Maintenance suggestion on an emacs mode

Maintenance suggestion on an emacs mode

[paul r]

Hello,

You are contacted because you are mentioned as the maintainer of a
mode now included in GNU Emacs, development version 23 as of today, on
which we suggest you to do some small modifications.
Those modifications concern some default key bindings on your mode
map, bound on the C-c[a-zA-Z] pattern, which is supposed to be
reserved for users, as explained in the last paragraph of the link
[1].
Each of you will receive an other email with the file to fix and the
matching lines. If possible, please send back a patch on emacs-devel
with those keybindings changed.
Thank you

-- Paul

[1] http://www.gnu.org/software/emacs/manual/html_node/emacs/Keymaps.html

Re: Maintenance suggestion on an emacs mode

[paul r]

Please find below the packages for which maintainer is FSF or
maintainer email does not work anymore.

Regards

-- Paul

FSF
lisp/progmodes/delphi.el:1896: '("\C-cu" delphi-find-unit)

FSF
lisp/textmodes/sgml-mode.el:1583: (define-key map "\C-co" 'html-ordered-list)
lisp/textmodes/sgml-mode.el:1584: (define-key map "\C-cu" 'html-unordered-list)
lisp/textmodes/sgml-mode.el:1585: (define-key map "\C-cr" 'html-radio-buttons)
lisp/textmodes/sgml-mode.el:1586: (define-key map "\C-cc" 'html-checkboxes)
lisp/textmodes/sgml-mode.el:1587: (define-key map "\C-cl" 'html-list-item)
lisp/textmodes/sgml-mode.el:1588: (define-key map "\C-ch" 'html-href-anchor)
lisp/textmodes/sgml-mode.el:1589: (define-key map "\C-cn" 'html-name-anchor)
lisp/textmodes/sgml-mode.el:1590: (define-key map "\C-ci" 'html-image))

Michael Schmidt <michael@pbinfo.UUCP> , email adress does not work
lisp/progmodes/modula2.el:78: (define-key map "\C-cb" 'm2-begin)
lisp/progmodes/modula2.el:79: (define-key map "\C-cc" 'm2-case)
lisp/progmodes/modula2.el:80: (define-key map "\C-cd" 'm2-definition)
lisp/progmodes/modula2.el:81: (define-key map "\C-ce" 'm2-else)
lisp/progmodes/modula2.el:82: (define-key map "\C-cf" 'm2-for)
lisp/progmodes/modula2.el:83: (define-key map "\C-ch" 'm2-header)
lisp/progmodes/modula2.el:84: (define-key map "\C-ci" 'm2-if)
lisp/progmodes/modula2.el:85: (define-key map "\C-cm" 'm2-module)
lisp/progmodes/modula2.el:86: (define-key map "\C-cl" 'm2-loop)
lisp/progmodes/modula2.el:87: (define-key map "\C-co" 'm2-or)
lisp/progmodes/modula2.el:88: (define-key map "\C-cp" 'm2-procedure)
lisp/progmodes/modula2.el:90: (define-key map "\C-cr" 'm2-record)
lisp/progmodes/modula2.el:91: (define-key map "\C-cs" 'm2-stdio)
lisp/progmodes/modula2.el:92: (define-key map "\C-ct" 'm2-type)
lisp/progmodes/modula2.el:93: (define-key map "\C-cu" 'm2-until)
lisp/progmodes/modula2.el:94: (define-key map "\C-cv" 'm2-var)
lisp/progmodes/modula2.el:95: (define-key map "\C-cw" 'm2-while)
lisp/progmodes/modula2.el:96: (define-key map "\C-cx" 'm2-export)
lisp/progmodes/modula2.el:97: (define-key map "\C-cy" 'm2-import)

Re: Maintenance suggestion on an emacs mode

[Richard Stallman]

I am pretty sure some of those people are long gone.
If some of them don't respond, we should change the files
to indicate those people are no longer maintaining them.

Re: Maintenance suggestion on an emacs mode

[Richard Stallman]

    Please find below the packages for which maintainer is FSF or
    maintainer email does not work anymore.

Thanks for reporting these.  They should indeed be fixed.
Would you like to write patches for them?

User reserved keybindings in `gnus-diary.el' (was: Maintenance suggestion on an emacs mode)

[Reiner Steib]

On Sat, Mar 29 2008, paul r wrote:

> Those modifications concern some default key bindings on your mode
> map, bound on the C-c[a-zA-Z] pattern, which is supposed to be
> reserved for users, as explained in the last paragraph of the link
> [1].
> Each of you will receive an other email with the file to fix and the
> matching lines. If possible, please send back a patch on emacs-devel
> with those keybindings changed.

`gnus-diary.el' uses `C-c D' as a prefix key:

(define-key message-mode-map "\C-cDc" 'gnus-diary-check-message)
(define-key message-mode-map "\C-cDv" 'gnus-diary-version)
(define-key gnus-article-edit-mode-map "\C-cDc" 'gnus-diary-check-message)
(define-key gnus-article-edit-mode-map "\C-cDv" 'gnus-diary-version)

Do we really need a key binding for `gnus-diary-version'?

I'd suggest to discuss an alternative prefix on the Gnus list.
(Mail-Followup-To set accordingly.)

> [1] http://www.gnu.org/software/emacs/manual/html_node/emacs/Keymaps.html

,----[ (info "(emacs)Keymaps") ]
|    As a user, you can redefine any key; but it is usually best to stick
| to key sequences that consist of `C-c' followed by a letter (upper or
| lower case).  These keys are "reserved for users," so they won't
| conflict with any properly designed Emacs extension.  The function keys
| <F5> through <F9> are also reserved for users.  If you redefine some
| other key, your definition may be overridden by certain extensions or
| major modes which redefine the same key.
`----

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: User reserved keybindings in `gnus-diary.el' (was: Maintenance suggestion on an emacs mode)

[paul r]

2008/3/30, Reiner Steib <reinersteib+gmane@imap.cc>:

>
>  `gnus-diary.el' uses `C-c D' as a prefix key:

Didier Verna is mentionned as the maintener for this file, and has
been informed of theses bindings. Its email is at xemacs.org, so I
guess this file has been imported from XEmacs. I do not know what
policy XEmacs applies for user key bindings, but in case it is
compatible with GNU Emacs policy, it could be done directly by Didier
for both flavours.

Re: User reserved keybindings in `gnus-diary.el' (was: Maintenance suggestion on an emacs mode)

[paul r]

oops, I did not notice you had already included Didier to the thread,
sorry for the noise.

Re: Maintenance suggestion on an emacs mode

[paul r]

2008/3/30, Richard Stallman <rms@gnu.org>:
>     Please find below the packages for which maintainer is FSF or
>     maintainer email does not work anymore.
>
>
> Thanks for reporting these.  They should indeed be fixed.
>  Would you like to write patches for them?
>

Here we go :

This one is fairly evident and can be applied straight.

diff -r 49597064d851 -r f00d6e37763a lisp/progmodes/delphi.el
--- a/lisp/progmodes/delphi.el  Sun Mar 30 16:22:49 2008 +0200
+++ b/lisp/progmodes/delphi.el  Sun Mar 30 16:31:19 2008 +0200
@@ -1890,10 +1890,10 @@ comment block. If not in a // comment, j
          (list '("\r" delphi-newline)
                '("\t" delphi-tab)
                '("\177" backward-delete-char-untabify)
-;;              '("\C-cd" delphi-find-current-def)
-;;              '("\C-cx" delphi-find-current-xdef)
-;;              '("\C-cb" delphi-find-current-body)
-               '("\C-cu" delphi-find-unit)
+;;              '("\C-c\C-d" delphi-find-current-def)
+;;              '("\C-c\C-x" delphi-find-current-xdef)
+;;              '("\C-c\C-b" delphi-find-current-body)
+               '("\C-c\C-f" delphi-find-unit)
                '("\M-q" delphi-fill-comment)
                '("\M-j" delphi-new-comment-line)
                ;; Debug bindings:



That one is less evident. Actually, HTML mode has a 'quick-keys'
option that is disabled by default. When enabled, C-c[a-z] keys
are bound. I do not know what users want, I'm not one of them,
but in case here is the patch. I just prefixed those keys by
"C-c C-q" (q for quick-key).

diff -r f00d6e37763a -r 7492320e7825 lisp/textmodes/sgml-mode.el
--- a/lisp/textmodes/sgml-mode.el       Sun Mar 30 16:31:19 2008 +0200
+++ b/lisp/textmodes/sgml-mode.el       Sun Mar 30 16:48:36 2008 +0200
@@ -1579,15 +1579,15 @@ This takes effect when first loading the
     (define-key map "\C-c\C-cn" 'html-name-anchor)
     (define-key map "\C-c\C-ci" 'html-image)
     (when html-quick-keys
-      (define-key map "\C-c-" 'html-horizontal-rule)
-      (define-key map "\C-co" 'html-ordered-list)
-      (define-key map "\C-cu" 'html-unordered-list)
-      (define-key map "\C-cr" 'html-radio-buttons)
-      (define-key map "\C-cc" 'html-checkboxes)
-      (define-key map "\C-cl" 'html-list-item)
-      (define-key map "\C-ch" 'html-href-anchor)
-      (define-key map "\C-cn" 'html-name-anchor)
-      (define-key map "\C-ci" 'html-image))
+      (define-key map "\C-c\C-q-" 'html-horizontal-rule)
+      (define-key map "\C-c\C-qo" 'html-ordered-list)
+      (define-key map "\C-c\C-qu" 'html-unordered-list)
+      (define-key map "\C-c\C-qr" 'html-radio-buttons)
+      (define-key map "\C-c\C-qc" 'html-checkboxes)
+      (define-key map "\C-c\C-ql" 'html-list-item)
+      (define-key map "\C-c\C-qh" 'html-href-anchor)
+      (define-key map "\C-c\C-qn" 'html-name-anchor)
+      (define-key map "\C-c\C-qi" 'html-image))
     (define-key map "\C-c\C-s" 'html-autoview-mode)
     (define-key map "\C-c\C-v" 'browse-url-of-buffer)
     (define-key map [menu-bar html] (cons "HTML" menu-map))




