GNU bug report logs - #10838
24.0.92; package.el shows a misleading version number

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 10838 in the body.
You can then email your comments to 10838 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Phil Hagelberg <phil <at> hagelb.org>
To: bug-gnu-emacs <at> gnu.org
Subject: 24.0.92; package.el shows a misleading version number
Date: Fri, 17 Feb 2012 18:40:47 +0000

I'm passing on this bug report from the repository where I was
maintaining package.el in preparation for merging it into Emacs itself:

https://github.com/technomancy/package.el/issues/5

> There is a version of package.el which is available from
> tromey.com/elpa, which is clearly the starting point for this
> package.el, which claims to be version 0.9 in the head of the file. The
> current (and emacs23) versions contain this same version number, despite
> having significantly different features. This is misleading and
> remarkably confusing. The version numbers in both the emacs24 and
> emacs23 versions should be updated to indicate that they contain new
> features.

I agree that incrementing the version number makes sense as a lot of
features have been added since the 0.9 release.

Message #8 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Monnier <monnier <at> iro.umontreal.ca>
To: Phil Hagelberg <phil <at> hagelb.org>
Cc: 10838 <at> debbugs.gnu.org
Subject: Re: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Fri, 17 Feb 2012 18:46:19 -0500

> I agree that incrementing the version number makes sense as a lot of
> features have been added since the 0.9 release.

IIUC the package.el we distribute is not synchronized with the unbundled
package.el any more, right?  So it doesn't need a version number any
more (it's version number is the one of Emacs).


        Stefan

Message #11 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Phil Hagelberg <phil <at> hagelb.org>
To: Stefan Monnier <monnier <at> iro.umontreal.ca>
Cc: 10838 <at> debbugs.gnu.org
Subject: Re: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Sat, 18 Feb 2012 00:01:36 +0000

Stefan Monnier <monnier <at> iro.umontreal.ca> writes:

> IIUC the package.el we distribute is not synchronized with the unbundled
> package.el any more, right?  So it doesn't need a version number any
> more (it's version number is the one of Emacs).

If we want to backport package.el to work in Emacs 23 then we should
probably keep the version number, but if we decide that's not worth it
then it could be dropped. I haven't looked into what's keeping it from
working on 23; I've just been pointing 23 users at an older version.
Would you like me to investigate to see how much it would take to make
it work?

-Phil

Message #14 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Monnier <monnier <at> iro.umontreal.ca>
To: Phil Hagelberg <phil <at> hagelb.org>
Cc: 10838 <at> debbugs.gnu.org
Subject: Re: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Sat, 18 Feb 2012 17:40:40 -0500

>> IIUC the package.el we distribute is not synchronized with the unbundled
>> package.el any more, right?  So it doesn't need a version number any
>> more (it's version number is the one of Emacs).
> If we want to backport package.el to work in Emacs 23 then we should
> probably keep the version number,

I don't think so: if the canonical package.el is the one that comes with
Emacs, then it's version number is Emacs's (i.e. there "package.el from
Emacs-24.1" and "package.el from Emacs-24.2, ...").

> but if we decide that's not worth it then it could be
> dropped. I haven't looked into what's keeping it from working on 23;
> I've just been pointing 23 users at an older version.  Would you like
> me to investigate to see how much it would take to make it work?

I haven't thought about this.  It'd be good to have a package.el that
works with Emacs-23 to access GNU ELPA, but it doesn't have to be sync'd
with Emacs-24's.  So unless package.el version 0.9 has problems
accessing GNU ELPA, I see no need to backport the current code.


        Stefan

Message #17 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Chong Yidong <cyd <at> gnu.org>
To: Tom Tromey <tromey <at> redhat.com>
Cc: Phil Hagelberg <phil <at> hagelb.org>, Stefan Monnier <monnier <at> iro.umontreal.ca>,
	10838 <at> debbugs.gnu.org
Subject: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Sat, 25 Feb 2012 12:56:48 +0800

> There is a version of package.el which is available from
> tromey.com/elpa, which is clearly the starting point for this
> package.el, which claims to be version 0.9 in the head of the file. The
> current (and emacs23) versions contain this same version number, despite
> having significantly different features. This is misleading and
> remarkably confusing. The version numbers in both the emacs24 and
> emacs23 versions should be updated to indicate that they contain new
> features.

IMO, it's desireable for the package.el in Emacs to retain a version
number header, just in case there is any package that in turn relies on
package.el (even though sounds pretty theoretical).  I propose bumping
the package.el in Emacs to 1.0.  Any objection?

Message #20 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Phil Hagelberg <phil <at> hagelb.org>
To: Chong Yidong <cyd <at> gnu.org>
Cc: Tom Tromey <tromey <at> redhat.com>, Stefan Monnier <monnier <at> iro.umontreal.ca>,
	10838 <at> debbugs.gnu.org
Subject: Re: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Fri, 24 Feb 2012 21:23:40 -0800

On Fri, Feb 24, 2012 at 8:56 PM, Chong Yidong <cyd <at> gnu.org> wrote:
> IMO, it's desireable for the package.el in Emacs to retain a version
> number header, just in case there is any package that in turn relies on
> package.el (even though sounds pretty theoretical).  I propose bumping
> the package.el in Emacs to 1.0.  Any objection?

