rst-mode deprecated keybindings: okay to remove? During freeze?

rst-mode deprecated keybindings: okay to remove? During freeze?

[Daniel Colascione]

Would anyone mind terribly if I removed rst-mode's support for its 
"deprecated" keybindings? If not, can I do it before the next release? 
These bindings (e.g., C-c 3) have emitted warnings for years, and right 
now, they show up as ugly "??" entries in describe-mode output. I have a 
patch that makes the presentation nicer, but it'd be better to just kill 
these bindings entirely.

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[Stefan Monnier]

> Would anyone mind terribly if I removed rst-mode's support for its
> "deprecated" keybindings? If not, can I do it before the next release? These
> bindings (e.g., C-c 3) have emitted warnings for years, and right now, they
> show up as ugly "??" entries in describe-mode output. I have a patch that
> makes the presentation nicer, but it'd be better to just kill these
> bindings entirely.

Just adding rst-mode's authors in Cc.


        Stefan

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[Stefan Merten]

[-- Attachment #1: Type: text/plain, Size: 679 bytes --]

Hi Daniel!

Today Stefan Monnier wrote:
>> Would anyone mind terribly if I removed rst-mode's support for its
>> "deprecated" keybindings? If not, can I do it before the next release?

Please keep them once more. But you are right: They should be removed
at some point.

>> These
>> bindings (e.g., C-c 3) have emitted warnings for years, and right now, they
>> show up as ugly "??" entries in describe-mode output. I have a patch that
>> makes the presentation nicer, but it'd be better to just kill these
>> bindings entirely.

Thanks for the patch. But I think it's better to remove them in the
next release.

I'll do this.


						Grüße

						Stefan

[-- Attachment #2: Type: application/pgp-signature, Size: 307 bytes --]

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[Daniel Colascione]

Thanks for the reply.

On 01/06/2014 12:35 PM, Stefan Merten wrote:
> Hi Daniel!
>
> Today Stefan Monnier wrote:
>>> Would anyone mind terribly if I removed rst-mode's support for its
>>> "deprecated" keybindings? If not, can I do it before the next release?
>
> Please keep them once more. But you are right: They should be removed
> at some point.
>
>>> These
>>> bindings (e.g., C-c 3) have emitted warnings for years, and right now, they
>>> show up as ugly "??" entries in describe-mode output. I have a patch that
>>> makes the presentation nicer, but it'd be better to just kill these
>>> bindings entirely.
>
> Thanks for the patch. But I think it's better to remove them in the
> next release.
>
> I'll do this.

Wait, I'm confused --- are you removing the deprecated bindings for 24.4 
or not? If we're keeping them for this release, we should at least apply 
my patch so that C-h m is less confusing for users.

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[Stefan Merten]

[-- Attachment #1: Type: text/plain, Size: 1367 bytes --]

Hi Daniel!

2 days ago Daniel Colascione wrote:
> On 01/06/2014 12:35 PM, Stefan Merten wrote:
>> Today Stefan Monnier wrote:
>>>> Would anyone mind terribly if I removed rst-mode's support for its
>>>> "deprecated" keybindings? If not, can I do it before the next release?
>> 
>> Please keep them once more. But you are right: They should be removed
>> at some point.
>> 
>>>> These
>>>> bindings (e.g., C-c 3) have emitted warnings for years, and right now, they
>>>> show up as ugly "??" entries in describe-mode output. I have a patch that
>>>> makes the presentation nicer, but it'd be better to just kill these
>>>> bindings entirely.
>> 
>> Thanks for the patch. But I think it's better to remove them in the
>> next release.
>> 
>> I'll do this.
> 
> Wait, I'm confused --- are you removing the deprecated bindings for
> 24.4 or not?

Sorry for being unclear: I meant in the next release (> 24.4) - not
now. I'm too busy with other things at the moment. At the same time
there are bigger changes in the whole code which are still brewing
though.

> If we're keeping them for this release, we should at
> least apply my patch so that C-h m is less confusing for users.

That may be a good move - though I doubt that anyone has a problem
with that. Please do the patch. I'll look at it later.


						Grüße

						Stefan

[-- Attachment #2: Type: application/pgp-signature, Size: 307 bytes --]

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[Stefan Merten]

[-- Attachment #1: Type: text/plain, Size: 1920 bytes --]

Hi Daniel!