For this mode, keys are bound by default, so the patch should be
applied. As they all are quick keys, I prefixed them by "C-c C-q"
as well.

diff -r 7492320e7825 -r 2fe60049a07e lisp/progmodes/modula2.el
--- a/lisp/progmodes/modula2.el Sun Mar 30 16:48:36 2008 +0200
+++ b/lisp/progmodes/modula2.el Sun Mar 30 16:58:58 2008 +0200
@@ -75,30 +75,29 @@
 (if m2-mode-map ()
   (let ((map (make-sparse-keymap)))
     (define-key map "\^i" 'm2-tab)
-    (define-key map "\C-cb" 'm2-begin)
-    (define-key map "\C-cc" 'm2-case)
-    (define-key map "\C-cd" 'm2-definition)
-    (define-key map "\C-ce" 'm2-else)
-    (define-key map "\C-cf" 'm2-for)
-    (define-key map "\C-ch" 'm2-header)
-    (define-key map "\C-ci" 'm2-if)
-    (define-key map "\C-cm" 'm2-module)
-    (define-key map "\C-cl" 'm2-loop)
-    (define-key map "\C-co" 'm2-or)
-    (define-key map "\C-cp" 'm2-procedure)
-    (define-key map "\C-c\C-w" 'm2-with)
-    (define-key map "\C-cr" 'm2-record)
-    (define-key map "\C-cs" 'm2-stdio)
-    (define-key map "\C-ct" 'm2-type)
-    (define-key map "\C-cu" 'm2-until)
-    (define-key map "\C-cv" 'm2-var)
-    (define-key map "\C-cw" 'm2-while)
-    (define-key map "\C-cx" 'm2-export)
-    (define-key map "\C-cy" 'm2-import)
+    (define-key map "\C-c\C-qb" 'm2-begin)
+    (define-key map "\C-c\C-qc" 'm2-case)
+    (define-key map "\C-c\C-qd" 'm2-definition)
+    (define-key map "\C-c\C-qe" 'm2-else)
+    (define-key map "\C-c\C-qf" 'm2-for)
+    (define-key map "\C-c\C-qh" 'm2-header)
+    (define-key map "\C-c\C-qi" 'm2-if)
+    (define-key map "\C-c\C-ql" 'm2-loop)
+    (define-key map "\C-c\C-qm" 'm2-module)
+    (define-key map "\C-c\C-qo" 'm2-or)
+    (define-key map "\C-c\C-qp" 'm2-procedure)
+    (define-key map "\C-c\C-qr" 'm2-record)
+    (define-key map "\C-c\C-qs" 'm2-stdio)
+    (define-key map "\C-c\C-qt" 'm2-type)
+    (define-key map "\C-c\C-qu" 'm2-until)
+    (define-key map "\C-c\C-qv" 'm2-var)
+    (define-key map "\C-c\C-qw" 'm2-while)
+    (define-key map "\C-c\C-cW" 'm2-with)
+    (define-key map "\C-c\C-qx" 'm2-export)
+    (define-key map "\C-c\C-qy" 'm2-import)
     (define-key map "\C-c{" 'm2-begin-comment)
     (define-key map "\C-c}" 'm2-end-comment)
     (define-key map "\C-j"  'm2-newline)
-    (define-key map "\C-c\C-z" 'suspend-emacs)
     (define-key map "\C-c\C-v" 'm2-visit)
     (define-key map "\C-c\C-t" 'm2-toggle)
     (define-key map "\C-c\C-l" 'm2-link)

Re: Maintenance suggestion on an emacs mode

[Richard Stallman]

These changes should be legally trivial, so we don't need papers to
install them.  But it would be a good idea for you to sign papers.
I will send you another message about that.

Re: Maintenance suggestion on an emacs mode

[Mathias Dahl]

>  That one is less evident. Actually, HTML mode has a 'quick-keys'
>  option that is disabled by default. When enabled, C-c[a-z] keys
>  are bound. I do not know what users want, I'm not one of them,
>  but in case here is the patch. I just prefixed those keys by
>  "C-c C-q" (q for quick-key).

> ...
>      (when html-quick-keys
> ...

In this case it is up to the user if he wants to set `html-quick-keys'
to true so I think we should leave that alone. The important thing is
that we do not bind any C-c[a-zA-Z] keys by default, right? If the
user wants this convenience feature we should not make it less
convenient.

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

> In this case it is up to the user if he wants to set `html-quick-keys'
> to true so I think we should leave that alone. The important thing is
> that we do not bind any C-c[a-zA-Z] keys by default, right? If the
> user wants this convenience feature we should not make it less
> convenient.

No, that's not how it works.


        Stefan

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

>> Please find below the packages for which maintainer is FSF or
>> maintainer email does not work anymore.
>> 
>> 
>> Thanks for reporting these.  They should indeed be fixed.
>> Would you like to write patches for them?
>> 

> Here we go :

> This one is fairly evident and can be applied straight.

Thanks, that looks good.  Please someone install it.

> That one is less evident. Actually, HTML mode has a 'quick-keys'
> option that is disabled by default. When enabled, C-c[a-z] keys
> are bound. I do not know what users want, I'm not one of them,
> but in case here is the patch. I just prefixed those keys by
> "C-c C-q" (q for quick-key).

> diff -r f00d6e37763a -r 7492320e7825 lisp/textmodes/sgml-mode.el
> --- a/lisp/textmodes/sgml-mode.el       Sun Mar 30 16:31:19 2008 +0200
> +++ b/lisp/textmodes/sgml-mode.el       Sun Mar 30 16:48:36 2008 +0200
> @@ -1579,15 +1579,15 @@ This takes effect when first loading the
>      (define-key map "\C-c\C-cn" 'html-name-anchor)
>      (define-key map "\C-c\C-ci" 'html-image)
>      (when html-quick-keys
> -      (define-key map "\C-c-" 'html-horizontal-rule)
> -      (define-key map "\C-co" 'html-ordered-list)
> -      (define-key map "\C-cu" 'html-unordered-list)
> -      (define-key map "\C-cr" 'html-radio-buttons)
> -      (define-key map "\C-cc" 'html-checkboxes)
> -      (define-key map "\C-cl" 'html-list-item)
> -      (define-key map "\C-ch" 'html-href-anchor)
> -      (define-key map "\C-cn" 'html-name-anchor)
> -      (define-key map "\C-ci" 'html-image))
> +      (define-key map "\C-c\C-q-" 'html-horizontal-rule)
> +      (define-key map "\C-c\C-qo" 'html-ordered-list)
> +      (define-key map "\C-c\C-qu" 'html-unordered-list)
> +      (define-key map "\C-c\C-qr" 'html-radio-buttons)
> +      (define-key map "\C-c\C-qc" 'html-checkboxes)
> +      (define-key map "\C-c\C-ql" 'html-list-item)
> +      (define-key map "\C-c\C-qh" 'html-href-anchor)
> +      (define-key map "\C-c\C-qn" 'html-name-anchor)
> +      (define-key map "\C-c\C-qi" 'html-image))
>      (define-key map "\C-c\C-s" 'html-autoview-mode)
>      (define-key map "\C-c\C-v" 'browse-url-of-buffer)
>      (define-key map [menu-bar html] (cons "HTML" menu-map))

I'd recommend to use a config `html-quick-keys-prefix' that defaults to
"\C-c\C-q" so the users can set it to "\C-c" (or anything else) if they
want to.  I guess a good idea would also be to introduce
html-quick-keys-map and then just do
(define-key map html-quick-keys-prefix html-quick-keys-map).

> For this mode, keys are bound by default, so the patch should be
> applied. As they all are quick keys, I prefixed them by "C-c C-q"
> as well.

> diff -r 7492320e7825 -r 2fe60049a07e lisp/progmodes/modula2.el
> --- a/lisp/progmodes/modula2.el Sun Mar 30 16:48:36 2008 +0200
> +++ b/lisp/progmodes/modula2.el Sun Mar 30 16:58:58 2008 +0200
> @@ -75,30 +75,29 @@
>  (if m2-mode-map ()
>    (let ((map (make-sparse-keymap)))
>      (define-key map "\^i" 'm2-tab)
> -    (define-key map "\C-cb" 'm2-begin)
> -    (define-key map "\C-cc" 'm2-case)
> -    (define-key map "\C-cd" 'm2-definition)
> -    (define-key map "\C-ce" 'm2-else)
> -    (define-key map "\C-cf" 'm2-for)
> -    (define-key map "\C-ch" 'm2-header)
> -    (define-key map "\C-ci" 'm2-if)
> -    (define-key map "\C-cm" 'm2-module)
> -    (define-key map "\C-cl" 'm2-loop)
> -    (define-key map "\C-co" 'm2-or)
> -    (define-key map "\C-cp" 'm2-procedure)
> -    (define-key map "\C-c\C-w" 'm2-with)
> -    (define-key map "\C-cr" 'm2-record)
> -    (define-key map "\C-cs" 'm2-stdio)
> -    (define-key map "\C-ct" 'm2-type)
> -    (define-key map "\C-cu" 'm2-until)
> -    (define-key map "\C-cv" 'm2-var)
> -    (define-key map "\C-cw" 'm2-while)
> -    (define-key map "\C-cx" 'm2-export)
> -    (define-key map "\C-cy" 'm2-import)
> +    (define-key map "\C-c\C-qb" 'm2-begin)
> +    (define-key map "\C-c\C-qc" 'm2-case)
> +    (define-key map "\C-c\C-qd" 'm2-definition)
> +    (define-key map "\C-c\C-qe" 'm2-else)
> +    (define-key map "\C-c\C-qf" 'm2-for)
> +    (define-key map "\C-c\C-qh" 'm2-header)
> +    (define-key map "\C-c\C-qi" 'm2-if)
> +    (define-key map "\C-c\C-ql" 'm2-loop)
> +    (define-key map "\C-c\C-qm" 'm2-module)
> +    (define-key map "\C-c\C-qo" 'm2-or)
> +    (define-key map "\C-c\C-qp" 'm2-procedure)
> +    (define-key map "\C-c\C-qr" 'm2-record)
> +    (define-key map "\C-c\C-qs" 'm2-stdio)
> +    (define-key map "\C-c\C-qt" 'm2-type)
> +    (define-key map "\C-c\C-qu" 'm2-until)
> +    (define-key map "\C-c\C-qv" 'm2-var)
> +    (define-key map "\C-c\C-qw" 'm2-while)
> +    (define-key map "\C-c\C-cW" 'm2-with)
> +    (define-key map "\C-c\C-qx" 'm2-export)
> +    (define-key map "\C-c\C-qy" 'm2-import)
>      (define-key map "\C-c{" 'm2-begin-comment)
>      (define-key map "\C-c}" 'm2-end-comment)
>      (define-key map "\C-j"  'm2-newline)
> -    (define-key map "\C-c\C-z" 'suspend-emacs)
>      (define-key map "\C-c\C-v" 'm2-visit)
>      (define-key map "\C-c\C-t" 'm2-toggle)
>      (define-key map "\C-c\C-l" 'm2-link)

You could do the same here.


