FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Gentoo > Gentoo Development

 
 
LinkBack Thread Tools
 
Old 06-26-2011, 07:02 AM
Ciaran McCreesh
 
Default Are tags just sets?

Here's a completely different way of doing tags:

First, standardise sets. We probably want to go with a format along the
lines of:

eapi = 4
description = Monkeys

dev-monkey/howler
dev-monkey/spider
>=dev-monkey/spanky-2.0
dev-monkey/squirrel

where eapi has to be on the first line.

Second, make a bunch of sets named kde-tag, editors-tag, xml-tag,
monkeys-tag etc.

Third, make tools that allow browsing, searching etc able to list
things by tag (or sets in general).

Advantages: dead easy to implement, backwards compatible, we need sets
anyway.

Disadvantages: doesn't use some horribly convoluted system of XML,
wikis and web 2.0.

--
Ciaran McCreesh
 
Old 06-26-2011, 08:41 AM
Michał Górny
 
Default Are tags just sets?

On Sun, 26 Jun 2011 08:02:57 +0100
Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:

> First, standardise sets. We probably want to go with a format along
> the lines of:
>
> eapi = 4
> description = Monkeys

A 'type' field would be useful as well, to support various kinds of
package sets (much like portage handles currently).

> dev-monkey/howler
> dev-monkey/spider
> >=dev-monkey/spanky-2.0
> dev-monkey/squirrel

We'd either want to add || ( ) here, or somehow explicitly specify that
this is a one-of set.

> Second, make a bunch of sets named kde-tag, editors-tag, xml-tag,
> monkeys-tag etc.

I'd go with prefix, not suffix. Should be easier group them together
then.

> Disadvantages: doesn't use some horribly convoluted system of XML,
> wikis and web 2.0.

And introduces another dedicated file format the PM has to implement
from scratch.

--
Best regards,
Michał Górny
 
Old 06-26-2011, 08:43 AM
Ciaran McCreesh
 
Default Are tags just sets?

On Sun, 26 Jun 2011 10:41:24 +0200
Michał Górny <mgorny@gentoo.org> wrote:
> A 'type' field would be useful as well, to support various kinds of
> package sets (much like portage handles currently).

I'm highly doubtful that there's any real need for different kinds of
repository-provided sets. We especially don't want sets to be code...

> > dev-monkey/howler
> > dev-monkey/spider
> > >=dev-monkey/spanky-2.0
> > dev-monkey/squirrel
>
> We'd either want to add || ( ) here, or somehow explicitly specify
> that this is a one-of set.

No, that's something that's determined by how the set's used, not by
what's in the set. There's no such thing as a "one-of" set; a set is
just a list of package dep specs.

> > Disadvantages: doesn't use some horribly convoluted system of XML,
> > wikis and web 2.0.
>
> And introduces another dedicated file format the PM has to implement
> from scratch.

It's a simple text file. It takes fewer lines of code to parse it
from scratch than it does to get the results out of an XML parser.

--
Ciaran McCreesh
 
Old 06-26-2011, 08:54 AM
Michał Górny
 
Default Are tags just sets?

On Sun, 26 Jun 2011 09:43:41 +0100
Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:

> On Sun, 26 Jun 2011 10:41:24 +0200
> Michał Górny <mgorny@gentoo.org> wrote:
> > A 'type' field would be useful as well, to support various kinds of
> > package sets (much like portage handles currently).
>
> I'm highly doubtful that there's any real need for different kinds of
> repository-provided sets. We especially don't want sets to be code...

Simple things like getting a list of packages which own a particular
file (for rebuilds) or grepping a variable are useful to users.
For example, the x11 overlay provides a set to rebuild the xorg server
modules after an update.

> > We'd either want to add || ( ) here, or somehow explicitly specify
> > that this is a one-of set.
>
> No, that's something that's determined by how the set's used, not by
> what's in the set. There's no such thing as a "one-of" set; a set is
> just a list of package dep specs.

Hm, true. I guess noone will want to merge 'a random package matching
a tag' :P.

--
Best regards,
Michał Górny
 
Old 06-26-2011, 09:00 AM
Ciaran McCreesh
 
Default Are tags just sets?

On Sun, 26 Jun 2011 10:54:44 +0200
Michał Górny <mgorny@gentoo.org> wrote:
> > I'm highly doubtful that there's any real need for different kinds
> > of repository-provided sets. We especially don't want sets to be
> > code...
>
> Simple things like getting a list of packages which own a particular
> file (for rebuilds) or grepping a variable are useful to users.

That's something done by sets as provided by the package mangler, not
something done by repository-specified sets.

> For example, the x11 overlay provides a set to rebuild the xorg server
> modules after an update.

That's just a list of package specs. The user then says "only the ones
of these that I have installed" by the way the set is used.

--
Ciaran McCreesh
 
Old 06-26-2011, 11:33 AM
Wyatt Epp
 
Default Are tags just sets?

On Sun, Jun 26, 2011 at 03:02, Ciaran McCreesh
<ciaran.mccreesh@googlemail.com> wrote:
> Here's a completely different way of doing tags:
>
You know, that's not a bad way of going about it. Truth be told, I
had sort of forgotten sets exists because they're a bit cumbersome at
the moment. But it's cheap and dead simple and gets us our 90%
immediately. Actually, it gets 100%, even, if you can include a set
as part of another set (implication) and symlinks function as aliases.
Very clever; I like it.

> where eapi has to be on the first line.
>
Looks fine but just to be clear, why is having the eapi necessary?

> Second, make a bunch of sets named kde-tag, editors-tag, xml-tag,
> monkeys-tag etc.
>
Don't even need the "-tag" part, really. But yes, a couple hundred
sets are in order. And some tool-glue.

