[gentoo-dev] RFC: emboss.eclass as replacement for embassy.eclass

[justin]

Hi all,

I unbundled a couple of libs from emboss and thus we need a new
installation scheme for the embassy package. For this purpose I rewrote
the embassy eclass  to handle now all emboss and embassy* packages.

Please comment if you have objections for the tree commit

http://git.overlays.gentoo.org/gitweb/?p=proj/sci.git;a=blob_plain;
f=eclass/emboss.eclass;hb=HEAD

thanks justin



signature.asc
Description: OpenPGP digital signature

[gentoo-dev] Last Rites: dev-cpp/libpqpp

[Aaron W. Swenson]

# Aaron W. Swenson  (15 Mar 2011)
# Upstream no longer exists. Superseded by dev-lib/libpqxx.
# Removal 15 Apr 2011 (bug #348359)
dev-cpp/libpqpp

Re: [gentoo-dev] RFC: emboss.eclass as replacement for embassy.eclass

[justin]

Donnie pointed out, that it might be clever to attach what I want to
have reviewed. So here it is.

justin
# Copyright 1999-2004 Gentoo Foundation
# Distributed under the terms of the GNU General Public License v2
# $Header: /var/cvsroot/gentoo-x86/eclass/embassy.eclass,v 1.17 2008/11/03 
22:17:50 ribosome Exp $

# Creator of the original eclass
# Author Olivier Fisette 
#
# Author of the next generation eclass
# Justin Lecher 

# @ECLASS: emboss.eclass
# @MAINTAINER:
# sci-biol...@gentoo.org
# j...@gentoo.org
# @BLURB: Use this to easy install EMBOSS and EMBASSY programs (EMBOSS add-ons).
# @DESCRIPTION:
# The inheriting ebuild should provide EBO_DESCRIPTION before the inherit line.
# KEYWORDS should be set. Additionally "(R|P)DEPEND"encies and other standard
# ebuild Variables can be extended (FOO+="BAR").
# The inheriting ebuild's name must begin with "emboss" or "embassy" and must 
be EAPI=4 conform.

# @ECLASS-VARIABLE: EBO_DESCRIPTION
# @DESCRIPTION:
# Should be set. Completes the describtion of the embassy module as follows:
#
# EMBOSS integrated version of EBO_DESCRIPTION"
#
# Defaults to the upstream name of the module.

# @ECLASS-VARIABLE: EBO_PATCH
# @DEFAULT_UNSET
# @DESCRIPTION: Specify the patch level of EMBOSS. Only available for the 
emboss ebuild.
# The patch wil be fetch from:
#
# ftp://emboss.open-bio.org/pub/EMBOSS/fixes/patches/patch-1-${EBO_PATCH}.gz.
#
# Embassy packages should create one patch package and place it in FILESDIR, 
e.g.
# "files/embassy-iprscan-4.3.1-r2.patch". The patch will be automatically used 
during src_prepare

# @ECLASS-VARIABLE: NO_RECONF
# @DEFAULT_UNSET
# @DESCRIPTION:
# Set this, if you do not want to have eautoreconf be run after patching.

# @ECLASS-VARIABLE: EBO_ECONF
# @DEFAULT_UNSET
# @DESCRIPTION:
# Extra config options passed to econf, similar to EXTRA_ECONF.

EAPI="4"

inherit autotools eutils multilib

HOMEPAGE="http://emboss.sourceforge.net";
LICENSE="LGPL-2 GPL-2"

SLOT="0"
IUSE="mysql pdf png postgres static-libs X "

DEPEND="
dev-libs/expat
dev-libs/libpcre:3
sci-libs/plplot
sys-libs/zlib
mysql? ( dev-db/mysql )
pdf? ( media-libs/libharu )
png? ( media-libs/gd[png] )
postgres? ( dev-db/postgresql-base )
X? ( x11-libs/libXt )
"

RDEPEND="
${DEPEND}
"