        Stefan

Re: Maintenance suggestion on an emacs mode

[Mathias Dahl]

> > In this case it is up to the user if he wants to set `html-quick-keys'
>  > to true so I think we should leave that alone. The important thing is
>  > that we do not bind any C-c[a-zA-Z] keys by default, right? If the
>  > user wants this convenience feature we should not make it less
>  > convenient.
>
>  No, that's not how it works.

Sorry, I don't understand what you mean. What part of what I write
above have I misunderstood?

Here is some code from sgml-mode.el:

(defvar html-quick-keys sgml-quick-keys
  "Use C-c X combinations for quick insertion of frequent tags when non-nil.
This defaults to `sgml-quick-keys'.
This takes effect when first loading the library.")

and later comes:

...
    (define-key map "\C-c\C-cn" 'html-name-anchor)
    (define-key map "\C-c\C-ci" 'html-image)
    (when html-quick-keys
      (define-key map "\C-c-" 'html-horizontal-rule)
      (define-key map "\C-co" 'html-ordered-list)
      (define-key map "\C-cu" 'html-unordered-list)
      (define-key map "\C-cr" 'html-radio-buttons)
      (define-key map "\C-cc" 'html-checkboxes)
      (define-key map "\C-cl" 'html-list-item)
      (define-key map "\C-ch" 'html-href-anchor)
      (define-key map "\C-cn" 'html-name-anchor)
      (define-key map "\C-ci" 'html-image))
    (define-key map "\C-c\C-s" 'html-autoview-mode)
    (define-key map "\C-c\C-v" 'browse-url-of-buffer)
...

It looks to me as the user makes a choice to have these bindings. Is
it then still not allowed to bind keys like this?

Re: Maintenance suggestion on an emacs mode

[paul r]

2008/3/31, Mathias Dahl <mathias.dahl@gmail.com>:

> Sorry, I don't understand what you mean. What part of what I write
>  above have I misunderstood?

Although my voice should not count too much because I'm not a sgml
user, I think the problem is not whether or not it is activated by
default. It is more that a mode should be fully usable, *including
optional keybindings it may provide*, without using the C-c[a-z],
which is in any case reserved to the user. In other words, one can
safely activate anything a mode provide without fearing its personnal
bindings will be overriden.

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

>> Sorry, I don't understand what you mean. What part of what I write
>> above have I misunderstood?

> Although my voice should not count too much because I'm not a sgml
> user, I think the problem is not whether or not it is activated by
> default. It is more that a mode should be fully usable, *including
> optional keybindings it may provide*, without using the C-c[a-z],
> which is in any case reserved to the user. In other words, one can
> safely activate anything a mode provide without fearing its personnal
> bindings will be overriden.

Exactly,


        Stefan

Re: User reserved keybindings in `gnus-diary.el' (was: Maintenance suggestion on an emacs mode)

[Stephen J. Turnbull]

paul r writes:

 > I do not know what policy XEmacs applies for user key bindings,

XEmacs policy is that the bindings (or prefixes, as the user chooses)
C-c <ASCII letter> are reserved to the user.

Other bindings should not collide with other modes (whether
distributed by XEmacs or not), be used sparingly as as scarce
resource, and be compatible with Emacs where possible.

This should usually be compatible with Emacs policy.

Re: Maintenance suggestion on an emacs mode

[Mathias Dahl]

>  > Although my voice should not count too much because I'm not a sgml
>  > user, I think the problem is not whether or not it is activated by
>  > default. It is more that a mode should be fully usable, *including
>  > optional keybindings it may provide*, without using the C-c[a-z],
>  > which is in any case reserved to the user. In other words, one can
>  > safely activate anything a mode provide without fearing its personnal
>  > bindings will be overriden.
>
>  Exactly,

Fair enough.

Re: Maintenance suggestion on an emacs mode

[paul r]

[-- Attachment #1: Type: text/plain, Size: 551 bytes --]

2008/3/31, Stefan Monnier <monnier@iro.umontreal.ca>:

> You could do the same here.

Please find attached patches.
3 questions :
 - why is sgml in textmodes ?
 - files are not indented according to emacs lisp default rules,
should I resend a patch with full reindentation according-to-mode ?
Attached patches do not change indentation.
 - When testing this kind of modifications, I could not find a way to
unload a mode so that I could make a modification, and reload the
mode. I ended up closing and reopening an emacs instance for that :)

-- Paul

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: sgml-keys.patch --]
[-- Type: text/x-patch; name=sgml-keys.patch, Size: 2660 bytes --]

diff -r 49597064d851 lisp/textmodes/sgml-mode.el
--- a/lisp/textmodes/sgml-mode.el	Sun Mar 30 16:22:49 2008 +0200
+++ b/lisp/textmodes/sgml-mode.el	Mon Mar 31 20:53:56 2008 +0200
@@ -1552,10 +1552,15 @@ Currently just returns (EMPTY-TAGS UNCLO
   :type 'hook
   :options '(html-autoview-mode))
 
-(defvar html-quick-keys sgml-quick-keys
-  "Use C-c X combinations for quick insertion of frequent tags when non-nil.
-This defaults to `sgml-quick-keys'.
-This takes effect when first loading the library.")
+(defvar html-use-quick-keys sgml-quick-keys
+  "Use `html-quick-keys-prefix'+letter combinations for quick
+insertion of frequent tags when non-nil. This defaults to
+`sgml-quick-keys'. This takes effect when first loading the
+library.")
+
+(defvar html-quick-keys-prefix "\C-c\C-q"
+  "When `html-use-quick-keys' is set to non-nil, quick keys can be
+  inserted using this prefix.")
 
 (defvar html-mode-map
   (let ((map (make-sparse-keymap))
@@ -1578,16 +1583,6 @@ This takes effect when first loading the
     (define-key map "\C-c\C-ch" 'html-href-anchor)
     (define-key map "\C-c\C-cn" 'html-name-anchor)
     (define-key map "\C-c\C-ci" 'html-image)
-    (when html-quick-keys
-      (define-key map "\C-c-" 'html-horizontal-rule)
-      (define-key map "\C-co" 'html-ordered-list)
-      (define-key map "\C-cu" 'html-unordered-list)
-      (define-key map "\C-cr" 'html-radio-buttons)
-      (define-key map "\C-cc" 'html-checkboxes)
-      (define-key map "\C-cl" 'html-list-item)
-      (define-key map "\C-ch" 'html-href-anchor)
-      (define-key map "\C-cn" 'html-name-anchor)
-      (define-key map "\C-ci" 'html-image))
     (define-key map "\C-c\C-s" 'html-autoview-mode)
     (define-key map "\C-c\C-v" 'browse-url-of-buffer)
     (define-key map [menu-bar html] (cons "HTML" menu-map))
@@ -1615,6 +1610,23 @@ This takes effect when first loading the
     (define-key menu-map "n" '("Name Anchor" . html-name-anchor))
     map)
   "Keymap for commands for use in HTML mode.")
+
+(defvar html-quick-keys-map
+  (let ((map (make-sparse-keymap)))
+    (define-key map "-" 'html-horizontal-rule)
+    (define-key map "o" 'html-ordered-list)
+    (define-key map "u" 'html-unordered-list)
+    (define-key map "r" 'html-radio-buttons)
+    (define-key map "c" 'html-checkboxes)
+    (define-key map "l" 'html-list-item)
+    (define-key map "h" 'html-href-anchor)
+    (define-key map "n" 'html-name-anchor)
+    (define-key map "i" 'html-image)
+    map)
+  "HTML quick keys keymap.")
+
+(when html-use-quick-keys
+  (define-key html-mode-map html-quick-keys-prefix html-quick-keys-map))
 
 (defvar html-face-tag-alist
   '((bold . "b")

