Re: [PATCH 1/3] setup: Provide GIT_PREFIX to built-ins

2 messages, 2 authors, 2016-06-15 · open the first message on its own page

Re: [PATCH 1/3] setup: Provide GIT_PREFIX to built-ins

From: Junio C Hamano <hidden>
Date: 2016-06-15 22:51:18

David Aguilar [off-list ref] writes:
I guess external scripts can call rev-parse --show-prefix themselves?
That has always been the case, I think, and it shouldn't be a problem.

The real reason you want the new GIT_PREFIX for alias/hooks is otherwise
they would not have a way to even say --show-prefix to figure it out
themselves.
quoted
Overall I think it's a good change, btw. But it leaves it up to the
(script) user to know whether git has actually changed the cwd or not,
i.e.: Is $(pwd) where the user called us from or $(pwd)/$GIT_PREFIX?
As long as there is a way for a script to figure it out when it wants to
know, I think it should be Ok.

Isn't it just the matter of reading --show-prefix and comparing it with
what came in $GIT_PREFIX?

Re: [PATCH 1/3] setup: Provide GIT_PREFIX to built-ins

From: Michael J Gruber <hidden>
Date: 2016-06-15 22:51:18

Junio C Hamano venit, vidit, dixit 23.05.2011 18:43:
David Aguilar [off-list ref] writes:
quoted
I guess external scripts can call rev-parse --show-prefix themselves?
That has always been the case, I think, and it shouldn't be a problem.

The real reason you want the new GIT_PREFIX for alias/hooks is otherwise
they would not have a way to even say --show-prefix to figure it out
themselves.
quoted
quoted
Overall I think it's a good change, btw. But it leaves it up to the
(script) user to know whether git has actually changed the cwd or not,
i.e.: Is $(pwd) where the user called us from or $(pwd)/$GIT_PREFIX?
As long as there is a way for a script to figure it out when it wants to
know, I think it should be Ok.

Isn't it just the matter of reading --show-prefix and comparing it with
what came in $GIT_PREFIX?
Yep, one is before and one is after any eventual cd'ing which git may
do. I just wanted to point out the difference. And the technical
difference (env var. vs. rev-parse option) is due to that difference
(and thus natural).

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