DOCS="AUTHORS ChangeLog NEWS README "

if [[ ${PN} == "emboss" ]] ; then
EBOV=${PV/_p*}
DESCRIPTION="The European Molecular Biology Open Software Suite - A 
sequence analysis package"
SRC_URI="ftp://emboss.open-bio.org/pub/EMBOSS/EMBOSS-${EBOV}.tar.gz";
[[ -n ${EBO_PATCH} ]] && SRC_URI+=" 
ftp://${PN}.open-bio.org/pub/EMBOSS/fixes/patches/patch-1-${EBO_PATCH}.gz -> 
${P}.patch.gz"
IUSE+="minimal "
RDEPEND+="
!sys-devel/cons
"
PDEPEND+="
!minimal? (
sci-biology/aaindex
sci-biology/cutg
sci-biology/prints
sci-biology/prosite
sci-biology/rebase
sci-biology/transfac
)
"
S=${WORKDIR}/EMBOSS-${EBOV}
DOCS+="FAQ THANKS "
else
# The EMBASSY package name, retrieved from the inheriting ebuild's name
EN=${PN:8}
# The full name and version of the EMBASSY package (excluding the Gentoo
# revision number)
EF=$(echo ${EN} | tr "[:lower:]" "[:upper:]")-${PV}
EBO_DESCRIPTION=${EBO_DESCRIPTION:=${EN}}
DESCRIPTION="EMBOSS integrated version of ${EBO_DESCRIPTION}"
SRC_URI="ftp://emboss.open-bio.org/pub/EMBOSS/${EF}.tar.gz -> 
embassy-${PN:8}-${PV}.tar.gz"

DEPEND+=">=sci-biology/emboss-6.3.1_p4[mysql=,pdf=,png=,postgres=,static-libs=,X=]
 "

S=${WORKDIR}/${EF}
fi

# @FUNCTION: emboss_src_prepare
# @USAGE:
# @RETURN:
# @MAINTAINER:
# @DESCRIPTION:
# Does three things
#
#  1. Patches EMBOSS if EBO_PATCH is set
#  2. Patches with "${FILESDIR}"/${PF}.patch, of present
#  3. Runs eautoreconf unless NO_RECONF is set
#

emboss_src_prepare() {
[[ ${PN} == emboss ]] && [[ -n ${EBO_PATCH} ]] && epatch 
"${WORKDIR}"/${P}.patch
[[ -f "${FILESDIR}"/${PF}.patch ]] && epatch "${FILESDIR}"/${PF}.patch
[[ -n ${NO_RECONF} ]] || eautoreconf
}

# @FUNCTION: emboss_src_prepare
# @USAGE:
# @RETURN:
# @MAINTAINER:
# @DESCRIPTION:
# runs econf with following options. Extra options can be passed by setting 
EBO_ECONF
#
#  $(use_with X x)
#  $(use_with png pngdriver "${EPREFIX}/usr")
#  $(use_with pdf hpdf "${EPREFIX}/usr")
#  $(use_with mysql mysql "${EPREFIX}/usr/bin/mysql_config")
#  $(use_with postgres postgresql "${EPREFIX}/usr/bin/pg_config")
#  $(use_enable amd64 64)
#  $(use_enable static-libs static)
#  --enable-large
#  --without-java
#  --enable-systemlibs
#  ${EBO_ECONF}

emboss_src_co

Re: [gentoo-dev] RFC: emboss.eclass as replacement for embassy.eclass

[Tomáš Chvátal]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Dne 15.3.2011 21:51, justin napsal(a):
> Donnie pointed out, that it might be clever to attach what I want to
> have reviewed. So here it is.
> 
> justin
EAPI CAN'T BE DEFINED IN ECLASS!
this should be craved into stones :)

Some whitespace is funny.

look on XFCONF_AUTORECONF or XORG_AUTORECONF variable, just create
EBO_AUTORECONF with yes by default and handle it correctly.

i am seriously hurt by $(use_enable amd64 64)

why are not those functions exported for embassy ebuild?

Cheers

Tom
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.17 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk1/2UcACgkQHB6c3gNBRYd7owCgsP74VUYO0oQ5w2UNoYNXS0aA
zoUAnRMrGLAA/8kj4hX7Jc+QCb5cPJdb
=lz/B
-END PGP SIGNATURE-

Re: [gentoo-dev] RFC: emboss.eclass as replacement for embassy.eclass

[Mike Frysinger]

> # Copyright 1999-2004 Gentoo Foundation

s/2004/2011/

have you looked at the man page output for this to make sure the white spacing 
you're putting into the comments properly carries over ?