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #3: modula2-keys.patch --]
[-- Type: text/x-patch; name=modula2-keys.patch, Size: 3386 bytes --]

diff -r 49597064d851 lisp/progmodes/modula2.el
--- a/lisp/progmodes/modula2.el	Sun Mar 30 16:22:49 2008 +0200
+++ b/lisp/progmodes/modula2.el	Mon Mar 31 20:56:51 2008 +0200
@@ -75,35 +75,51 @@
 (if m2-mode-map ()
   (let ((map (make-sparse-keymap)))
     (define-key map "\^i" 'm2-tab)
-    (define-key map "\C-cb" 'm2-begin)
-    (define-key map "\C-cc" 'm2-case)
-    (define-key map "\C-cd" 'm2-definition)
-    (define-key map "\C-ce" 'm2-else)
-    (define-key map "\C-cf" 'm2-for)
-    (define-key map "\C-ch" 'm2-header)
-    (define-key map "\C-ci" 'm2-if)
-    (define-key map "\C-cm" 'm2-module)
-    (define-key map "\C-cl" 'm2-loop)
-    (define-key map "\C-co" 'm2-or)
-    (define-key map "\C-cp" 'm2-procedure)
-    (define-key map "\C-c\C-w" 'm2-with)
-    (define-key map "\C-cr" 'm2-record)
-    (define-key map "\C-cs" 'm2-stdio)
-    (define-key map "\C-ct" 'm2-type)
-    (define-key map "\C-cu" 'm2-until)
-    (define-key map "\C-cv" 'm2-var)
-    (define-key map "\C-cw" 'm2-while)
-    (define-key map "\C-cx" 'm2-export)
-    (define-key map "\C-cy" 'm2-import)
     (define-key map "\C-c{" 'm2-begin-comment)
     (define-key map "\C-c}" 'm2-end-comment)
     (define-key map "\C-j"  'm2-newline)
-    (define-key map "\C-c\C-z" 'suspend-emacs)
     (define-key map "\C-c\C-v" 'm2-visit)
     (define-key map "\C-c\C-t" 'm2-toggle)
     (define-key map "\C-c\C-l" 'm2-link)
     (define-key map "\C-c\C-c" 'm2-compile)
     (setq m2-mode-map map)))
+
+(defvar m2-use-quick-keys nil
+  "Use `m2-quick-keys-prefix'+letter combinations for quick
+insertion of frequent tags when non-nil. This defaults to nil.
+This takes effect when first loading the library.")
+
+(defvar m2-quick-keys-prefix "\C-c\C-q"
+  "When `m2-use-quick-keys' is set to non-nil, quick keys can be
+  inserted using this prefix.")
+
+(defvar m2-quick-keys-map
+  (let ((map (make-sparse-keymap)))
+    (define-key map "b" 'm2-begin)
+    (define-key map "c" 'm2-case)
+    (define-key map "d" 'm2-definition)
+    (define-key map "e" 'm2-else)
+    (define-key map "f" 'm2-for)
+    (define-key map "h" 'm2-header)
+    (define-key map "i" 'm2-if)
+    (define-key map "l" 'm2-loop)
+    (define-key map "m" 'm2-module)
+    (define-key map "o" 'm2-or)
+    (define-key map "p" 'm2-procedure)
+    (define-key map "r" 'm2-record)
+    (define-key map "s" 'm2-stdio)
+    (define-key map "t" 'm2-type)
+    (define-key map "u" 'm2-until)
+    (define-key map "v" 'm2-var)
+    (define-key map "w" 'm2-while)
+    (define-key map "W" 'm2-with)
+    (define-key map "x" 'm2-export)
+    (define-key map "y" 'm2-import)
+    map)
+  "Modula2 quick keys keymap")
+
+(when m2-use-quick-keys
+  (define-key m2-mode-map m2-quick-keys-prefix m2-quick-keys-map))
 
 (defcustom m2-indent 5
   "*This variable gives the indentation in Modula-2-Mode."
@@ -149,8 +165,8 @@ followed by the first character of the c
   (setq paragraph-separate paragraph-start)
   (make-local-variable 'paragraph-ignore-fill-prefix)
   (setq paragraph-ignore-fill-prefix t)
-;  (make-local-variable 'indent-line-function)
-;  (setq indent-line-function 'c-indent-line)
+					;  (make-local-variable 'indent-line-function)
+					;  (setq indent-line-function 'c-indent-line)
   (make-local-variable 'require-final-newline)
   (setq require-final-newline mode-require-final-newline)
   (make-local-variable 'comment-start)

Re: Maintenance suggestion on an emacs mode

[Reto Zimmermann]

Please apply the following patch to vhdl-mode.el:

2726,2731c2726,2731
<   (define-key vhdl-mode-map "\C-c\C-c\C-n" 'vhdl-compose-new-component)
<   (define-key vhdl-mode-map "\C-c\C-c\C-p" 'vhdl-compose-place-component)
<   (define-key vhdl-mode-map "\C-c\C-c\C-w" 'vhdl-compose-wire-components)
<   (define-key vhdl-mode-map "\C-c\C-c\C-f" 'vhdl-compose-configuration)
<   (define-key vhdl-mode-map "\C-c\C-c\C-k" 
'vhdl-compose-components-package)
<   (define-key vhdl-mode-map "\C-cc"       'vhdl-comment-uncomment-region)
---
 >   (define-key vhdl-mode-map "\C-c\C-m\C-n" 'vhdl-compose-new-component)
 >   (define-key vhdl-mode-map "\C-c\C-m\C-p" 'vhdl-compose-place-component)
 >   (define-key vhdl-mode-map "\C-c\C-m\C-w" 'vhdl-compose-wire-components)
 >   (define-key vhdl-mode-map "\C-c\C-m\C-f" 'vhdl-compose-configuration)
 >   (define-key vhdl-mode-map "\C-c\C-m\C-k" 
'vhdl-compose-components-package)
 >   (define-key vhdl-mode-map "\C-c\C-c" 
'vhdl-comment-uncomment-region)
2770,2771c2770,2771
<   (define-key vhdl-mode-map "\C-cf"       'vhdl-fontify-buffer)
<   (define-key vhdl-mode-map "\C-cs"       'vhdl-statistics-buffer)
---
 >   (define-key vhdl-mode-map "\C-c\C-i\C-f" 'vhdl-fontify-buffer)
 >   (define-key vhdl-mode-map "\C-c\C-i\C-s" 'vhdl-statistics-buffer)

Reto


paul r wrote:

> Hello,
> 
> You are contacted because you are mentioned as the maintainer of a
> mode now included in GNU Emacs, development version 23 as of today, on
> which we suggest you to do some small modifications.
> Those modifications concern some default key bindings on your mode
> map, bound on the C-c[a-zA-Z] pattern, which is supposed to be
> reserved for users, as explained in the last paragraph of the link
> [1].
> Each of you will receive an other email with the file to fix and the
> matching lines. If possible, please send back a patch on emacs-devel
> with those keybindings changed.
> Thank you
> 
> -- Paul
> 
> [1] http://www.gnu.org/software/emacs/manual/html_node/emacs/Keymaps.html

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

> Please apply the following patch to vhdl-mode.el:

The patch was line-wrapped and failed.  Also please use (unified)
context diffs when sending patches.  And of course, please include
a ChangeLog entry along with the patch.  Thank you,


        Stefan


> 2726,2731c2726,2731
> <   (define-key vhdl-mode-map "\C-c\C-c\C-n" 'vhdl-compose-new-component)
> <   (define-key vhdl-mode-map "\C-c\C-c\C-p" 'vhdl-compose-place-component)
> <   (define-key vhdl-mode-map "\C-c\C-c\C-w" 'vhdl-compose-wire-components)
> <   (define-key vhdl-mode-map "\C-c\C-c\C-f" 'vhdl-compose-configuration)
> <   (define-key vhdl-mode-map "\C-c\C-c\C-k"
> vhdl-compose-components-package)
> <   (define-key vhdl-mode-map "\C-cc"       'vhdl-comment-uncomment-region)
> ---
>> (define-key vhdl-mode-map "\C-c\C-m\C-n" 'vhdl-compose-new-component)
>> (define-key vhdl-mode-map "\C-c\C-m\C-p" 'vhdl-compose-place-component)
>> (define-key vhdl-mode-map "\C-c\C-m\C-w" 'vhdl-compose-wire-components)
>> (define-key vhdl-mode-map "\C-c\C-m\C-f" 'vhdl-compose-configuration)
>> (define-key vhdl-mode-map "\C-c\C-m\C-k" 
> 'vhdl-compose-components-package)
>> (define-key vhdl-mode-map "\C-c\C-c" 
> 'vhdl-comment-uncomment-region)
> 2770,2771c2770,2771
> <   (define-key vhdl-mode-map "\C-cf"       'vhdl-fontify-buffer)
> <   (define-key vhdl-mode-map "\C-cs"       'vhdl-statistics-buffer)
> ---
>> (define-key vhdl-mode-map "\C-c\C-i\C-f" 'vhdl-fontify-buffer)
>> (define-key vhdl-mode-map "\C-c\C-i\C-s" 'vhdl-statistics-buffer)

> Reto


> paul r wrote:

>> Hello,
>> 
>> You are contacted because you are mentioned as the maintainer of a
>> mode now included in GNU Emacs, development version 23 as of today, on
>> which we suggest you to do some small modifications.
>> Those modifications concern some default key bindings on your mode
>> map, bound on the C-c[a-zA-Z] pattern, which is supposed to be
>> reserved for users, as explained in the last paragraph of the link
>> [1].
>> Each of you will receive an other email with the file to fix and the
>> matching lines. If possible, please send back a patch on emacs-devel
>> with those keybindings changed.
>> Thank you
>> 
>> -- Paul
>> 
>> [1] http://www.gnu.org/software/emacs/manual/html_node/emacs/Keymaps.html

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

> Please find attached patches.
> 3 questions :
>  - why is sgml in textmodes ?
>  - files are not indented according to emacs lisp default rules,
> should I resend a patch with full reindentation according-to-mode ?
> Attached patches do not change indentation.
>  - When testing this kind of modifications, I could not find a way to
> unload a mode so that I could make a modification, and reload the
> mode. I ended up closing and reopening an emacs instance for that :)

Seeing how you're participating actively, you should sign
copyright paperwork.  I might be able to consider the current patch as
a "tiny change", but that's already pushing it.

Would you be willing to assign your copyright to the FSF?
If so, please fill&send the form below,


        Stefan


Please email the following information to assign@gnu.org, and we
will send you the assignment form for your past and future changes.

Please use your full legal name (in ASCII characters) as the subject
line of the message.
----------------------------------------------------------------------
REQUEST: SEND FORM FOR PAST AND FUTURE CHANGES

[What is the name of the program or package you're contributing to?]
Emacs

[Did you copy any files or text written by someone else in these changes?
Even if that material is free software, we need to know about it.]


[Do you have an employer who might have a basis to claim to own
your changes?  Do you attend a school which might make such a claim?]


[For the copyright registration, what country are you a citizen of?]


[What year were you born?]


[Please write your email address here.]


[Please write your postal address here.]





