Re: [gentoo-dev] devfs is dead, let's move on

2005-07-11 Thread Martin Schlemmer
On Sat, 2005-07-09 at 20:34 +0200, Richard Fish wrote:
 I.o.w. is it still necessary to have RC_DEVICE_TARBALL=yes as a
 default or can we move to a pure udev system and change the default to
 no.
   
 
 I've been running my boxes successfully with no since the option
 showed up just fine :)
 
 
 
 
 I think people is under a misconception about this option and ... you
 really only need to enable this for a driver that is not sysfs aware
 (nvidia comes to mind - any others?), or if you have some custom nodes
 in /dev that you cannot do via udev ...  And I am pretty sure (correct
 me if I am wrong) that all (or most?) in-kernel drivers are sysfs aware,
 and only a handful outside are not.
 
   
 
 
 Well, I do have a small issue with the software RAID (md) driver, in
 that when autodetection is not performed by the driver (due to either
 being a module or booting the system through an initramfs), no sysfs
 entries or device nodes are created.
 
 Normally my RAID system is brought up inside my initramfs with static
 nodes, so this really only affects my recovery CD, where I need to run:
 
 for d in 0 1 2 3; do
 /sbin/mdadm --assemble --config=partitions --auto=md
 --super-minor=$d /dev/md$d /dev/null 21
 done
 
 Maybe something similar will be required in /sbin/rc, like you currently
 do for LVM and the device mapper?  It isn't a critical problem
 though...I am pretty sure there are only a few Gentoo users who will
 ever see this...maybe as few as 1!!!
 

Mike, what do you think?  This viable?  We could maybe add an init addon
for md, and move the lvm/whatever stuff to that as well?


-- 
Martin Schlemmer
Gentoo Linux Developer, Desktop/System Team Developer
Cape Town, South Africa



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


[gentoo-dev] G/FBSD compatibility: root:root and cp -a

2005-07-11 Thread Diego 'Flameeyes' Pettenò
Ok, to help devels know which are the requisites of Gentoo/FreeBSD porting (as 
we're going near to a release, I hope), I think it's better telling here what 
we hope you all will avoid (and we must replace when keywording).
Those problems are also valid for ppc-macos AFAIK, so you won't help just us, 
but also ppc-macos project.

The first one is the root:root problem (there's no root group in *BSD and 
BSD-based systems); the solution is usually using 0:0 to chown. If nobody is 
going to disagrees with this, I think this solution is what we should use.

The other problem is with cp -a: -a is a GNU option which is not supported by 
non-GNU cp commands. As the man pages says, it's equivalent to -dpPR. 
Unfortuately -dpPR is not portable in itself, but usually what you need is 
-pPR which works just fine usually.

I'm not going to have a big immediate change of all the ebuilds in portage, 
but at least I hope new ebuilds doesn't get added with those problems in 
them :)

HTH,
-- 
Diego Flameeyes Pettenò
Gentoo Developer - http://dev.gentoo.org/~flameeyes/
(Gentoo/FreeBSD, Video, Gentoo/AMD64, Sound, PAM)


pgpMmczU3W1Hp.pgp
Description: PGP signature


[gentoo-dev] make and wrappers

2005-07-11 Thread Diego 'Flameeyes' Pettenò
Following the discussion me and Azarah had about make commands, I've started 
thinking of a way to fix this without being too intrusive and I found a bit 
of a solution following Az's suggestion to use a wrapper script.

Let's see: we already have an emake script, but using it for make install is 
a no-go because it uses -jX from make.conf and that's not good.
A solution can be to improve emake wrapper to check if install is in its 
commandline, in which case it can automatically add a -j1 to be safe.

The the ebuilds can just use emake install to do the work, and let the profile 
define the default MAKE variable (gmake or make, we'll see in the mean time).
BDEPEND in this case should not list any kind of make.
If somethign needs exactly GNU make, it will list gmake in BDEPEND 
(sys-libs/make), and will set MAKE=gmake in the ebuild, so that emake will 
use it.
The same when we'll have something needing bsdmake, we'll add a 
dev-util/bsdmake ebuild sometime in the future, and then just BDEPEND on || 
( sys-apps/freebsd-ubin dev-util/bsdmake ) [or just bsdmake depending if the 
bsdmake ebuild will be used on g/fbsd, too].

