we will force all metadata.xml files have strict order of tags: first
<herd/> then other tags. Currently there are about 200 ebuilds with
different order http://bugs.gentoo.org/show_bug.cgi?id=279206#c4 .
Hans's suggestion http://bugs.gentoo.org/show_bug.cgi?id=279206#c7 works
but it is not nice and prohibits another order: herd, other tags, herd.
Personally I'd kept dtd simple and forced ordering.
So actually two questions here:
1. How should we fix dtd?
2. Are there any problems if we fix dtd first and let maintainers fix
metadata.xml on bumps (iow, what will became broken if metadata.xml
became not valid)?
--
Peter.
12-15-2009, 04:16 PM
Alex Alexander
metdata.dtd should require
On Tue, Dec 15, 2009 at 06:19:00PM +0300, Peter Volkov wrote:
> So what we will do with this? It'll be great to fix dtd to follow our
> requirements, but there is a problem:
>
> if we change dtd like this:
>
> <!ELEMENT pkgmetadata ( herd+, (maintainer|longdescription|use|upstream)* )>
>
> we will force all metadata.xml files have strict order of tags: first
> <herd/> then other tags. Currently there are about 200 ebuilds with
> different order http://bugs.gentoo.org/show_bug.cgi?id=279206#c4 .
Forced ordering looks fine to me... we could announce the change, let
devs fix their metadata during a short period of time (say, 2 weeks),
then force-fix the ones left (200 is a small number) and apply the dtd fix.
--
Alex Alexander :: wired
Gentoo Developer
www.linuxized.com
12-15-2009, 04:27 PM
Mike Frysinger
metdata.dtd should require
On Tuesday 15 December 2009 10:19:00 Peter Volkov wrote:
> we will force all metadata.xml files have strict order of tags: first
> <herd/> then other tags. Currently there are about 200 ebuilds with
> different order http://bugs.gentoo.org/show_bug.cgi?id=279206#c4 .
>
> Hans's suggestion http://bugs.gentoo.org/show_bug.cgi?id=279206#c7 works
> but it is not nice and prohibits another order: herd, other tags, herd.
>
> Personally I'd kept dtd simple and forced ordering.
+1 ... i dont see any need/reason to have fields scattered
> So actually two questions here:
>
> 1. How should we fix dtd?
> 2. Are there any problems if we fix dtd first and let maintainers fix
> metadata.xml on bumps (iow, what will became broken if metadata.xml
> became not valid)?
write a script to fix the tree and then update the dtd
-mike
12-15-2009, 04:48 PM
Jonathan Callen
metdata.dtd should require
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Peter Volkov wrote:
> В Пнд, 07/12/2009 в 21:20 +0100, Thilo Bangert пишет:
>> Hans de Graaff <graaff@gentoo.org> said:
>>> On Mon, 2009-12-07 at 12:56 +0100, Thilo Bangert wrote:
>>>> dev-util/cucumber <herd> missing
>>> Fixed, but this is really a bug in metadata.dtd, which specifies
>>> <!ELEMENT pkgmetadata ( (herd|maintainer|longdescription|use|
>>> upstream)* )>
>
>> indeed: http://bugs.gentoo.org/show_bug.cgi?id=279206
>
> So what we will do with this? It'll be great to fix dtd to follow our
> requirements, but there is a problem:
>
> if we change dtd like this:
>
> <!ELEMENT pkgmetadata ( herd+, (maintainer|longdescription|use|upstream)* )>
>
> we will force all metadata.xml files have strict order of tags: first
> <herd/> then other tags. Currently there are about 200 ebuilds with
> different order http://bugs.gentoo.org/show_bug.cgi?id=279206#c4 .
>
> Hans's suggestion http://bugs.gentoo.org/show_bug.cgi?id=279206#c7 works
> but it is not nice and prohibits another order: herd, other tags, herd.
>
> Personally I'd kept dtd simple and forced ordering.
>
> So actually two questions here:
>
> 1. How should we fix dtd?
