diff mbox series

[bug#60564] scripts: refresh: Add -T option.

Message ID 20230303162135.2864-1-jgart@dismail.de
State New
Headers show
Series [bug#60564] scripts: refresh: Add -T option. | expand

Commit Message

jgart March 3, 2023, 4:21 p.m. UTC 
* doc/guix.texi (Invoking guix refresh): Document the -T option.
* guix/scripts/refresh.scm (%options): Add the -T flag.
---

Hi Josselin and Maxim

> I wonder if this option is common enough to need a dedicated short
> switch?

I would probably use the option more if it had a short option ;()

I think that it is useful. What would be the negative of adding the short option for it?

all best,

jgart

 doc/guix.texi            | 1 +
 guix/scripts/refresh.scm | 4 ++--
 2 files changed, 3 insertions(+), 2 deletions(-)

Comments

Maxim Cournoyer March 4, 2023, 3:30 a.m. UTC | #1 
Hi jgart,

jgart <jgart@dismail.de> writes:

> * doc/guix.texi (Invoking guix refresh): Document the -T option.
> * guix/scripts/refresh.scm (%options): Add the -T flag.
> ---
>
> Hi Josselin and Maxim
>
>> I wonder if this option is common enough to need a dedicated short
>> switch?
>
> I would probably use the option more if it had a short option ;()
>
> I think that it is useful. What would be the negative of adding the short option for it?

Nothing immediately, but we'd have to honor T forever as it'd be part of
the public CLI, loosing that letter for something more useful that may
yet to appear.

Probably an overblown concern, to be honest, eh.

I'll see what Josselin says about it, otherwise I'll push it soon.
Josselin Poiret March 4, 2023, 10:51 a.m. UTC | #2 
Hi jgart and Maxim,

Maxim Cournoyer <maxim.cournoyer@gmail.com> writes:

> Nothing immediately, but we'd have to honor T forever as it'd be part of
> the public CLI, loosing that letter for something more useful that may
> yet to appear.
>
> Probably an overblown concern, to be honest, eh.
>
> I'll see what Josselin says about it, otherwise I'll push it soon.

I'm on the fence as well, it doesn't seem like something you would use
very often, compared to eg. `-l` or `-u`, especially if you mention that
you would use the option more often if this was implemented (I'd expect
more of a "I use it all the time but typing it is annoying"). `guix
size` and `guix graph` are the more useful commands imho. What do you
use it for specifically (maybe that would help us understand the
use-case a bit more)?

At the same time, it's probably fine, we don't have many options for
`guix refresh` and it's not like dozens of them will get added overnight
(and they maybe don't deserve a short argument either).  I'll leave it
in your hands 🤠.

Best,
diff mbox series

Patch

diff --git a/doc/guix.texi b/doc/guix.texi
index 6ff525d0ff..25bbd75f04 100644
--- a/doc/guix.texi
+++ b/doc/guix.texi
@@ -14327,6 +14327,7 @@  for compatibility with an upgraded @code{flex} package.
 @table @code
 
 @item --list-transitive
+@item -T
 List all the packages which one or more packages depend upon.
 
 @example
diff --git a/guix/scripts/refresh.scm b/guix/scripts/refresh.scm
index 6498d73c2b..6b633840e0 100644
--- a/guix/scripts/refresh.scm
+++ b/guix/scripts/refresh.scm
@@ -101,7 +101,7 @@  (define %options
         (option '(#\r "recursive") #f #f
                 (lambda (opt name arg result)
                   (alist-cons 'recursive? #t result)))
-        (option '("list-transitive") #f #f
+        (option '(#\T "list-transitive") #f #f
                 (lambda (opt name arg result)
                   (alist-cons 'list-transitive? #t result)))
 
@@ -159,7 +159,7 @@  (define (show-help)
   (display (G_ "
   -r, --recursive        check the PACKAGE and its inputs for upgrades"))
   (display (G_ "
-      --list-transitive  list all the packages that PACKAGE depends on"))
+  -T, --list-transitive  list all the packages that PACKAGE depends on"))
   (newline)
   (display (G_ "
       --keyring=FILE     use FILE as the keyring of upstream OpenPGP keys"))