1.0 makes sense considering it will be the first version distributed with Emacs.

-Phil

Message #23 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Phil Hagelberg <phil <at> hagelb.org>
To: Stefan Monnier <monnier <at> iro.umontreal.ca>
Cc: 10838 <at> debbugs.gnu.org
Subject: Re: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Sat, 25 Feb 2012 05:23:09 +0000

Stefan Monnier <monnier <at> iro.umontreal.ca> writes:

>> but if we decide that's not worth it then it could be
>> dropped. I haven't looked into what's keeping it from working on 23;
>> I've just been pointing 23 users at an older version.  Would you like
>> me to investigate to see how much it would take to make it work?
>
> I haven't thought about this.  It'd be good to have a package.el that
> works with Emacs-23 to access GNU ELPA, but it doesn't have to be sync'd
> with Emacs-24's.  So unless package.el version 0.9 has problems
> accessing GNU ELPA, I see no need to backport the current code.

The problem is there are many versions claiming to be 0.9. The original
one on tromey.com (which did not support multiple sources) probably has
the best claim of being "the real 0.9". I think Chong's suggestion of
releasing 24 with package.el 1.0 makes sense, and if we decide to
backport a few changes for official 23-compatibility it could be 0.9.1
or something.

-Phil

Message #26 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Monnier <monnier <at> iro.umontreal.ca>
To: Phil Hagelberg <phil <at> hagelb.org>
Cc: 10838 <at> debbugs.gnu.org
Subject: Re: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Sat, 25 Feb 2012 04:46:48 -0500

> The problem is there are many versions claiming to be 0.9. The original
> one on tromey.com (which did not support multiple sources) probably has
> the best claim of being "the real 0.9". I think Chong's suggestion of
> releasing 24 with package.el 1.0 makes sense, and if we decide to
> backport a few changes for official 23-compatibility it could be 0.9.1
> or something.

From what I understand, our package.el is the authoritative package.el,
so it can just use Emacs's own version number: no need to name it
"version 1.0".
As for the one that works with Emacs-23: it's just a stop-gap, so it
really doesn't matter what it's called and whether or not it's based on
our code: as long as it exists and works well enough, we're set.


        Stefan

Message #29 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Phil Hagelberg <phil <at> hagelb.org>
To: Stefan Monnier <monnier <at> iro.umontreal.ca>
Cc: 10838 <at> debbugs.gnu.org
Subject: Re: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Sat, 25 Feb 2012 22:21:31 +0000

Stefan Monnier <monnier <at> iro.umontreal.ca> writes:

> From what I understand, our package.el is the authoritative package.el,
> so it can just use Emacs's own version number: no need to name it
> "version 1.0".

Chong had a good point that you could have another package that depends
on a particular feature in package.el; in that case having a version
number to pin it on so you get a clearer error might be advantageous.

-Phil

Message #32 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Tom Tromey <tromey <at> redhat.com>
To: Chong Yidong <cyd <at> gnu.org>
Cc: Phil Hagelberg <phil <at> hagelb.org>, Stefan Monnier <monnier <at> iro.umontreal.ca>,
	10838 <at> debbugs.gnu.org
Subject: Re: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Thu, 01 Mar 2012 09:49:59 -0700

>>>>> "Chong" == Chong Yidong <cyd <at> gnu.org> writes:

>> There is a version of package.el which is available from
>> tromey.com/elpa, which is clearly the starting point for this
>> package.el, which claims to be version 0.9 in the head of the file. The
>> current (and emacs23) versions contain this same version number, despite
>> having significantly different features. This is misleading and
>> remarkably confusing. The version numbers in both the emacs24 and
>> emacs23 versions should be updated to indicate that they contain new
>> features.

Chong> IMO, it's desireable for the package.el in Emacs to retain a version
Chong> number header, just in case there is any package that in turn relies on
Chong> package.el (even though sounds pretty theoretical).  I propose bumping
Chong> the package.el in Emacs to 1.0.  Any objection?

It is fine by me.

FWIW, package.el has a version number so it can be updated using itself.

Tom

Message #35 received at 10838 <at> debbugs.gnu.org (full text, mbox):

From: Chong Yidong <cyd <at> gnu.org>
To: Tom Tromey <tromey <at> redhat.com>
Cc: Phil Hagelberg <phil <at> hagelb.org>, Stefan Monnier <monnier <at> iro.umontreal.ca>,
	10838 <at> debbugs.gnu.org
Subject: Re: bug#10838: 24.0.92; package.el shows a misleading version number
Date: Sun, 04 Mar 2012 17:45:25 +0800

Tom Tromey <tromey <at> redhat.com> writes:

> Chong> IMO, it's desireable for the package.el in Emacs to retain a version
> Chong> number header, just in case there is any package that in turn relies on
> Chong> package.el (even though sounds pretty theoretical).  I propose bumping
> Chong> the package.el in Emacs to 1.0.  Any objection?
>
> It is fine by me.

Done.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.