Comments?
-- 
Diego Flameeyes Pettenò
Gentoo Developer - http://dev.gentoo.org/~flameeyes/
(Gentoo/FreeBSD, Video, Gentoo/AMD64, Sound, PAM)


pgpPTv3nPmONg.pgp
Description: PGP signature


Re: [gentoo-dev] make and wrappers

2005-07-11 Thread Ciaran McCreesh
On Mon, 11 Jul 2005 14:20:20 +0200 Diego 'Flameeyes' Pettenò
[EMAIL PROTECTED] wrote:
| Let's see: we already have an emake script, but using it for make
| install is  a no-go because it uses -jX from make.conf and that's not
| good. A solution can be to improve emake wrapper to check if install
| is in its  commandline, in which case it can automatically add a -j1
| to be safe.

Why not check which phase you're in instead?

-- 
Ciaran McCreesh

-- 
gentoo-dev@gentoo.org mailing list



Re: [gentoo-dev] make and wrappers

2005-07-11 Thread Henrik Brix Andersen
On Mon, 2005-07-11 at 14:20 +0200, Diego 'Flameeyes' Pettenò wrote:
 Let's see: we already have an emake script, but using it for make install 
 is 
 a no-go because it uses -jX from make.conf and that's not good.
 A solution can be to improve emake wrapper to check if install is in its 
 commandline, in which case it can automatically add a -j1 to be safe.

What is the problem with using -jX for `make install`?

Regards,
Brix
-- 
Henrik Brix Andersen [EMAIL PROTECTED]
Gentoo Metadistribution | Mobile computing herd


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


Re: [gentoo-dev] make and wrappers

2005-07-11 Thread Diego 'Flameeyes' Pettenò
On Monday 11 July 2005 14:32, Ciaran McCreesh wrote:
 Why not check which phase you're in instead?
That also can be done, true.

-- 
Diego Flameeyes Pettenò
Gentoo Developer - http://dev.gentoo.org/~flameeyes/
(Gentoo/FreeBSD, Video, Gentoo/AMD64, Sound, PAM)


pgpC1VHMPecuU.pgp
Description: PGP signature


[gentoo-dev] ebookmerge unmask and ebook-*.ebuilds remove

2005-07-11 Thread José Alberto Suárez López
Hola :)

after a week w/o bugs for ebookmerge i will unmask it and i will remove
all the ebook-*.ebuilds if nobody say the secret word ;)

regards


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


Re: [gentoo-dev] make and wrappers

2005-07-11 Thread Ciaran McCreesh
On Mon, 11 Jul 2005 14:32:06 +0200 Henrik Brix Andersen
[EMAIL PROTECTED] wrote:
| On Mon, 2005-07-11 at 14:20 +0200, Diego 'Flameeyes' Pettenò wrote:
|  Let's see: we already have an emake script, but using it for make
|  install is  a no-go because it uses -jX from make.conf and that's
|  not good. A solution can be to improve emake wrapper to check if
|  install is in its  commandline, in which case it can automatically
|  add a -j1 to be safe.
| 
| What is the problem with using -jX for `make install`?

Anything using a standardish autotools install can potentially explode
when parallelised.

-- 
Ciaran McCreesh

-- 
gentoo-dev@gentoo.org mailing list



Re: [gentoo-dev] ebookmerge unmask and ebook-*.ebuilds remove

2005-07-11 Thread José Alberto Suárez López
El lun, 11-07-2005 a las 14:58 +0200, Henrik Brix Andersen escribió:
 On Mon, 2005-07-11 at 14:53 +0200, José Alberto Suárez López wrote:
  after a week w/o bugs for ebookmerge i will unmask it and i will remove
  all the ebook-*.ebuilds if nobody say the secret word ;)
 
 A week? Any particular reason not to follow policy and wait 30 days?

a few :

* this app will not broke anything in the system or userside if it
fails.
* i will go in vacations soon
* and i forget about 30 days sorry :P


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


Re: [gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Henrik Brix Andersen
On Mon, 2005-07-11 at 15:13 +0200, Karl Trygve Kalleberg wrote:
 over the course of the next few days, we will finally be replacing the
 'mozilla' useflag in all the Java runtimes with a new USE flag, called
 'browserplugin'.

Yay!

./Brix
-- 
Henrik Brix Andersen [EMAIL PROTECTED]
Gentoo Metadistribution | Mobile computing herd


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


[gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Karl Trygve Kalleberg
Hi gang,

over the course of the next few days, we will finally be replacing the
'mozilla' useflag in all the Java runtimes with a new USE flag, called
'browserplugin'.

The change means that in order to have continued Java browser plugin
support, you must add the useflag 'browserplugin' to the USE= line
in your /etc/make.conf file.

The reason for the change is that:

) all browser plugins now work with more than just the Mozilla
  webbrowser, thus naming it 'mozilla' has become rather inaccurate.

