Bug 744868

Summary:	[Future EAPI] usev with a second optional argument
Product:	Gentoo Hosted Projects	Reporter:	David Seifert <soap>
Component:	PMS/EAPI	Assignee:	Package Manager Specification <pms>
Status:	RESOLVED FIXED
Severity:	normal	CC:	esigra, mgorny, pacho
Priority:	Normal
Version:	unspecified
Hardware:	All
OS:	Linux
See Also:	https://bugs.gentoo.org/show_bug.cgi?id=199722
Whiteboard:	in-eapi-8
Package list:		Runtime testing required:	---
Bug Depends on:
Bug Blocks:	174380

Description David Seifert gentoo-dev

2020-09-26 20:02:09 UTC

Currently, there's no simple solution for the use case of "echo'ing X if USE=Y is set, but otherwise return nothing". The alternative $(usex Y X '') is terrible for readability in lists, and I have encountered multiple ebuilds where the empty part has been omitted (causing weird bogus warnings). $(usev Y X) is the natural extension of usev, and is consistent with how the second argument in use_with/use_enable changes the name of the option.

Reproducible: Always

Comment 1 Ulrich Müller gentoo-dev

2020-09-27 14:01:03 UTC

Presumably we should also add this to hasv, to keep behaviour of usev and hasv consistent?

Comment 2 David Seifert gentoo-dev

2020-09-27 14:03:37 UTC

(In reply to Ulrich Müller from comment #1)
> Presumably we should also add this to hasv, to keep behaviour of usev and
> hasv consistent?

Agreed

Comment 3 Michał Górny archtester

2020-09-27 14:22:08 UTC

How can you add it to hasv if hasv takes a variable number of arguments?

Comment 4 Ulrich Müller gentoo-dev

2020-09-27 15:20:56 UTC

(In reply to Michał Górny from comment #3)
> How can you add it to hasv if hasv takes a variable number of arguments?

Good point. I guess using an option to specify the output string would be awkward, and offer no advantage over "has && echo"?

So, ignore my comment #1 please.

Comment 5 Ulrich Müller gentoo-dev

2020-10-26 11:48:20 UTC

(In reply to Ulrich Müller from comment #4)
> Good point. I guess using an option to specify the output string would be
> awkward, and offer no advantage over "has && echo"?

Actually, hasv is not used in the Gentoo repository (and never was, historically). So another way to get rid of the inconsistency would be banning hasv altogether.

Comment 6 Larry the Git Cow gentoo-dev

2021-05-24 04:37:23 UTC

The bug has been referenced in the following commit(s):

https://gitweb.gentoo.org/proj/portage.git/commit/?id=96c2cf5ac78f0886e23186dd84f9080572d98adf

commit 96c2cf5ac78f0886e23186dd84f9080572d98adf
Author:     Michał Górny <mgorny@gentoo.org>
AuthorDate: 2021-05-12 12:36:53 +0000
Commit:     Zac Medico <zmedico@gentoo.org>
CommitDate: 2021-05-24 04:36:52 +0000

    Support second usev arg in EAPI 8
    
    Bug: https://bugs.gentoo.org/744868
    Signed-off-by: Michał Górny <mgorny@gentoo.org>
    Signed-off-by: Zac Medico <zmedico@gentoo.org>

 bin/eapi.sh          | 4 ++++
 bin/phase-helpers.sh | 6 +++++-
 2 files changed, 9 insertions(+), 1 deletion(-)

Comment 7 Larry the Git Cow gentoo-dev

2021-06-13 19:18:59 UTC

The bug has been referenced in the following commit(s):

https://gitweb.gentoo.org/proj/pms.git/commit/?id=c2847df81db51bc1402aa9a4339d53e5a0179208

commit c2847df81db51bc1402aa9a4339d53e5a0179208
Author:     Ulrich Müller <ulm@gentoo.org>
AuthorDate: 2021-05-16 18:59:37 +0000
Commit:     Ulrich Müller <ulm@gentoo.org>
CommitDate: 2021-06-05 08:55:41 +0000

    EAPI 8: usev has an optional second argument
    
    Bug: https://bugs.gentoo.org/744868
    Signed-off-by: Ulrich Müller <ulm@gentoo.org>

 eapi-differences.tex |  4 ++++
 pkg-mgr-commands.tex | 23 ++++++++++++++---------
 2 files changed, 18 insertions(+), 9 deletions(-)