Re: [RFC PATCH] GIT-VERSION-GEN: set --abbrev=9 to match auto-scaling

2 messages, 2 authors, 2016-12-05 · open the first message on its own page

Re: [RFC PATCH] GIT-VERSION-GEN: set --abbrev=9 to match auto-scaling

From: Junio C Hamano <hidden>
Date: 2016-12-05 22:25:13

Ramsay Jones [off-list ref] writes:
As I said, the original version of the patch just removed the
--abbrev=7, but then I started to think about why you might have
used --abbrev in the first place (first in commit 9b88fcef7 and
again in commit bf505158d). Making sure to override the configuration
was the only thing I could come up with. So, I was hoping you could
remember why! :-P
Nope.  As a maintainer support script, the only thing I cared about
it is that there is no -gXXXX at the end for anything I release ;-)
(I assumed it was to force a measure of uniformity/reproducibility).
You cannot force uniformity/reproducibility with fixed abbrev,
unless you set abbreviation length to 40, so you are correct to add
"a measure of" there ;-) The first choice (i.e. 4) may have had a
justification to force absolute minimum, and the second one (i.e. 7)
may have had a justifiation to make it clear that we are using the
same setting as the default, so in post-1.7.10 era, I think it is
fine for us to just say "we have been using the same as default, so
let's not specify anything explicitly".

Re: [RFC PATCH] GIT-VERSION-GEN: set --abbrev=9 to match auto-scaling

From: Ramsay Jones <hidden>
Date: 2016-12-05 22:43:21


On 05/12/16 22:24, Junio C Hamano wrote:
Ramsay Jones [off-list ref] writes:
quoted
As I said, the original version of the patch just removed the
--abbrev=7, but then I started to think about why you might have
used --abbrev in the first place (first in commit 9b88fcef7 and
again in commit bf505158d). Making sure to override the configuration
was the only thing I could come up with. So, I was hoping you could
remember why! :-P
Nope.  As a maintainer support script, the only thing I cared about
it is that there is no -gXXXX at the end for anything I release ;-)
quoted
(I assumed it was to force a measure of uniformity/reproducibility).
You cannot force uniformity/reproducibility with fixed abbrev,
unless you set abbreviation length to 40, so you are correct to add
"a measure of" there ;-)
Indeed. ;-)
                           The first choice (i.e. 4) may have had a
justification to force absolute minimum, and the second one (i.e. 7)
may have had a justifiation to make it clear that we are using the
same setting as the default, so in post-1.7.10 era, I think it is
fine for us to just say "we have been using the same as default, so
let's not specify anything explicitly".
So, you would be happy with just removing the '--abbrev' argument?

(That's fine by me; I don't set core.abbrev!)

ATB,
Ramsay Jones
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help