) the 'mozilla' useflag is primarily used for embedding Mozilla
  into other applications.

) it should be possible to enable the Java browser plugin system-wide,
  without turning on Mozilla embedding for random other applications.
  As such, the 'mozilla' and 'mozilla' useflags are different, but
  at the present time have the same name, which is rather confusing.


(For now, the new useflag is not a global useflag, so as to not upset
certain elements in the developer community. It is marked as a local
useflag for all packages supporting it. However, the difference is
merely academic, but I'm sure a flamewar will erupt over this as usual.)

Refer to [1] for tracking of this issue.


We will keep supporting the 'mozilla' useflag for a minimum period of a
month, and you will be amply warned about the change when installing or
upgrading a Java runtime.


-- Karl T


[1] https://bugs.gentoo.org/show_bug.cgi?id=94056
-- 
gentoo-dev@gentoo.org mailing list



Re: [gentoo-dev] make and wrappers

2005-07-11 Thread Kito


On Jul 11, 2005, at 7:32 AM, Henrik Brix Andersen wrote:


On Mon, 2005-07-11 at 14:20 +0200, Diego 'Flameeyes' Pettenò wrote:

Let's see: we already have an emake script, but using it for make  
install is

a no-go because it uses -jX from make.conf and that's not good.
A solution can be to improve emake wrapper to check if install  
is in its

commandline, in which case it can automatically add a -j1 to be safe.



What is the problem with using -jX for `make install`?



THink about what happens in most `make install` phases, files get put  
in their DESTROOT, and many times permissions are set etc. when its  
done in parallel the chances of the makefile trying to operate on non- 
existent targets is quite high, and things start to explode.



Regards,
Brix
--
Henrik Brix Andersen [EMAIL PROTECTED]
Gentoo Metadistribution | Mobile computing herd




--
gentoo-dev@gentoo.org mailing list



Re: [gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Aron Griffis
KarlTK wrote:   [Mon Jul 11 2005, 09:13:03AM EDT]
 over the course of the next few days, we will finally be replacing
 the 'mozilla' useflag in all the Java runtimes with a new USE flag,
 called 'browserplugin'.

I think I can speak for the mozilla team... this change looks good.

Regards,
Aron

--
Aron Griffis
Gentoo Linux Developer



pgpzzTOse0cYj.pgp
Description: PGP signature


Re: [gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Stefan Schweizer
Am Montag 11 Juli 2005 15:13 schrieb Karl Trygve Kalleberg:
 Hi gang,

 over the course of the next few days, we will finally be replacing the
 'mozilla' useflag in all the Java runtimes with a new USE flag, called
 'browserplugin'.

Are you expecting the other browser-plugin-ebuilds to follow this step?
acroread, gplflash, netscape-flash, netscape-plugger, mozplugger, 
mplayerplug-in?

Are you planning to make that a global use flag or will it just be used 
locally for java?

I vote for global use flag and useage in all browserplugins.

Regards,
  Stefan


pgp5S6TBhmIp1.pgp
Description: PGP signature


Re: [gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Henrik Brix Andersen
On Mon, 2005-07-11 at 18:38 +0200, Stefan Schweizer wrote:
 I vote for global use flag and useage in all browserplugins.

Yes, I second that.

./Brix
-- 
Henrik Brix Andersen [EMAIL PROTECTED]
Gentoo Metadistribution | Mobile computing herd


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


Re: [gentoo-dev] GET ME OF THIS LIST!

2005-07-11 Thread Lance Albertson
[EMAIL PROTECTED] wrote:
 Hi,
 
 I don't want to be an ignorant asss, but it is impossible for me to
 unsubscribe by myself right now.
 
 I currently only have webmail access, and the webmail program of my ISP
 will not
 accept an e-mail address in the form of gentoo-dev
 [EMAIL PROTECTED]
 for whatever reason, and since a reply to just that address is required
 to get me out of here,
 I can't do it.

Gentoo's Infrastructure has its own product in bugzilla for these types
of requests. Please create a new bug there and it will be seen rather
than spamming the list.