[Which files have you changed so far, and which new files have you written
so far?]
sgml-mode.el, modula2.el

Re: Maintenance suggestion on an emacs mode

[paul r]

2008/4/4, Stefan Monnier <monnier@iro.umontreal.ca>:
> > Please find attached patches.
>  > 3 questions :
>  >  - why is sgml in textmodes ?
>  >  - files are not indented according to emacs lisp default rules,
>  > should I resend a patch with full reindentation according-to-mode ?
>  > Attached patches do not change indentation.
>  >  - When testing this kind of modifications, I could not find a way to
>  > unload a mode so that I could make a modification, and reload the
>  > mode. I ended up closing and reopening an emacs instance for that :)
>
>
> Seeing how you're participating actively, you should sign
>  copyright paperwork.  I might be able to consider the current patch as
>  a "tiny change", but that's already pushing it.
>
>  Would you be willing to assign your copyright to the FSF?
>  If so, please fill&send the form below,
>
>

I'm already in this process, papers are probably somewhere between
Boston and Paris.
Would you answer my questions, please ? It might save some of my time
later, particulary if I decide to fix myself all those modes for which
maintainers do not seem to be availables anymore. thx

paul

>         Stefan
>
>
>  Please email the following information to assign@gnu.org, and we
>
> will send you the assignment form for your past and future changes.
>
>  Please use your full legal name (in ASCII characters) as the subject
>  line of the message.
>  ----------------------------------------------------------------------
>  REQUEST: SEND FORM FOR PAST AND FUTURE CHANGES
>
>  [What is the name of the program or package you're contributing to?]
>  Emacs
>
>  [Did you copy any files or text written by someone else in these changes?
>  Even if that material is free software, we need to know about it.]
>
>
>  [Do you have an employer who might have a basis to claim to own
>  your changes?  Do you attend a school which might make such a claim?]
>
>
>  [For the copyright registration, what country are you a citizen of?]
>
>
>  [What year were you born?]
>
>
>  [Please write your email address here.]
>
>
>  [Please write your postal address here.]
>
>
>
>
>
>  [Which files have you changed so far, and which new files have you written
>  so far?]
>
> sgml-mode.el, modula2.el
>
>
>

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

> Please find attached patches.
> 3 questions :
>  - why is sgml in textmodes ?

Why not?  Although XML is now used for more than just text, SGML was
originally mostly used for text (and mostly for HTML, actually).

>  - files are not indented according to emacs lisp default rules,
> should I resend a patch with full reindentation according-to-mode ?
> Attached patches do not change indentation.

If there's incorrect indentation, yes, feel free to fix it (but only in
the last patch, not the ones submitted for review so the actual changes
don't get drowned in irrelevant reindentation).

>  - When testing this kind of modifications, I could not find a way to
> unload a mode so that I could make a modification, and reload the
> mode.  I ended up closing and reopening an emacs instance for that :)

You can try M-x unload-feature, but I typically just use C-M-x to
evaluate the function I just changed.


        Stefan

Re: Maintenance suggestion on an emacs mode

[Reto Zimmermann]

[-- Attachment #1: Type: text/plain, Size: 433 bytes --]

Stefan Monnier wrote:

>>Please apply the following patch to vhdl-mode.el:
> 
> The patch was line-wrapped and failed.  Also please use (unified)
> context diffs when sending patches.  And of course, please include
> a ChangeLog entry along with the patch.  Thank you,

Diff is attached.

ChangeLog is:

2008-04-01  Reto Zimmermann  <reto@gnu.org>

	* progmodes/vhdl-mode.el (vhdl-mode-map-init): Replace
	C-c[a-zA-Z] key bindings.


[-- Attachment #2: diff --]
[-- Type: text/plain, Size: 3062 bytes --]

*** vhdl-mode-cvs.el	Tue Apr  1 11:50:13 2008
--- vhdl-mode-cvs-keyb.el	Tue Apr  1 11:49:55 2008
***************
*** 2723,2734 ****
    (define-key vhdl-mode-map "\C-c\C-s\C-b" 'vhdl-subprog-paste-body)
    (define-key vhdl-mode-map "\C-c\C-s\C-c" 'vhdl-subprog-paste-call)
    (define-key vhdl-mode-map "\C-c\C-s\C-f" 'vhdl-subprog-flatten)
!   (define-key vhdl-mode-map "\C-c\C-c\C-n" 'vhdl-compose-new-component)
!   (define-key vhdl-mode-map "\C-c\C-c\C-p" 'vhdl-compose-place-component)
!   (define-key vhdl-mode-map "\C-c\C-c\C-w" 'vhdl-compose-wire-components)
!   (define-key vhdl-mode-map "\C-c\C-c\C-f" 'vhdl-compose-configuration)
!   (define-key vhdl-mode-map "\C-c\C-c\C-k" 'vhdl-compose-components-package)
!   (define-key vhdl-mode-map "\C-cc"	   'vhdl-comment-uncomment-region)
    (define-key vhdl-mode-map "\C-c-"	   'vhdl-comment-append-inline)
    (define-key vhdl-mode-map "\C-c\M--"	   'vhdl-comment-display-line)
    (define-key vhdl-mode-map "\C-c\C-i\C-l" 'indent-according-to-mode)
--- 2723,2734 ----
    (define-key vhdl-mode-map "\C-c\C-s\C-b" 'vhdl-subprog-paste-body)
    (define-key vhdl-mode-map "\C-c\C-s\C-c" 'vhdl-subprog-paste-call)
    (define-key vhdl-mode-map "\C-c\C-s\C-f" 'vhdl-subprog-flatten)
!   (define-key vhdl-mode-map "\C-c\C-m\C-n" 'vhdl-compose-new-component)
!   (define-key vhdl-mode-map "\C-c\C-m\C-p" 'vhdl-compose-place-component)
!   (define-key vhdl-mode-map "\C-c\C-m\C-w" 'vhdl-compose-wire-components)
!   (define-key vhdl-mode-map "\C-c\C-m\C-f" 'vhdl-compose-configuration)
!   (define-key vhdl-mode-map "\C-c\C-m\C-k" 'vhdl-compose-components-package)
!   (define-key vhdl-mode-map "\C-c\C-c"	   'vhdl-comment-uncomment-region)
    (define-key vhdl-mode-map "\C-c-"	   'vhdl-comment-append-inline)
    (define-key vhdl-mode-map "\C-c\M--"	   'vhdl-comment-display-line)
    (define-key vhdl-mode-map "\C-c\C-i\C-l" 'indent-according-to-mode)
***************
*** 2767,2774 ****
    (define-key vhdl-mode-map "\C-c\C-b"	   'vhdl-beautify-buffer)
    (define-key vhdl-mode-map "\C-c\C-u\C-s" 'vhdl-update-sensitivity-list-process)
    (define-key vhdl-mode-map "\C-c\C-u\M-s" 'vhdl-update-sensitivity-list-buffer)
!   (define-key vhdl-mode-map "\C-cf"	   'vhdl-fontify-buffer)
!   (define-key vhdl-mode-map "\C-cs"	   'vhdl-statistics-buffer)
    (define-key vhdl-mode-map "\C-c\M-m"	   'vhdl-show-messages)
    (define-key vhdl-mode-map "\C-c\C-h"	   'vhdl-doc-mode)
    (define-key vhdl-mode-map "\C-c\C-v"	   'vhdl-version)
--- 2767,2774 ----
    (define-key vhdl-mode-map "\C-c\C-b"	   'vhdl-beautify-buffer)
    (define-key vhdl-mode-map "\C-c\C-u\C-s" 'vhdl-update-sensitivity-list-process)
    (define-key vhdl-mode-map "\C-c\C-u\M-s" 'vhdl-update-sensitivity-list-buffer)
!   (define-key vhdl-mode-map "\C-c\C-i\C-f" 'vhdl-fontify-buffer)
!   (define-key vhdl-mode-map "\C-c\C-i\C-s" 'vhdl-statistics-buffer)
    (define-key vhdl-mode-map "\C-c\M-m"	   'vhdl-show-messages)
    (define-key vhdl-mode-map "\C-c\C-h"	   'vhdl-doc-mode)
    (define-key vhdl-mode-map "\C-c\C-v"	   'vhdl-version)

Re: Maintenance suggestion on an emacs mode

[paul r]

2008/4/4, Stefan Monnier <monnier@iro.umontreal.ca>:
> > Please find attached patches.
>  > 3 questions :
>  >  - why is sgml in textmodes ?
>
>
> Why not?  Although XML is now used for more than just text, SGML was
>  originally mostly used for text (and mostly for HTML, actually).

Fair enough

>  >  - When testing this kind of modifications, I could not find a way to
>  > unload a mode so that I could make a modification, and reload the
>  > mode.  I ended up closing and reopening an emacs instance for that :)
>
>
> You can try M-x unload-feature, but I typically just use C-M-x to
>  evaluate the function I just changed.

unload-feature is what I need, don't know how I missed it.
C-M-x, or re-loading the file, can not help for (defvar ...)

Thank you stefan

>
>
>
>         Stefan
>

Re: Maintenance suggestion on an emacs mode

[Andreas Schwab]

"paul r" <paul.r.ml@gmail.com> writes:

> C-M-x, or re-loading the file, can not help for (defvar ...)

eval-defun handles defvar specially.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Maintenance suggestion on an emacs mode

[Paul Rivier]

Andreas Schwab <schwab@suse.de> writes:

> "paul r" <paul.r.ml@gmail.com> writes:
>
>> C-M-x, or re-loading the file, can not help for (defvar ...)
>
> eval-defun handles defvar specially.
>

Thank you, I was mislead by the name and took it for a wrapper around
(end-of-defun)(eval-last-sexp). Handy function for developpment,
thanks !

> Andreas.
>
> -- 
> Andreas Schwab, SuSE Labs, schwab@suse.de
> SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
> PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
> "And now for something completely different."

-- 
Paul Rivier
Demotera
06.31.56.48.89

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

>> You can try M-x unload-feature, but I typically just use C-M-x to
>> evaluate the function I just changed.

> unload-feature is what I need,

But be aware that it is very brittle.  Sometimes it works.


        Stefan

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

>>> Please apply the following patch to vhdl-mode.el:
>> 
>> The patch was line-wrapped and failed.  Also please use (unified)
>> context diffs when sending patches.  And of course, please include
>> a ChangeLog entry along with the patch.  Thank you,

> Diff is attached.

