281000 – sci-libs/gdal-1.6.1 always wants ruby, why?

Bug 281000 - sci-libs/gdal-1.6.1 always wants ruby, why?

Summary: sci-libs/gdal-1.6.1 always wants ruby, why?

Status:	RESOLVED FIXED

Alias:	None

Product:	Gentoo Linux
Classification:	Unclassified
Component:	New packages (show other bugs)
Hardware:	All Linux

Importance:	High normal (vote)
Assignee:	Steve Arnold

URL:
Whiteboard:
Keywords:

Depends on:
Blocks:

Reported:	2009-08-10 13:52 UTC by Stefan de Konink
Modified:	2009-10-18 23:12 UTC (History)
CC List:	1 user (show)

See Also:
Package list:
Runtime testing required:	---

Attachments
makes USE=-ruby gdal not depending on ruby (gdal-1.6.2-optional_ruby.diff,566 bytes, patch) 2009-09-21 13:33 UTC, Petr Pisar	Details \| Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description Stefan de Konink 2009-08-10 13:52:52 UTC

as $subj

Reproducible: Always

Comment 1 James Earl Spahlinger 2009-08-10 21:37:29 UTC

package has ruby? ( >=dev-lang/ruby-1.8.4.20060226 ) in DEPENDS but also has virtual/ruby in DEPENDS as well. Question appears to be: Is virtual/ruby required for the package to build when ruby useflag is *disabled*

Comment 2 Steve Arnold archtester

2009-08-14 05:49:52 UTC

I'm assuming the eclass RUBY_OPTIONAL="yes" means what it says, and I don't see a virtual/ruby in that version or anywhere else, so I'm confused.  Care to elaborate?

Comment 3 Petr Pisar 2009-09-21 13:33:31 UTC

Created attachment 204816 [details, diff]
makes USE=-ruby gdal not depending on ruby

ruby eclass is inherited before RUBY_OPTIONAL is set to "yes", thus ruby eclass adds virtual/ruby into {R,}DEPEND as stated inside eclass code always.

This patch moves RUBY_OPTINAL definition before inherit clause.

Comment 4 Ryan May 2009-10-12 18:28:54 UTC

This patch fixes the problem for me.

Comment 5 Steve Arnold archtester

2009-10-18 23:12:47 UTC

The ruby bindings aren't that current, at least as far as upstream maintenance, but they appear to build okay.  Thanks for pointing out the RUBY_OPTIONAL thing.