to a 100% diff to anything semantically relevant to the software version (i.e., apart maintainer and such).
> 
> The -devel version is as clean as can be, quite different in term of generated package and Make machinery, and I really see no value in keeping the tortuous history of the original port.
> 
> What gain do you see?

The reason we bother with doing svn cp in the first place is to denote
ancestry, to say "this is related".  The amount of diff this would
generate is really irrelevant.

-- 
Mathieu Arnold
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 963 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/svn-ports-head/attachments/20180807/c81137cd/attachment.sig>