Perfect, installed, thank you,


        Stefan

Re: Maintenance suggestion on an emacs mode

[Paul R]

Some news, see below :

> FSF
> lisp/progmodes/delphi.el:1896: '("\C-cu" delphi-find-unit)
Patch sent, should be applied soon

> FSF
> lisp/textmodes/sgml-mode.el:1583: (define-key map "\C-co" 'html-ordered-list)
> (...)
> lisp/textmodes/sgml-mode.el:1590: (define-key map "\C-ci" 'html-image))
Patch sent, should be applied soon

> Michael Schmidt <michael@pbinfo.UUCP> , email adress does not work
> lisp/progmodes/modula2.el:78: (define-key map "\C-cb" 'm2-begin)
> (...)
> lisp/progmodes/modula2.el:97: (define-key map "\C-cy" 'm2-import)
Patch sent, should be applied soon

Vhdl-mode is fixed, its maintainer sent a patch now applied.

Concerning other modes :

,---- RefTex :
| lisp/textmodes/reftex.el:2409: '(("\C-ct" . reftex-toc)
| lisp/textmodes/reftex.el:2410: ("\C-cl" . reftex-label)
| lisp/textmodes/reftex.el:2411: ("\C-cr" . reftex-reference)
| lisp/textmodes/reftex.el:2412: ("\C-cc" . reftex-citation)
| lisp/textmodes/reftex.el:2413: ("\C-cv" . reftex-view-crossref)
| lisp/textmodes/reftex.el:2414: ("\C-cg" . reftex-grep-document)
| lisp/textmodes/reftex.el:2415: ("\C-cs" . reftex-search-document))
`----


Those bindings are not activated by default, in this sens they are
much like those of modula2 and html-mode. Should they be changed in
the same way ?

,---- Gnus-Diary 
| Someone wrote something about it but I can't remember what. Emacs 
| really needs its bug-tracking system :)
`----

,---- Inf-lisp
| lisp/progmodes/inf-lisp.el:126: (define-key lisp-mode-map "\C-ce"
|                                'lisp-eval-defun-and-go)
| lisp/progmodes/inf-lisp.el:127: (define-key lisp-mode-map "\C-cr"
|                                 'lisp-eval-region-and-go)
| lisp/progmodes/inf-lisp.el:128: (define-key lisp-mode-map "\C-cc"
|                                 'lisp-compile-defun-and-go)
| lisp/progmodes/inf-lisp.el:129: (define-key lisp-mode-map "\C-cz"
|                                 'switch-to-lisp)
| lisp/progmodes/inf-lisp.el:130: (define-key lisp-mode-map "\C-cl"
|                                 'lisp-load-file)
| lisp/progmodes/inf-lisp.el:131: (define-key lisp-mode-map "\C-ck"
|                                 'lisp-compile-file)
| lisp/progmodes/inf-lisp.el:132: (define-key lisp-mode-map "\C-ca"
|                                 'lisp-show-arglist)
| lisp/progmodes/inf-lisp.el:133: (define-key lisp-mode-map "\C-cd"
|                                 'lisp-describe-sym)
| lisp/progmodes/inf-lisp.el:134: (define-key lisp-mode-map "\C-cf"
|                                 'lisp-show-function-documentation)
| lisp/progmodes/inf-lisp.el:135: (define-key lisp-mode-map "\C-cv"
|                                 'lisp-show-variable-documentation)
| lisp/progmodes/inf-lisp.el:137: (define-key inferior-lisp-mode-map
|                                 "\C-cl" 'lisp-load-file)
| lisp/progmodes/inf-lisp.el:138: (define-key inferior-lisp-mode-map
|                                 "\C-ck" 'lisp-compile-file)
| lisp/progmodes/inf-lisp.el:139: (define-key inferior-lisp-mode-map
|                                 "\C-ca" 'lisp-show-arglist)
| lisp/progmodes/inf-lisp.el:140: (define-key inferior-lisp-mode-map
|                                 "\C-cd" 'lisp-describe-sym)
| lisp/progmodes/inf-lisp.el:141: (define-key inferior-lisp-mode-map
|                                 "\C-cf" 'lisp-show-function-documentation)
| lisp/progmodes/inf-lisp.el:142: (define-key inferior-lisp-mode-map "\C-cv"
`----
Maintainer did not answer, should we fix that ?

,---- Ada Mode
| isp/progmodes/ada-mode.el:4590: (define-key ada-mode-map "\C-co" 'ff-find-other-file)
| lisp/progmodes/ada-mode.el:4595: (define-key ada-mode-map "\C-cc" 'ada-change-prj)
| lisp/progmodes/ada-mode.el:4596: (define-key ada-mode-map "\C-cd" 'ada-set-default-project-file)
| lisp/progmodes/ada-mode.el:4597: (define-key ada-mode-map "\C-cg" 'ada-gdb-application)
| lisp/progmodes/ada-mode.el:4599: (define-key ada-mode-map "\C-cr" 'ada-run-application)
| lisp/progmodes/ada-mode.el:4602: (define-key ada-mode-map "\C-cl" 'ada-find-local-references)
| lisp/progmodes/ada-mode.el:4604: (define-key ada-mode-map "\C-cf" 'ada-find-file)
| lisp/progmodes/ada-mode.el:4606: (define-key ada-mode-map "\C-cu" 'ada-prj-edit)
| lisp/progmodes/ada-mode.el:4639: (define-key ada-mode-map "\C-ct" map))
`----
Same as above

,---- rmail-spam-filter
| lisp/mail/rmail-spam-filter.el:566:(define-key rmail-summary-mode-map "\C-cSg" 'rsf-customize-group)
| lisp/mail/rmail-spam-filter.el:567:(define-key rmail-mode-map "\C-cSg" 'rsf-customize-group)
| lisp/mail/rmail-spam-filter.el:573:(define-key rmail-summary-mode-map "\C-cSd" 'rsf-customize-spam-definitions)
| lisp/mail/rmail-spam-filter.el:574:(define-key rmail-mode-map "\C-cSd" 'rsf-customize-spam-definitions)
| lisp/mail/rmail-spam-filter.el:583:(define-key rmail-summary-mode-map "\C-cSa" 'rsf-custom-save-all)
| lisp/mail/rmail-spam-filter.el:584:(define-key rmail-mode-map "\C-cSa" 'rsf-custom-save-all)
| lisp/mail/rmail-spam-filter.el:590:(define-key rmail-summary-mode-map "\C-cSn" 'rsf-add-region-to-spam-list)
| lisp/mail/rmail-spam-filter.el:591:(define-key rmail-mode-map "\C-cSn" 'rsf-add-region-to-spam-list)
| lisp/mail/rmail-spam-filter.el:597:(define-key rmail-summary-mode-map "\C-cSr" 'rsf-add-sender-to-spam-list)
| lisp/mail/rmail-spam-filter.el:598:(define-key rmail-mode-map "\C-cSr" 'rsf-add-sender-to-spam-list)
| lisp/mail/rmail-spam-filter.el:604:(define-key rmail-summary-mode-map "\C-cSt" 'rsf-add-subject-to-spam-list)
| lisp/mail/rmail-spam-filter.el:605:(define-key rmail-mode-map "\C-cSt" 'rsf-add-subject-to-spam-list)
`----
Same as above

,---- mixal-mode
| lisp/progmodes/mixal-mode.el:82: (define-key map "\C-cc" 'compile)
| lisp/progmodes/mixal-mode.el:83: (define-key map "\C-cr" 'mixal-run)
| lisp/progmodes/mixal-mode.el:84: (define-key map "\C-cd" 'mixal-debug)
`----
Same as above


-- Paul

Re: Maintenance suggestion on an emacs mode

[Paul R]

Some missing patches can be found below.
I did my best to make acceptable changes, but I might have missed
something, let me know if you find something bad.
I received the fsf papers, I'll send them back tomorrow.

> ,---- RefTex :
>
> Those bindings are not activated by default, in this sens they are
> much like those of modula2 and html-mode. Should they be changed in
> the same way ?

This patch change them.

Index: lisp/textmodes/reftex-vars.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/textmodes/reftex-vars.el,v
retrieving revision 1.44
diff -U 8 -r1.44 reftex-vars.el
--- lisp/textmodes/reftex-vars.el	10 Apr 2008 14:10:44 -0000	1.44
+++ lisp/textmodes/reftex-vars.el	21 Apr 2008 17:21:49 -0000
@@ -1844,17 +1844,26 @@
 ;; Miscellaneous configurations -----------------------------------------
 
 (defgroup reftex-miscellaneous-configurations nil
   "Collection of further configurations."
   :group 'reftex)
 
 (defcustom reftex-extra-bindings nil
   "Non-nil means, make additional key bindings on startup.
-These extra bindings are located in the users `C-c letter' map."
+These extra bindings are located in the
+`reftex-extra-bindings-map' map, bound to
+`reftex-extra-bindings-prefix'."
+  :group 'reftex-miscellaneous-configurations
+  :type 'boolean)
+
+;; below, default is C-c C-y because it is free in LaTeX mode.
+(defcustom reftex-extra-bindings-prefix "\C-c\C-y"
+  "When `reftex-extra-bindings' is set to non-nil, use extra
+bindings with this prefix bound to `reftex-extra-bindings-map'."
   :group 'reftex-miscellaneous-configurations
   :type 'boolean)
 
 (defcustom reftex-plug-into-AUCTeX nil
   "*Plug-in flags for AUCTeX interface.
 This variable is a list of 4 boolean flags.  When a flag is non-nil,
 RefTeX will
 
Index: lisp/textmodes/reftex.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/textmodes/reftex.el,v
retrieving revision 1.86
diff -U 8 -r1.86 reftex.el
--- lisp/textmodes/reftex.el	10 Apr 2008 14:10:45 -0000	1.86
+++ lisp/textmodes/reftex.el	21 Apr 2008 17:21:50 -0000
@@ -2399,26 +2399,33 @@
       'reftex-mouse-view-crossref)))
 