> # @ECLASS-VARIABLE: EBO_DESCRIPTION
> # @DESCRIPTION:
> # Should be set. Completes the describtion of the embassy module as follows:

i dont understand what this is trying to say (ignoring the spelling mistake in 
description)

> # EMBOSS integrated version of EBO_DESCRIPTION"

i see just one quote.  it needs a friend.

> # @ECLASS-VARIABLE: EBO_PATCH
> # @DEFAULT_UNSET
> # @DESCRIPTION: Specify the patch level of EMBOSS. Only available for the 
emboss ebuild.
> # The patch wil be fetch from:

wil -> will

> # @ECLASS-VARIABLE: NO_RECONF
> # @DEFAULT_UNSET
> # @DESCRIPTION:
> # Set this, if you do not want to have eautoreconf be run after patching.

please namespace all variables consistently

> # @ECLASS-VARIABLE: EBO_ECONF

EBO_EXTRA_ECONF is better for consistency sake too

> EAPI="4"

erp!

> HOMEPAGE="http://emboss.sourceforge.net";

could do with a trailing "/"

> IUSE="mysql pdf png postgres static-libs X "

no trailing space.  let the person appending the variable add the space:
IUSE+=" moo"

> DOCS="AUTHORS ChangeLog NEWS README "

no need for that trailing space, and usually DOCS is left for end ebuilds isnt 
it ?  eclasses should append it.

> 
> if [[ ${PN} == "emboss" ]] ; then
>   EBOV=${PV/_p*}
>   DESCRIPTION="The European Molecular Biology Open Software Suite - A 
sequence analysis package"
>   SRC_URI="ftp://emboss.open-bio.org/pub/EMBOSS/EMBOSS-${EBOV}.tar.gz";
>   [[ -n ${EBO_PATCH} ]] && SRC_URI+=" ftp://${PN}.open-
bio.org/pub/EMBOSS/fixes/patches/patch-1-${EBO_PATCH}.gz -> ${P}.patch.gz"
>   IUSE+="minimal "
>   RDEPEND+="
>   !sys-devel/cons
>   "
>   PDEPEND+="
>   !minimal? (
>   sci-biology/aaindex
>   sci-biology/cutg
>   sci-biology/prints
>   sci-biology/prosite
>   sci-biology/rebase
>   sci-biology/transfac
>   )
>   "
>   S=${WORKDIR}/EMBOSS-${EBOV}
>   DOCS+="FAQ THANKS "

this doesnt make much sense to me.  shouldnt this all be in the emboss ebuild 
instead of the eclass ?

>   # The EMBASSY package name, retrieved from the inheriting ebuild's name
>   EN=${PN:8}

should there be some checking here to make sure it starts with "embassy-" or 
whatever you're expecting ?

>   # The full name and version of the EMBASSY package (excluding the Gentoo
>   # revision number)
>   EF=$(echo ${EN} | tr "[:lower:]" "[:upper:]")-${PV}

ugh, but i guess we cant do much until we have newer bash

>   EBO_DESCRIPTION=${EBO_DESCRIPTION:=${EN}}

tricky, but wrong.  i guess you actually want:
: ${EBO_DESCRIPTION:=${EN}}

>   
> DEPEND+=">=sci-biology/emboss-6.3.1_p4[mysql=,pdf=,png=,postgres=,static-
libs=,X=] "

space should be at the start of the string, not the end

> # @FUNCTION: emboss_src_prepare
> # @USAGE:
> # @RETURN:
> # @MAINTAINER:

i dont think these three tags make sense.  not even sure if this last one is 
even allowed.  best to just punt.

same goes for all the other funcs below.

> emboss_src_prepare() {
>   [[ ${PN} == emboss ]] && [[ -n ${EBO_PATCH} ]] && epatch 
"${WORKDIR}"/${P}.patch

guess that would be moved to the emboss ebuild

>   [[ -f "${FILESDIR}"/${PF}.patch ]] && epatch "${FILESDIR}"/${PF}.patch

technically, no need to quote FILESDIR here as the [[...]] takes care of it

> #  $(use_with png pngdriver "${EPREFIX}/usr")
> #  $(use_with pdf hpdf "${EPREFIX}/usr")
> #  $(use_with mysql mysql "${EPREFIX}/usr/bin/mysql_config")
> #  $(use_with postgres postgresql "${EPREFIX}/usr/bin/pg_config")

do you really need full paths here ?  are the configure scripts that screwed 
up ?  ideally libs would be found via .pc files.