> Disadvantages: doesn't use some horribly convoluted system of XML,
> wikis and web 2.0.
>
That's not a disadvantage at all. Thank you for noticing the third path.

While I still don't really believe categories to be necessary, this
will be a fine intermediate step.

Cheers,
Wyatt
 
Old 06-26-2011, 11:33 AM
Wyatt Epp
 
Default Are tags just sets?

On Sun, Jun 26, 2011 at 03:02, Ciaran McCreesh
<ciaran.mccreesh@googlemail.com> wrote:
> Here's a completely different way of doing tags:
>
You know, that's not a bad way of going about it. Truth be told, I
had sort of forgotten sets exists because they're a bit cumbersome at
the moment. But it's cheap and dead simple and gets us our 90%
immediately. Actually, it gets 100%, even, if you can include a set
as part of another set (implication) and symlinks function as aliases.
Very clever; I like it.

> where eapi has to be on the first line.
>
Looks fine but just to be clear, why is having the eapi necessary?

> Second, make a bunch of sets named kde-tag, editors-tag, xml-tag,
> monkeys-tag etc.
>
Don't even need the "-tag" part, really. But yes, a couple hundred
sets are in order. And some tool-glue.

> Disadvantages: doesn't use some horribly convoluted system of XML,
> wikis and web 2.0.
>
That's not a disadvantage at all. Thank you for noticing the third path.

While I still don't really believe categories to be necessary, this
will be a fine intermediate step.

Cheers,
Wyatt
 
Old 06-26-2011, 12:17 PM
Kent Fredric
 
Default Are tags just sets?

On 26 June 2011 19:02, Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:
> Here's a completely different way of doing tags:
>
> First, standardise sets. We probably want to go with a format along the
> lines of:
>
> * *eapi = 4
> * *description = Monkeys
>
> * *dev-monkey/howler
> * *dev-monkey/spider
> * *>=dev-monkey/spanky-2.0
> * *dev-monkey/squirrel
>
> where eapi has to be on the first line.
>

I initially didn't like the idea somewhat, but then I figured the
amount of impact and retooling required to make this work is virtually
zero, its not complicated, and its text based.

So why don't we just implement it, even if it sucks balls, the amount
of downsides it has are zero really, it doesn't affect how portage
currently works at all, so if we prove it to suck or decide it needs
replacing, we can throw it out and put something else in with very
little pain.

So +1.

( Yes, I understand the concerns of Yet Another format, I myself would
suggest JSON for a plethora of reasons were it up to me, and all
though it is /mostly/ just a list of package specs, those "first
lines" with the = in them make this more "format" than just a text
file, but I think we should see whether or not the concept works FIRST
before debating whether or not we've bikeshedded the right format to
put it in ).


--
Kent

perl -e* "print substr( "edrgmaM* SPA NOcomil.ic@tfrken", $_ * 3,
3 ) for ( 9,8,0,7,1,6,5,4,3,2 );"

http://kent-fredric.fox.geek.nz
 
Old 06-26-2011, 12:17 PM
Kent Fredric
 
Default Are tags just sets?

On 26 June 2011 19:02, Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:
> Here's a completely different way of doing tags:
>
> First, standardise sets. We probably want to go with a format along the
> lines of:
>
> * *eapi = 4
> * *description = Monkeys
>
> * *dev-monkey/howler
> * *dev-monkey/spider
> * *>=dev-monkey/spanky-2.0
> * *dev-monkey/squirrel
>
> where eapi has to be on the first line.
>

I initially didn't like the idea somewhat, but then I figured the
amount of impact and retooling required to make this work is virtually
zero, its not complicated, and its text based.

So why don't we just implement it, even if it sucks balls, the amount
of downsides it has are zero really, it doesn't affect how portage
currently works at all, so if we prove it to suck or decide it needs
replacing, we can throw it out and put something else in with very
little pain.

So +1.

( Yes, I understand the concerns of Yet Another format, I myself would
suggest JSON for a plethora of reasons were it up to me, and all
though it is /mostly/ just a list of package specs, those "first
lines" with the = in them make this more "format" than just a text
file, but I think we should see whether or not the concept works FIRST
before debating whether or not we've bikeshedded the right format to
put it in ).


--
Kent

perl -e* "print substr( "edrgmaM* SPA NOcomil.ic@tfrken", $_ * 3,
3 ) for ( 9,8,0,7,1,6,5,4,3,2 );"

http://kent-fredric.fox.geek.nz
 
Old 06-26-2011, 12:48 PM
Michał Górny
 
Default Are tags just sets?

On Sun, 26 Jun 2011 10:00:43 +0100
Ciaran McCreesh <ciaran.mccreesh@googlemail.com> wrote:

> On Sun, 26 Jun 2011 10:54:44 +0200
> Michał Górny <mgorny@gentoo.org> wrote:
> > > I'm highly doubtful that there's any real need for different kinds
> > > of repository-provided sets. We especially don't want sets to be
> > > code...
> >
> > Simple things like getting a list of packages which own a particular
> > file (for rebuilds) or grepping a variable are useful to users.
>
> That's something done by sets as provided by the package mangler, not
> something done by repository-specified sets.

So we should provide separate copies of the same sets for each package
mangler?

> > For example, the x11 overlay provides a set to rebuild the xorg
> > server modules after an update.
>
> That's just a list of package specs. The user then says "only the ones
> of these that I have installed" by the way the set is used.

Well, I think a simple specification saying 'all installed packages
which install to /usr/lib/foo' is much simpler to write and maintain
than a random number of package names.

--
Best regards,
Michał Górny
 

Thread Tools




All times are GMT. The time now is 09:30 PM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org