Cheers,

-- 
Lance Albertson [EMAIL PROTECTED]
Gentoo Infrastructure | Operations Manager

---
GPG Public Key:  http://www.ramereth.net/lance.asc
Key fingerprint: 0423 92F3 544A 1282 5AB1  4D07 416F A15D 27F4 B742

ramereth/irc.freenode.net


signature.asc
Description: OpenPGP digital signature


Re: [gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Jeremy Maitin-Shepard
One issue I see with a `browserplugin' USE flag is that there may at
some point be web browsers in the tree that support plugins that are not
Mozilla-compatible.  Although there are multiple Mozilla-compatible
browsers (mainly browsers which use Mozilla code or have Mozilla
embedded), the plugins in question are nonetheless essentially
``Mozilla'' plugins.  Thus it is probably better for the USE flag to be
called `mozillaplugin', to allow for the future introduction of other
browser-specific plugin USE flags.  (It would be undesirable for ebuilds
supporting multiple types of web browser plugins to, if the single
`browserplugin' USE flag is enabled, install plugins for all of the
supported browsers that portage says are installed.)

-- 
Jeremy Maitin-Shepard
-- 
gentoo-dev@gentoo.org mailing list



Re: [gentoo-dev] ebookmerge unmask and ebook-*.ebuilds remove

2005-07-11 Thread Henrik Brix Andersen
On Mon, 2005-07-11 at 15:10 +0200, José Alberto Suárez López wrote:
 * this app will not broke anything in the system or userside if it
 fails.
 * i will go in vacations soon
 * and i forget about 30 days sorry :P

Those are not really valid points for breaking the policy - but it's too
late now, I see you've already marked the package stable after being in
the tree for only 7 days. So much for QA...

./Brix
-- 
Henrik Brix Andersen [EMAIL PROTECTED]
Gentoo Metadistribution | Mobile computing herd


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


Re: [gentoo-dev] ebookmerge unmask and ebook-*.ebuilds remove

2005-07-11 Thread Robert Paskowitz
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

 a few :
 
 * this app will not broke anything in the system or userside if it
 fails.

But users will get angry if a package is not working and start filling bugs.

 * i will go in vacations soon

All the more reason _not_ to mark stable, since any bugs will then be
unhandled.

 * and i forget about 30 days sorry :P
Consider a todo list or calender, paper or software or otherwise.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (Darwin)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFC0sV/ZwjIiODIZ4oRAjjCAJ9C/38MmzOML0d0/LPqy43Cx0uR6QCfSPCd
Oj9iROXGXUv+tg4YwhHbNUM=
=mjhJ
-END PGP SIGNATURE-
-- 
gentoo-dev@gentoo.org mailing list



Re: [gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Diego 'Flameeyes' Pettenò
On Monday 11 July 2005 20:56, Jeremy Maitin-Shepard wrote:
 Thus it is probably better for the USE flag to be
 called `mozillaplugin', to allow for the future introduction of other
 browser-specific plugin USE flags.
profiles/use.local.desc:app-text/djvu:nsplugin - Builds plugins for Netscape 
compatible browsers

We already have this that is the same at the end.

-- 
Diego Flameeyes Pettenò
Gentoo Developer - http://dev.gentoo.org/~flameeyes/
(Gentoo/FreeBSD, Video, Gentoo/AMD64, Sound, PAM)


pgpeDjoBEIJbk.pgp
Description: PGP signature


[gentoo-dev] New MySQL doc

2005-07-11 Thread Chris White
Here's the initial devspace draft of the new MySQL draft I've been working on:

http://dev.gentoo.org/~chriswhite/mysql.html

Comments, etc are welcome.

Chris White


pgpuXmAvblVwE.pgp
Description: PGP signature


Re: [gentoo-dev] New MySQL doc

2005-07-11 Thread John Myers
On Monday 11 July 2005 21:38, Chris White wrote:
 http://dev.gentoo.org/~chriswhite/mysql.html

 Comments, etc are welcome.

You might want to make
Code listing 4.19: Finding our guest user in the user table
a little narrower, as it makes the page much wider than my 1,280 pixel wide 
screen can show.

I would suggest 
SELECT Host, User, Password FROM user WHERE User = 'guest';
rather than
SELECT * FROM user WHERE User = 'guest';
which will make the output much, much smaller. You might want to mention the 
other query in passing for the user to try on their own, however.


pgpX5fWFA9EGF.pgp
Description: PGP signature


Re: [gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Aron Griffis
Diego 'Flameeyes' Pettenò wrote:[Mon Jul 11 2005, 03:25:22PM EDT]
 profiles/use.local.desc:app-text/djvu:nsplugin - Builds plugins for Netscape 
 compatible browsers
 
 We already have this that is the same at the end.

This looks like the right thing to do instead of the browserplugin
USE-flag.  Karl, does this make sense from your perspective for the
java stuff?