> #  $(use_enable amd64 64)

oh god it burns so bad when i poo

> emboss_src_install() {
>   default
>   mv "${ED}"/usr/share/EMBOSS/doc/* "${ED}"/usr/share/doc/${PF}/
>   rm -rf "${ED}"/usr/share/EMBOSS/doc

does the configure script respect --docdir= ?  then you wouldnt need this 
clean up logic ...

> [[ ${PN} == embassy ]] || EXPORT_FUNCTIONS src_prepare src_configure 
src_install

this $PN logic should get punted
-mike


signature.asc
Description: This is a digitally signed message part.

[gentoo-dev] Re: RFC: emboss.eclass as replacement for embassy.eclass

[Duncan]

Mike Frysinger posted on Tue, 15 Mar 2011 18:06:36 -0400 as excerpted:

>> # The patch wil be fetch from:
> 
> wil -> will

While you're at it, s/fetch/fetched/

IOW:

# The patch will be fetched from:

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman

[gentoo-dev] Why a betagarden overlay

[Sebastian Pipping]

Hello!


First: If betagarden were a normal overlay, I would not be writing about
it here.

If you're in a hurry just skip the introduction and jump down to section
"Betagarden overlay".


Introduction

The betagarden overlay has been around for a while.  I always wanted to
write about its purpose and invite you to collaboration but I haven't
got to it before.

I understand betagarden as a third place supplementing the Gentoo main
tree (sometimes known as "gentoo-x86" or "portage") and the special
overlay of Project Sunrise [1].  It fills a gap that these other two
repositories leave open.  Let's have a look:


  Gentoo Main tree
  
  - Post-publishing review
  - Territorial write access: Gentoo Developers (only)
  - Full write access: Gentoo QA maybe?
  - High quality standards

  sunrise overlay
  ---
  - Pre-publishing review
  - Reduced write access: Anyone passing a simple test [2]
  - Full write access: Project Sunrise developers (only)
  - High quality standards


>From these lists a few things can be observed:

 1. Both trees require high quality from ebuilds.  This includes
- Full integration with Gentoo (menu entries, init scripts, etc.)
- Cleaning the ebuild
- Support for LDFLAGS
- ...

 2. Gentoo developers who are not fully committed to sunrise
do not have full write access to it


-->  Wouldn't it be nice to have a place where polishing is optional
 (as long as the ebuilds are still safe) with more liberal write
 access?

But there's another group of repositories that I would also like to have
a look at:


Gentoo developer overlays
-
When you go to  you see them
instantly - most Gentoo devs have one:

  dev/aballier.git  Developer overlay  Alexis Ballier
  dev/alexxy.gitDeveloper overlay  Alexey Shvetsov
  dev/anarchy.git   Developer overlay  Jory Pratt
  dev/angelos.git   Developer overlay  Christoph Mende
  [..]

Many of these overlays currently combine two groups of ebuilds:

 - Stuff useful to themselves, only

 - Stuff useful to a wider audience
   (that they didn't feel like adding to the Gentoo main tree)


With such a mix it often makes no sense for somebody else to keep that
overlay installed over time.

-->  Wouldn't it be nice to have the stuff useful to others in a more
 central place (and reduce your developer to stuff that basically is
 only interesting to you)?

Hollow and I (sping) have been trying to do that with our overlays:
moving stuff useful to others over to betagarden, a shared overlay.



Betagarden overlay
==
So now that I have shared my view on the Gentoo main tree, the sunrise
overlay and developer overlays let me summarize how betagarden fits in:

 - Full write access to all Gentoo Developers
   That means more freedom than in the main tree or sunrise.

 - Reduced (but essential) quality standards
   (hence the "beta" in "betagarden")

 - Keeping really useful stuff off the developer overlays


How to join
---
All devs have write access to betagarden already.

 1. Clone git+ssh://g...@git.overlays.gentoo.org/proj/betagarden.git

 2. Add yourself to the betagar...@gentoo.org alias:
# ssh dev.gentoo.org
# nano -w /var/mail/alias/misc/betagarden

 3. Start adding (or moving over existing) ebuilds

If you have trouble pushing commits please contact .

In bugzilla, you can assign bugs to  by now.


Expected criticism
--
I expect some of you to be worried: does that mean people stop adding
quality ebuilds to the Gentoo main tree and move on to betagarden?

