335506 – Enable dev-db/sqlite[extensions] by default

Bug 335506 - Enable dev-db/sqlite[extensions] by default

Summary: Enable dev-db/sqlite[extensions] by default

Status:	RESOLVED FIXED

Alias:	None

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

Importance:	High normal (vote)
Assignee:	Petteri Räty (RETIRED)

URL:
Whiteboard:
Keywords:

Depends on:
Blocks:

Reported:	2010-09-01 12:50 UTC by Johan Bergström
Modified:	2010-10-31 09:27 UTC (History)
CC List:	2 users (show)

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

Attachments
sqlite-3.7.2-r1.ebuild (sqlite-3.7.2-r1.ebuild,4.32 KB, text/plain) 2010-09-01 13:18 UTC, Johan Bergström	Details
sqlite-3.7.2-r1.ebuild (sqlite-3.7.2-r1.ebuild,4.34 KB, text/plain) 2010-09-10 08:04 UTC, Johan Bergström	Details
Show Obsolete (1) 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 Johan Bergström 2010-09-01 12:50:13 UTC

Upstream (at least the amalgation tarball) nowadays defaults to building dynamic extensions. We should do this too!

The only dep i could find in tree that needs to be changed is dev-perl/DBD-SQLite/DBD-SQLite-1.29-r2.ebuild

Comment 1 Johan Bergström 2010-09-01 13:18:11 UTC

Created attachment 245586 [details]
sqlite-3.7.2-r1.ebuild

This ebuild removes the extensions use flag and builds it  (read: doesn't modify away) by default

Comment 2 m_zwart 2010-09-01 23:34:30 UTC

Did you doublecheck this actually enables extension support for the non-amalgamation (USE=tcl or FEATURES=test) case? My own quick check indicated the switch defaults to off there, but it's possible I was confused.

Comment 3 Johan Bergström 2010-09-10 08:04:04 UTC

Created attachment 246662 [details]
sqlite-3.7.2-r1.ebuild

(In reply to comment #2)
> Did you doublecheck this actually enables extension support for the
> non-amalgamation (USE=tcl or FEATURES=test) case? My own quick check indicated
> the switch defaults to off there, but it's possible I was confused.
> 

You're entirely correct. Upstream isn't consistent here, so here's a new ebuild that adds a flag to configure to make sure its always enabled

Comment 4 Jan Psota 2010-09-27 13:00:00 UTC

Could you consider enabling +unlock-notify too?

It does cost no more packages sqlite depends on
and firefox-4 depends on this.

Comment 5 Johan Bergström 2010-09-27 13:46:32 UTC

(In reply to comment #4)
> Could you consider enabling +unlock-notify too?
> 
> It does cost no more packages sqlite depends on
> and firefox-4 depends on this.
> 

I tried it locally without any issues, lets leave the decision to maintainer though.

Comment 6 Petteri Räty (RETIRED) gentoo-dev

2010-10-31 09:26:54 UTC

(In reply to comment #1)
> 
> This ebuild removes the extensions use flag and builds it  (read: doesn't
> modify away) by default
> 

I don't see why users should not be given the choice to disable extensions when upstream provides such a switch so I just turned it on by default.(In reply to comment #4)

> Could you consider enabling +unlock-notify too?
> 
> It does cost no more packages sqlite depends on
> and firefox-4 depends on this.
> 

Our policy is to stick close to upstream. As far as I can see unlock-notify is not a default upstream. For firefox-4 shouldn't Portage take care of enabling it automatically through proper dependencies?