[gentoo-portage-dev] [PATCH] Allow virtuals in package.provided

[gentoo-portage-dev] [PATCH] Allow virtuals in package.provided

[Michał Górny]

With new-style virtuals, there is no reason to enforce special rules to
virtuals in package.provided. If user wishes to implicitly provide
the virual package, we should not forbid him. Of course, he knows
the implications.
---
 man/portage.5                        | 7 -------
 pym/portage/package/ebuild/config.py | 6 ------
 2 files changed, 13 deletions(-)

diff --git a/man/portage.5 b/man/portage.5
index 150294b..46835b5 100644
--- a/man/portage.5
+++ b/man/portage.5
@@ -400,13 +400,6 @@ entries may cause installed packages satisfying equivalent dependencies
 to be removed by \fBemerge\fR(1) \fB\-\-depclean\fR actions (see the
 \fBACTIONS\fR section of the \fBemerge\fR(1) man page for more information).
 
-Virtual packages (virtual/*) should not be specified in package.provided,
-since virtual packages themselves do not provide any files, and
-package.provided is intended to represent packages that do provide files.
-Depending on the type of virtual, it may be necessary to add an entry to the
-virtuals file and/or add a package that satisfies a virtual to
-package.provided.
-
 .I Format:
 .nf
 \- comments begin with # (no inline comments)
diff --git a/pym/portage/package/ebuild/config.py b/pym/portage/package/ebuild/config.py
index 59e239b..4d1f1d5 100644
--- a/pym/portage/package/ebuild/config.py
+++ b/pym/portage/package/ebuild/config.py
@@ -803,12 +803,6 @@ class config(object):
 					has_invalid_data = True
 					del pkgprovidedlines[x]
 					continue
-				if cpvr[0] == "virtual":
-					writemsg(_("Virtual package in package.provided: %s\n") % \
-						myline, noiselevel=-1)
-					has_invalid_data = True
-					del pkgprovidedlines[x]
-					continue
 			if has_invalid_data:
 				writemsg(_("See portage(5) for correct package.provided usage.\n"),
 					noiselevel=-1)
-- 
2.2.0

Re: [gentoo-portage-dev] [PATCH] Allow virtuals in package.provided

[Zac Medico]

@Robin: This patch reverts the changes from bug 161003. Maybe this
constraint is not needed anymore?

On 12/05/2014 08:43 AM, Michał Górny wrote:
> With new-style virtuals, there is no reason to enforce special rules to
> virtuals in package.provided. If user wishes to implicitly provide
> the virual package, we should not forbid him. Of course, he knows
> the implications.
> ---
>  man/portage.5                        | 7 -------
>  pym/portage/package/ebuild/config.py | 6 ------
>  2 files changed, 13 deletions(-)
> 
> diff --git a/man/portage.5 b/man/portage.5
> index 150294b..46835b5 100644
> --- a/man/portage.5
> +++ b/man/portage.5
> @@ -400,13 +400,6 @@ entries may cause installed packages satisfying equivalent dependencies
>  to be removed by \fBemerge\fR(1) \fB\-\-depclean\fR actions (see the
>  \fBACTIONS\fR section of the \fBemerge\fR(1) man page for more information).
>  
> -Virtual packages (virtual/*) should not be specified in package.provided,
> -since virtual packages themselves do not provide any files, and
> -package.provided is intended to represent packages that do provide files.
> -Depending on the type of virtual, it may be necessary to add an entry to the
> -virtuals file and/or add a package that satisfies a virtual to
> -package.provided.

LGTM. I guess we can mark bug 161003 resolved as "OBSOLETE".

[1] https://bugs.gentoo.org/show_bug.cgi?id=161003
-- 
Thanks,
Zac

Re: [gentoo-portage-dev] [PATCH] Allow virtuals in package.provided

[Michał Górny]

[-- Attachment #1: Type: text/plain, Size: 297 bytes --]

Dnia 2014-12-06, o godz. 22:31:42
Zac Medico <zmedico@gentoo.org> napisał(a):

> @Robin: This patch reverts the changes from bug 161003. Maybe this
> constraint is not needed anymore?

Considering the feedback on the bug, ok to push this one finally?

-- 
Best regards,
Michał Górny

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 949 bytes --]

Re: [gentoo-portage-dev] [PATCH] Allow virtuals in package.provided

[Zac Medico]

On 01/12/2015 01:34 AM, Michał Górny wrote:
> Dnia 2014-12-06, o godz. 22:31:42
> Zac Medico <zmedico@gentoo.org> napisał(a):
> 
>> @Robin: This patch reverts the changes from bug 161003. Maybe this
>> constraint is not needed anymore?
> 
> Considering the feedback on the bug, ok to push this one finally?
> 

Yes, please do.
-- 
Thanks,
Zac