 ;; Bind `reftex-view-crossref-from-bibtex' in BibTeX mode map
 (eval-after-load
  "bibtex"
  '(define-key bibtex-mode-map "\C-c&" 'reftex-view-crossref-from-bibtex))
 
 ;; If the user requests so, she can have a few more bindings:
+(defvar reftex-extra-bindings-map
+  (let ((map (make-sparse-keymap)))
+    (define-key map "t" 'reftex-toc)
+    (define-key map "l" 'reftex-label)
+    (define-key map "r" 'reftex-reference)
+    (define-key map "c" 'reftex-citation)
+    (define-key map "v" 'reftex-view-crossref)
+    (define-key map "g" 'reftex-grep-document)
+    (define-key map "s" 'reftex-search-document)
+    map)
+  "Reftex extra bindings map")
+
 (when reftex-extra-bindings
-  (loop for x in
-        '(("\C-ct" . reftex-toc)
-          ("\C-cl" . reftex-label)
-          ("\C-cr" . reftex-reference)
-          ("\C-cc" . reftex-citation)
-          ("\C-cv" . reftex-view-crossref)
-          ("\C-cg" . reftex-grep-document)
-          ("\C-cs" . reftex-search-document))
-        do (define-key reftex-mode-map (car x) (cdr x))))
+  (define-key reftex-mode-map
+    reftex-extra-bindings-prefix
+    reftex-extra-bindings-map))
+    
 
 ;;; =========================================================================
 ;;;
 ;;; Menu
 
 ;; Define a menu for the menu bar if Emacs is running under X
 
 (defvar reftex-isearch-minor-mode nil)


> ,---- Gnus-Diary 
> | Someone wrote something about it but I can't remember what. Emacs 
> | really needs its bug-tracking system :)
> `----

No fixed.

> ,---- Inf-lisp
> `----
> Maintainer did not answer, should we fix that ?

Not fixed.

> ,---- Ada Mode
> `----
Fixed by the patch below.


Index: lisp/progmodes/ada-mode.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/progmodes/ada-mode.el,v
retrieving revision 1.94
diff -U 8 -r1.94 ada-mode.el
--- lisp/progmodes/ada-mode.el	24 Feb 2008 10:08:51 -0000	1.94
+++ lisp/progmodes/ada-mode.el	21 Apr 2008 17:21:48 -0000
@@ -445,16 +445,23 @@
 The extensions should include a `.' if needed.")
 
 (defvar ada-mode-menu (make-sparse-keymap "Ada")
   "Menu for Ada mode.")
 
 (defvar ada-mode-map (make-sparse-keymap)
   "Local keymap used for Ada mode.")
 
+(defvar ada-mode-extra-map (make-sparse-keymap)
+  "Keymap used for non-standard keybindings.")
+
+;; default is C-c C-q because it's free in ada-mode-map
+(defvar ada-mode-extra-prefix "\C-c\C-q"
+  "Prefix key to access `ada-mode-extra-map' functions.")
+
 (defvar ada-mode-abbrev-table nil
   "Local abbrev table for Ada mode.")
 
 (defvar ada-mode-syntax-table nil
   "Syntax table to be used for editing Ada source code.")
 
 (defvar ada-mode-symbol-syntax-table nil
   "Syntax table for Ada, where `_' is a word constituent.")
@@ -4534,16 +4541,19 @@
 \f
 ;; ------------------------------------------------------------
 ;; --  Define keymap and menus for Ada
 ;; -------------------------------------------------------------
 
 (defun ada-create-keymap ()
   "Create the keymap associated with the Ada mode."
 
+  ;; All non-standard keys go into ada-mode-extra-map
+  (define-key ada-mode-map ada-mode-extra-prefix ada-mode-extra-map)
+
   ;; Indentation and Formatting
   (define-key ada-mode-map "\C-j"     'ada-indent-newline-indent-conditional)
   (define-key ada-mode-map "\C-m"     'ada-indent-newline-indent-conditional)
   (define-key ada-mode-map "\t"       'ada-tab)
   (define-key ada-mode-map "\C-c\t"   'ada-justified-indent-current)
   (define-key ada-mode-map "\C-c\C-l" 'ada-indent-region)
   (define-key ada-mode-map [(shift tab)]    'ada-untab)
   (define-key ada-mode-map "\C-c\C-f" 'ada-format-paramlist)
@@ -4582,33 +4592,33 @@
   ;; The following keys are bound to functions defined in ada-xref.el or
   ;; ada-prj,el., However, RMS rightly thinks that the code should be shared,
   ;; and activated only if the right compiler is used
 
   (define-key ada-mode-map (if (featurep 'xemacs) '(shift button3) [S-mouse-3])
     'ada-point-and-xref)
   (define-key ada-mode-map [(control tab)] 'ada-complete-identifier)
 
-  (define-key ada-mode-map "\C-co"     'ff-find-other-file)
+  (define-key ada-mode-extra-map "o"     'ff-find-other-file)
   (define-key ada-mode-map "\C-c5\C-d" 'ada-goto-declaration-other-frame)
   (define-key ada-mode-map "\C-c\C-d"  'ada-goto-declaration)
   (define-key ada-mode-map "\C-c\C-s"  'ada-xref-goto-previous-reference)
   (define-key ada-mode-map "\C-c\C-c"  'ada-compile-application)
-  (define-key ada-mode-map "\C-cc"     'ada-change-prj)
-  (define-key ada-mode-map "\C-cd"     'ada-set-default-project-file)
-  (define-key ada-mode-map "\C-cg"     'ada-gdb-application)
+  (define-key ada-mode-extra-map "c"     'ada-change-prj)
+  (define-key ada-mode-extra-map "d"     'ada-set-default-project-file)
+  (define-key ada-mode-extra-map "g"     'ada-gdb-application)
   (define-key ada-mode-map "\C-c\C-m"  'ada-set-main-compile-application)
-  (define-key ada-mode-map "\C-cr"     'ada-run-application)
+  (define-key ada-mode-extra-map "r"     'ada-run-application)
   (define-key ada-mode-map "\C-c\C-o"  'ada-goto-parent)
   (define-key ada-mode-map "\C-c\C-r"  'ada-find-references)
-  (define-key ada-mode-map "\C-cl"     'ada-find-local-references)
+  (define-key ada-mode-extra-map "l"     'ada-find-local-references)
   (define-key ada-mode-map "\C-c\C-v"  'ada-check-current)
-  (define-key ada-mode-map "\C-cf"     'ada-find-file)
+  (define-key ada-mode-extra-map "f"     'ada-find-file)
 
-  (define-key ada-mode-map "\C-cu"  'ada-prj-edit)
+  (define-key ada-mode-extra-map "u"  'ada-prj-edit)
 
   ;;  The templates, defined in ada-stmt.el
 
   (let ((map (make-sparse-keymap)))
     (define-key map "h"    'ada-header)
     (define-key map "\C-a" 'ada-array)
     (define-key map "b"    'ada-exception-block)
     (define-key map "d"    'ada-declare-block)
@@ -4631,17 +4641,17 @@
     (define-key map "\C-y" 'ada-type)
     (define-key map "\C-v" 'ada-private)
     (define-key map "u"    'ada-use)
     (define-key map "\C-u" 'ada-with)
     (define-key map "\C-w" 'ada-when)
     (define-key map "w"    'ada-while-loop)
     (define-key map "\C-x" 'ada-exception)
     (define-key map "x"    'ada-exit)
-    (define-key ada-mode-map "\C-ct" map))
+    (define-key ada-mode-extra-map "t" map))
   )
 
 
 (defun ada-create-menu ()
   "Create the Ada menu as shown in the menu bar."
   (let ((m '("Ada"
 	     ("Help"
 	      ["Ada Mode"               (info "ada-mode") t]

>
> ,---- rmail-spam-filter
> `----
Not fixed.

> ,---- mixal-mode
> `----
Fixed by the patch below.

Index: lisp/progmodes/mixal-mode.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/progmodes/mixal-mode.el,v
retrieving revision 1.16
diff -U 8 -r1.16 mixal-mode.el
--- lisp/progmodes/mixal-mode.el	8 Jan 2008 20:46:20 -0000	1.16
+++ lisp/progmodes/mixal-mode.el	21 Apr 2008 17:21:48 -0000
@@ -74,20 +74,20 @@
 ;; 09/11/02: started mixal-mode.
 
 ;;; Code:
 (defvar compile-command)
 
 ;;; Key map
 (defvar mixal-mode-map
   (let ((map (make-sparse-keymap)))
-    (define-key map "\C-cc" 'compile)
-    (define-key map "\C-cr" 'mixal-run)
-    (define-key map "\C-cd" 'mixal-debug)
-    (define-key map "\C-ho" 'mixal-describe-operation-code)
+    (define-key map "\C-c\C-c" 'compile)
+    (define-key map "\C-c\C-r" 'mixal-run)
+    (define-key map "\C-c\C-d" 'mixal-debug)
+    (define-key map "\C-h\C-o" 'mixal-describe-operation-code)
     map)
   "Keymap for `mixal-mode'.")
 ;; (makunbound 'mixal-mode-map)
 
 ;;; Syntax table
 (defvar mixal-mode-syntax-table
   (let ((st (make-syntax-table)))
     ;; We need to do a bit more to make fontlocking for comments work.




-- 
      Paul

Re: Maintenance suggestion on an emacs mode

[Ralf Angeli]

* Paul R. (2008-04-21) writes:

>> ,---- RefTex :
>>
>> Those bindings are not activated by default, in this sens they are
>> much like those of modula2 and html-mode. Should they be changed in
>> the same way ?
>
> This patch change them.

>  (when reftex-extra-bindings
> -  (loop for x in
> -        '(("\C-ct" . reftex-toc)
> -          ("\C-cl" . reftex-label)
> -          ("\C-cr" . reftex-reference)
> -          ("\C-cc" . reftex-citation)
> -          ("\C-cv" . reftex-view-crossref)
> -          ("\C-cg" . reftex-grep-document)
> -          ("\C-cs" . reftex-search-document))

Please don't change those.  There are already bindings for them (except
for the last two) in place.  Setting `reftex-extra-bindings' really is
only there to spare users the hassle of defining bindings in the user
space themselves.