Regards,
Aron

--
Aron Griffis
Gentoo Linux Developer



pgpS5vPFdFhef.pgp
Description: PGP signature


Re: [gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Aron Griffis
Stefan Schweizer wrote: [Mon Jul 11 2005, 12:38:46PM EDT]
 Are you expecting the other browser-plugin-ebuilds to follow this step?
 acroread

makes sense here

 gplflash, netscape-flash, netscape-plugger, mozplugger, 
 mplayerplug-in?

These are all plugins by definition, there's nothing to do here.

Regards,
Aron

--
Aron Griffis
Gentoo Linux Developer



pgpr4Wf76l8Kq.pgp
Description: PGP signature


Re: [gentoo-dev] New MySQL doc

2005-07-11 Thread John Myers
On Monday 11 July 2005 21:38, Chris White wrote:
 http://dev.gentoo.org/~chriswhite/mysql.html

 Comments, etc are welcome.
Sorry for posting twice, but
Code listing 4.11: REVOKE format
is incorrect. It probably should say
REVOKE [privleges] ON database.* FROM '[user]'@'[host]';
rather than
PRIVLEGES [privleges] ON database.* FROM '[user]'@'[host]';


pgpBmcq6zi22G.pgp
Description: PGP signature


Re: [gentoo-dev] New MySQL doc

2005-07-11 Thread Julien Allanos
Chris White wrote:
 Here's the initial devspace draft of the new MySQL draft I've been working on:
 
 http://dev.gentoo.org/~chriswhite/mysql.html
 
 Comments, etc are welcome.
 
 Chris White

Nice guide! You might mention phpmyadmin as well, as it's a popular
web-based gui to mysql.
-- 
dju`
Julien Allanos
Gentoo Developer (web-apps)
-- 
gentoo-dev@gentoo.org mailing list



Re: [gentoo-dev] New MySQL doc

2005-07-11 Thread Francesco R
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
 
Chris White wrote:

Here's the initial devspace draft of the new MySQL draft I've been
working on:

http://dev.gentoo.org/~chriswhite/mysql.html

Comments, etc are welcome.

Chris White

one reminder:
Code listing 1.3: MySQL configuration probably will change for
mysql-4.1 and better (when unmasked)


and having the possibility to scroll the screen,

Code listing 4.19: Finding our guest user in the user table
mysql SELECT * FROM user WHERE User = 'guest';

may become

Code listing 4.19: Finding our guest user in the user table
mysql SELECT * FROM user WHERE User = 'guest' \G

to make it more readable. See J.M. email too


the default my.cnf has the following line
bind-address= 127.0.0.1
that actually restrict the server access to local  machine only. You
may want to signal this in the Granting Privleges with GRANT section.


Last but not last, thankyou - nice intro document
Francesco R.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
 
iD8DBQFC0tyw1wNBTLVPMuARAroiAJsHl8+M1cOD0Yq2Zlti8kVaMxDnIQCdFIke
NS4BndQanBMCysEuApHAwG0=
=dy9e
-END PGP SIGNATURE-

-- 
gentoo-dev@gentoo.org mailing list



Re: [gentoo-dev] ebookmerge unmask and ebook-*.ebuilds remove

2005-07-11 Thread José Alberto Suárez López
El lun, 11-07-2005 a las 20:57 +0200, Henrik Brix Andersen escribió:
 On Mon, 2005-07-11 at 15:10 +0200, José Alberto Suárez López wrote:
  * this app will not broke anything in the system or userside if it
  fails.
  * i will go in vacations soon
  * and i forget about 30 days sorry :P
 
 Those are not really valid points for breaking the policy - but it's too
 late now, I see you've already marked the package stable after being in
 the tree for only 7 days. So much for QA...
 
 ./Brix

is masked again... don't kill me please...


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


Re: [gentoo-dev] Notice: Changing mozilla useflag to browserplugin for all Java runtimes

2005-07-11 Thread Karl Trygve Kalleberg
Stefan Schweizer wrote:

 Are you expecting the other browser-plugin-ebuilds to follow this step?
 acroread, gplflash, netscape-flash, netscape-plugger, mozplugger, 
 mplayerplug-in?

As Aaron points out, mplayerplug-in, mozplugger, netscape-plugger,
netscape-flash and gplflash do not have the mozilla useflag. They are
pure plug-ins, so the useflag doesn't make sense here.

For acroread, I suggest we make the same change.

 Are you planning to make that a global use flag or will it just be used 
 locally for java?

We can add it locally to acroread, too, if people are worried about
having it as a global USE flag.

 I vote for global use flag and useage in all browserplugins.

That would indeed be very attractive (apart from it having to be
global, which is not strictly necessary).

-- Karl T
-- 
gentoo-dev@gentoo.org mailing list