No.  If an ebuild is really important it belongs into the main tree.  In
that case someone will take the time to ensure high quality standards
and move it from betagarden to the main tree.


I hope some of you do see something good in this project.

Thanks for your interest,



Sebastian


[1] http://overlays.gentoo.org/proj/sunrise
[2] http://overlays.gentoo.org/proj/sunrise/wiki/HowToCommit#Password

[gentoo-dev] Re: Last rites: net-p2p/bittorrent

[Ryan Hill]

On Sat, 12 Mar 2011 00:27:21 -0600
Ryan Hill  wrote:

> # Ryan Hill  (12 Mar 2011)
> # Mask net-p2p/bittorrent for removal 20110412 (#336166)
> # Dead upstream. Requires wxpython-2.6 which is being removed.
> # Doesn't build with python-2.6.  Doesn't start.
> net-p2p/bittorrent

Instead, we're going to drop the 5.x series and downgrade to 4.4.0, which was
the last version to use pygtk rather than wxpython.  

Does anyone want to maintain it after I fix the outstanding issues?


-- 
fonts, gcc-porting,  it makes no sense how it makes no sense
toolchain, wxwidgets   but i'll take it free anytime
@ gentoo.orgEFFD 380E 047A 4B51 D2BD C64F 8AA8 8346 F9A4 0662


signature.asc
Description: PGP signature

Re: [gentoo-dev] RFC: emboss.eclass as replacement for embassy.eclass

[Jeroen Roovers]

On Tue, 15 Mar 2011 18:06:36 -0400
Mike Frysinger  wrote:

> > # The full name and version of the EMBASSY package
> > (excluding the Gentoo # revision number)
> > EF=$(echo ${EN} | tr "[:lower:]" "[:upper:]")-${PV}
> 
> ugh, but i guess we cant do much until we have newer bash

bash 4.1 has been stable for half a year now. Not long enough?

EF="${EN^^}"


 jer

Re: [gentoo-dev] RFC: emboss.eclass as replacement for embassy.eclass

[Jeremy Olexa]

On 03/15/2011 10:00 PM, Jeroen Roovers wrote:

On Tue, 15 Mar 2011 18:06:36 -0400
Mike Frysinger  wrote:


# The full name and version of the EMBASSY package
(excluding the Gentoo # revision number)
EF=$(echo ${EN} | tr "[:lower:]" "[:upper:]")-${PV}


ugh, but i guess we cant do much until we have newer bash


bash 4.1 has been stable for half a year now. Not long enough?

EF="${EN^^}"


Not until PMS specifies[1] something other than: "The interpreter is 
assumed to be GNU bash, version 3.2 or later"


[1]: http://dev.gentoo.org/~ulm/pms/head/pms.html#x1-650007

-Jeremy

Re: [gentoo-dev] Re: Quantity of open bugs

[Jeroen Roovers]

On Sat, 12 Mar 2011 15:45:34 +0100
Diego Elio Pettenò  wrote:

> And you can challenge that with the council if you wish as I'm
> weighting that in as part of QA, thank you very much.

Again that language. Why do you talk like that?


 jer

Re: [gentoo-dev] RFC: emboss.eclass as replacement for embassy.eclass

[justin]

On 15/03/11 22:25, Tomáš Chvátal wrote:
> Dne 15.3.2011 21:51, justin napsal(a):
>> Donnie pointed out, that it might be clever to attach what I want to
>> have reviewed. So here it is.
> 
>> justin
> EAPI CAN'T BE DEFINED IN ECLASS!
> this should be craved into stones :)

I didn't see this mentioned anywhere in the docs. But okay, this is fixable.

> 
> Some whitespace is funny.

I love funny whitespaces.

> 
> look on XFCONF_AUTORECONF or XORG_AUTORECONF variable, just create
> EBO_AUTORECONF with yes by default and handle it correctly.

Thanks for that hint. I will change it accordingly.

> 
> i am seriously hurt by $(use_enable amd64 64)

Where is the problem? The purpose of the eclass is not to repeat the
same code 20 times. Why can't this be here?

> 
> why are not those functions exported for embassy ebuild?

Just a Meta ebuild to install all modules (emabassy-*).

> 
> Cheers
> 
> Tom

thanks justin



signature.asc
Description: OpenPGP digital signature