> 2. Are there any problems if we fix dtd first and let maintainers fix
> metadata.xml on bumps (iow, what will became broken if metadata.xml
> became not valid)?
>
we will force all metadata.xml files have strict order of tags: first
<herd/> then other tags. Currently there are about 200 ebuilds with
different order http://bugs.gentoo.org/show_bug.cgi?id=279206#c4 .
Others and I actually make use of the order in metadata.xml. The first
entry is the one that will get bugs assignment in bugzilla, and the
others will get CCed.
So if we're really going with herds first in metadata.xml, could we have
an optional attribute - or whatever else you see fit - to convey that
_this_ herd/maintainer is the "main" herd/maintainer?
Thanks
Rémi
12-23-2009, 05:36 PM
Paul de Vrieze
metdata.dtd should require
On Wed, Dec 16, 2009 at 7:49 AM, Rémi Cardona <remi@gentoo.org> wrote:
> Le 15/12/2009 16:19, Peter Volkov a écrit :
>>
>> we will force all metadata.xml files have strict order of tags: first
>> <herd/> *then other tags. Currently there are about 200 ebuilds with
>> different order http://bugs.gentoo.org/show_bug.cgi?id=279206#c4 .
>
> Others and I actually make use of the order in metadata.xml. The first entry
> is the one that will get bugs assignment in bugzilla, and the others will
> get CCed.
>
> So if we're really going with herds first in metadata.xml, could we have an
> optional attribute - or whatever else you see fit - to convey that _this_
> herd/maintainer is the "main" herd/maintainer?
>
> Thanks
>
Perhaps we should create a schema to validate the file. XMLSchema (or
any of the other standards) allows for much more flexibility in
specifying these things. Btw. I did not design the metadata DTD for
order to be significant. The only priority is that maintainer goes
before herd, that's all.
Paul
--
Paul de Vrieze
12-24-2009, 11:37 AM
Richard Freeman
metdata.dtd should require
On 12/23/2009 01:36 PM, Paul de Vrieze wrote:
Perhaps we should create a schema to validate the file. XMLSchema (or
any of the other standards) allows for much more flexibility in
specifying these things. Btw. I did not design the metadata DTD for
order to be significant. The only priority is that maintainer goes
before herd, that's all.
I think we should definitely have some way of designating which should
be the contact for bugs. I've had some bugs sit around for a while
without being noticed because they were assigned to the herd the package
is in, and not to me personally, and I don't generally work with that
herd, and the project associated with the herd doesn't generally
maintain the package.
I'm sure there are many cases where a similar situation exists.
Another way to handle this is at least CC EVERYBODY in the metadata in
new bugs, and not assume that copying the project will get all the
maintainers.
12-24-2009, 03:12 PM
Jeroen Roovers
metdata.dtd should require
On Thu, 24 Dec 2009 07:37:26 -0500
Richard Freeman <rich0@gentoo.org> wrote:
> I think we should definitely have some way of designating which
> should be the contact for bugs. I've had some bugs sit around for a
> while without being noticed because they were assigned to the herd
> the package is in, and not to me personally, and I don't generally
> work with that herd, and the project associated with the herd doesn't
> generally maintain the package.
That means the bug wasn't properly assigned... I've set up some rules
[1] that might help there. It comes down to this:
1) If one or more <maintainer> tags are listed, assign to the first one
mentioned.
2) If no <maintainer> tag is listed, assign to the first <herd>
mentioned.
3) CC everyone else listed (<maintainer>s and <herd>s).
The rationale is that if one or more <maintainer>s are listed as well as
one or more <herd>s, a bug wrangler shouldn't need to go find out if
that maintainer happens to be on the <herd>'s alias as well (otherwise,
why would it mention those maintainers separately?).
The order in which <maintainer> and <herd> tags are listed isn't
special, because <maintainer> goes before <herd> for the reason set out
above. The order of multiple <maintainer> tags (or multiple <herd> tags)
does matter: the first one you find is the Assignee, the others are
CC'd.
metdata.dtd should require 