-- 
Ralf

Re: Maintenance suggestion on an emacs mode

[Paul R]

Ralf Angeli <angeli@caeruleus.net> writes:

> * Paul R. (2008-04-21) writes:
>
>>> ,---- RefTex :
>>>
>>> Those bindings are not activated by default, in this sens they are
>>> much like those of modula2 and html-mode. Should they be changed in
>>> the same way ?
>>
>> This patch change them.
> (...)
> Please don't change those.  There are already bindings for them (except
> for the last two) in place.  Setting `reftex-extra-bindings' really is
> only there to spare users the hassle of defining bindings in the user
> space themselves.

I have no particular position on that. It is in a move to clean up
sources from C-c[a-Z] bindings.
As far as this particular modification is concerned, I think it is
acceptable because the user before had to do :
  (setq reftex-extra-bindings t)
to have them automatically bound to C-c, and now she would have to do
  (setq reftex-extra-bindings t)
  (setq reftex-extra-bindings-prefix "\C-c")
for the same effect. Not that much hassle added, and more
Info-compliant. But again, I won't push it, so please feel free to
argue against this change.


-- 
      Paul

Re: Maintenance suggestion on an emacs mode

[Chong Yidong]

Ralf Angeli <angeli@caeruleus.net> writes:

> Please don't change those.  There are already bindings for them (except
> for the last two) in place.  Setting `reftex-extra-bindings' really is
> only there to spare users the hassle of defining bindings in the user
> space themselves.

Could you add a comment to reftex.el saying that we are aware of the
violation of the "no C-c [letter] binding" rule, with the explanation
you gave above?


Paul R <paul.r.ml@gmail.com> writes:

>> Please don't change those.  There are already bindings for them (except
>> for the last two) in place.  Setting `reftex-extra-bindings' really is
>> only there to spare users the hassle of defining bindings in the user
>> space themselves.
>
> I have no particular position on that. It is in a move to clean up
> sources from C-c[a-Z] bindings.
> As far as this particular modification is concerned, I think it is
> acceptable because ...

With the above change, we can leave these alone.  Thanks.

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

>> Please don't change those.  There are already bindings for them (except
>> for the last two) in place.  Setting `reftex-extra-bindings' really is
>> only there to spare users the hassle of defining bindings in the user
>> space themselves.

> Could you add a comment to reftex.el saying that we are aware of the
> violation of the "no C-c [letter] binding" rule, with the explanation
> you gave above?


> Paul R <paul.r.ml@gmail.com> writes:

>>> Please don't change those.  There are already bindings for them (except
>>> for the last two) in place.  Setting `reftex-extra-bindings' really is
>>> only there to spare users the hassle of defining bindings in the user
>>> space themselves.
>> 
>> I have no particular position on that. It is in a move to clean up
>> sources from C-c[a-Z] bindings.
>> As far as this particular modification is concerned, I think it is
>> acceptable because ...

> With the above change, we can leave these alone.  Thanks.

Maybe we could apply his change, but make C-c the default (i.e. the
default is to break the convention).  So it would still be easy for
someone to unbreak the conventions.


        Stefan

Re: Maintenance suggestion on an emacs mode

[Ralf Angeli]

* Stefan Monnier (2008-04-23) writes:

> Maybe we could apply his change, but make C-c the default (i.e. the
> default is to break the convention).  So it would still be easy for
> someone to unbreak the conventions.

I cannot imagine somebody changing the default.  And if somebody did and
used something like the proposed `C-c C-y', the bindings would be in the
namespace of major modes.  By the way, the proposal of `C-c C-y' would
also violate the key binding recommendations for minor modes.

Another option would be to remove those bindings altogether and put the
code for defining the key bindings into the manual for easy copying and
pasting into the init file.

-- 
Ralf

Re: Maintenance suggestion on an emacs mode

[Stefan Monnier]

>> Maybe we could apply his change, but make C-c the default (i.e. the
>> default is to break the convention).  So it would still be easy for
>> someone to unbreak the conventions.

> I cannot imagine somebody changing the default.  And if somebody did and
> used something like the proposed `C-c C-y', the bindings would be in the
> namespace of major modes.

The user is allowed to use any binding whatsoever, os if she wants
"\C-c\C-y" or "\e" or even just "", that's fine by me.  The "reserved
for users" namespace is just there to guarantee that this is something
that will never conflict.

> By the way, the proposal of `C-c C-y' would also violate the key
> binding recommendations for minor modes.

Indeed.  Which is why a "\C-c" default is no worse.

> Another option would be to remove those bindings altogether and put the
> code for defining the key bindings into the manual for easy copying and
> pasting into the init file.

Doesn't seem like an improvement,


        Stefan

Re: Maintenance suggestion on an emacs mode

[Paul R]

Ralf Angeli <angeli@caeruleus.net> writes:

> * Stefan Monnier (2008-04-23) writes:
>
> I cannot imagine somebody changing the default.  And if somebody did and
> used something like the proposed `C-c C-y', the bindings would be in the
> namespace of major modes.  By the way, the proposal of `C-c C-y' would
> also violate the key binding recommendations for minor modes.

I tried reftex + emacs latex mode, and reftex + auctex latex mode. In
both cases, C-c C-y was free, that's why I took it.
As for key bindings recommandations for minor modes, I don't know them,
may you point me to them please ? If the problem is the 3 strokes
needed, how to circumvent this when there is no space left in 2 stokes
keybindings space ?

>
> Another option would be to remove those bindings altogether and put the
> code for defining the key bindings into the manual for easy copying and
> pasting into the init file.

Indeed.

-- 
      Paul

Re: Maintenance suggestion on an emacs mode

[Lennart Borgman (gmail)]

Stefan Monnier wrote:
>>> Maybe we could apply his change, but make C-c the default (i.e. the
>>> default is to break the convention).  So it would still be easy for
>>> someone to unbreak the conventions.

Moving the code doing the key bindings to the defcustom 
reftex-extra-bindings makes it a little bit easier to change the 
bindings. It seems easy enough to also add code to remove the key bindings.

Would not that be enough instead of breaking the key binding 
conventions? And should not the doc string for this defcustom say 
something about that it nearly breaks the conventions?

Re: Maintenance suggestion on an emacs mode

[Ralf Angeli]

* Paul R. (2008-04-23) writes:

> As for key bindings recommandations for minor modes, I don't know them,
> may you point me to them please ?

(info "(elisp)Keymaps and Minor Modes")

-- 
Ralf

Re: Maintenance suggestion on an emacs mode

[Ralf Angeli]

* Stefan Monnier (2008-04-23) writes:

>> Another option would be to remove those bindings altogether and put the
>> code for defining the key bindings into the manual for easy copying and
>> pasting into the init file.
>
> Doesn't seem like an improvement,

Then I'd suggest to just add a comment in reftex.el as proposed earlier
in this thread.  Following is a ChangeLog entry and patch.


2008-04-24  Ralf Angeli  <angeli@caeruleus.net>

	* textmodes/reftex.el: Explain in a comment why
	`reftex-extra-bindings' activates key bindings reserverd for
	users.


--- reftex.el.~1.86.~	2008-04-19 08:52:29.000000000 +0200
+++ reftex.el	2008-04-24 22:19:18.000000000 +0200
@@ -2403,7 +2403,10 @@
  "bibtex"
  '(define-key bibtex-mode-map "\C-c&" 'reftex-view-crossref-from-bibtex))
 
-;; If the user requests so, she can have a few more bindings:
+;; For most of these commands there are already bindings in place.
+;; Setting `reftex-extra-bindings' really is only there to spare users
+;; the hassle of defining bindings in the user space themselves.  This
+;; is why they violate the key binding recommendations.
 (when reftex-extra-bindings
   (loop for x in
         '(("\C-ct" . reftex-toc)


-- 
Ralf

Re: Maintenance suggestion on an emacs mode

[Lennart Borgman (gmail)]

Ralf Angeli wrote:
> * Stefan Monnier (2008-04-23) writes:
> 
>>> Another option would be to remove those bindings altogether and put the
>>> code for defining the key bindings into the manual for easy copying and
>>> pasting into the init file.
>> Doesn't seem like an improvement,
> 
> Then I'd suggest to just add a comment in reftex.el as proposed earlier
> in this thread.  

I suggested moving this to the defcustom instead (and of course updating 
its doc string). Is not that easier for the user?

> Following is a ChangeLog entry and patch.
> 
> 
> 2008-04-24  Ralf Angeli  <angeli@caeruleus.net>
> 
> 	* textmodes/reftex.el: Explain in a comment why
> 	`reftex-extra-bindings' activates key bindings reserverd for
> 	users.
> 
> 
> --- reftex.el.~1.86.~	2008-04-19 08:52:29.000000000 +0200
> +++ reftex.el	2008-04-24 22:19:18.000000000 +0200
> @@ -2403,7 +2403,10 @@
>   "bibtex"
>   '(define-key bibtex-mode-map "\C-c&" 'reftex-view-crossref-from-bibtex))
>  
> -;; If the user requests so, she can have a few more bindings:
> +;; For most of these commands there are already bindings in place.
> +;; Setting `reftex-extra-bindings' really is only there to spare users
> +;; the hassle of defining bindings in the user space themselves.  This
> +;; is why they violate the key binding recommendations.
>  (when reftex-extra-bindings
>    (loop for x in
>          '(("\C-ct" . reftex-toc)
> 
> 

Re: Maintenance suggestion on an emacs mode

[Reiner Steib]

On Mon, Apr 21 2008, Paul R wrote:

>> ,---- Gnus-Diary 
>> | Someone wrote something about it but I can't remember what. Emacs 
>> | really needs its bug-tracking system :)
>> `----
>
> No fixed.

Fixed in Gnus CVS (not sure if it has been synced to Emacs CVS
already.  See <news:v9mynysrsi.fsf@marauder.physik.uni-ulm.de> or
<http://thread.gmane.org/v9mynysrsi.fsf@marauder.physik.uni-ulm.de>

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/