2 days ago Stefan Merten wrote:
> 2 days ago Daniel Colascione wrote:
>> On 01/06/2014 12:35 PM, Stefan Merten wrote:
>>> Today Stefan Monnier wrote:
>>>>> Would anyone mind terribly if I removed rst-mode's support for its
>>>>> "deprecated" keybindings? If not, can I do it before the next release?
>>> 
>>> Please keep them once more. But you are right: They should be removed
>>> at some point.
>>> 
>>>>> These
>>>>> bindings (e.g., C-c 3) have emitted warnings for years, and right now, they
>>>>> show up as ugly "??" entries in describe-mode output. I have a patch that
>>>>> makes the presentation nicer, but it'd be better to just kill these
>>>>> bindings entirely.
>>> 
>>> Thanks for the patch. But I think it's better to remove them in the
>>> next release.
>>> 
>>> I'll do this.
>> 
>> Wait, I'm confused --- are you removing the deprecated bindings for
>> 24.4 or not?
> 
> Sorry for being unclear: I meant in the next release (> 24.4) - not
> now. I'm too busy with other things at the moment. At the same time
> there are bigger changes in the whole code which are still brewing
> though.
> 
>> If we're keeping them for this release, we should at
>> least apply my patch so that C-h m is less confusing for users.
> 
> That may be a good move - though I doubt that anyone has a problem
> with that. Please do the patch. I'll look at it later.

I now checked your patch. In general I like your patch.

However, at this time I refuse using `lexical-let' in `rst.el' for
backward compatibility. If your patch works without `lexical-let' then
go ahead. Otherwise please do not commit the patch to the Emacs main
line.

PS: The Emacs development seems to be moving to GIT. For me this would
    be yet another VCS I need to learn (sigh). When this happens I
    might be unable to use the Emacs VCS for some time.


						Grüße

						Stefan

[-- Attachment #2: Type: application/pgp-signature, Size: 307 bytes --]

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[David Kastrup]

Stefan Merten <stefan@merten-home.de> writes:

> I now checked your patch. In general I like your patch.
>
> However, at this time I refuse using `lexical-let' in `rst.el' for
> backward compatibility.

Backward compatibility with _what_?  lexical-let was available in
Emacs 21.1 already.  It's conceivable that this requires something like

(eval-when-compile (require 'cl-macs))

for some non-current versions, but it's certainly available.

What isn't, is setting the lexical-binding file variable to t in order
to have all "let" interpreted as "lexical-let" by default.

-- 
David Kastrup

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[Stefan Monnier]

> (eval-when-compile (require 'cl-macs))

Please never do that.  Use (eval-when-compile (require 'cl)) instead.


        Stefan

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[Stefan Merten]

[-- Attachment #1: Type: text/plain, Size: 746 bytes --]

Hi David, Daniel, Stefan, all!

1 hours ago David Kastrup wrote:
> Backward compatibility with _what_?  lexical-let was available in
> Emacs 21.1 already.

Ah, ok. That was in the code which were considered doomed at this
time. Yes, I know, macros were allowed, ...

In fact I'm using it already ;-) .

> What isn't, is setting the lexical-binding file variable to t in order
> to have all "let" interpreted as "lexical-let" by default.

Thanks for clarifying this. Indeed it was this aspect I probably had
in mind.

12 minutes ago Stefan Monnier wrote:
> Please never do that.  Use (eval-when-compile (require 'cl)) instead.

Indeed that is already in the code.

@Daniel: Please go ahead!


						Grüße

						Stefan

[-- Attachment #2: Type: application/pgp-signature, Size: 307 bytes --]

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[Thierry Volpiatto]

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

>> (eval-when-compile (require 'cl-macs))
>
> Please never do that.  Use (eval-when-compile (require 'cl)) instead.

Why? Why inside Emacs-24.4 people would use this ?
We had a discussion about this sometime ago:
Why people are not using "cl-lib+lexical-binding" especially in packages
that are embeded in Emacs-24.4 ?
How do you want people adopt cl-lib and lex if in Emacs itself people
continue to use (require 'cl) ?

-- 
Thierry
Get my Gnupg key:
gpg --keyserver pgp.mit.edu --recv-keys 59F29997 

Re: rst-mode deprecated keybindings: okay to remove? During freeze?

[Stefan Monnier]

>>> (eval-when-compile (require 'cl-macs))
>> Please never do that.  Use (eval-when-compile (require 'cl)) instead.
> Why? Why inside Emacs-24.4 people would use this ?

Then they should use (eval-when-compile (require 'cl-lib))

The point is that the division of cl-lib (and cl) into cl.el,
cl-macs.el, cl-seq.el, etc... is an internal detail that can change.


        Stefan