PCL-CVS: M-x cvs-update vs. key binding in *cvs* buffer

PCL-CVS: M-x cvs-update vs. key binding in *cvs* buffer

[Reiner Steib]

Hi,

`M-x cvs-update RET' prompts for a directory and creates a *cvs*
buffer.  Do `M-x cvs-update RET' in the *cvs* buffer.  Again, it
prompts for a directory.

Now eval (global-set-key (kbd "C-c v u") 'cvs-update).  Then, when
pressing `C-c v u' in the *cvs* buffer, it won't ask for a directory.

I'd expect `cvs-update' to behave the same, no matter how it is called
(M-x vs. key binding).

It seems that `cvs-query-directory' tries to figure out if it was
called via M-x.  But I don't quite understand the purpose.

--8<---------------cut here---------------start------------->8---
(defun cvs-query-directory (msg)
  ;; last-command-char = ?\r hints that the command was run via M-x
  (if (and (cvs-buffer-p)

	   (not current-prefix-arg)
	   (not (eq last-command-char ?\r)))
      default-directory
    (read-directory-name msg nil default-directory nil)))
--8<---------------cut here---------------end--------------->8---

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

Re: PCL-CVS: M-x cvs-update vs. key binding in *cvs* buffer

[Stefan Monnier]

> It seems that `cvs-query-directory' tries to figure out if it was
> called via M-x.  But I don't quite understand the purpose.

I can't remember the original reason.  It may have been to preserve the
original behavior which was to always prompt for a directory.
But since C-u does the job as well we don't need that kind of hack.
Feel free to remove it,


        Stefan

Re: PCL-CVS: M-x cvs-update vs. key binding in *cvs* buffer

[Reiner Steib]

On Tue, Aug 21 2007, Stefan Monnier wrote:

>> It seems that `cvs-query-directory' tries to figure out if it was
>> called via M-x.  But I don't quite understand the purpose.
>
> I can't remember the original reason.  It may have been to preserve the
> original behavior which was to always prompt for a directory.
> But since C-u does the job as well we don't need that kind of hack.
> Feel free to remove it,

Like this (maybe someone could suggest a nicer doc string)?
Should it go to go to Emacs 22.2 as well?

--8<---------------cut here---------------start------------->8---
(defun cvs-query-directory (prompt)
  "Read directory name, prompting with PROMPT.
If in a *cvs* buffer, don't prompt unless a prefix argument is given."
  (if (and (cvs-buffer-p)
	   (not current-prefix-arg))
      default-directory
    (read-directory-name prompt nil default-directory nil)))
--8<---------------cut here---------------end--------------->8---

--8<---------------cut here---------------start------------->8---
--- pcvs.el	12 Aug 2007 18:07:07 -0000	1.98.2.7
+++ pcvs.el	22 Aug 2007 20:01:16 -0000
@@ -981,13 +981,13 @@
   (interactive)
   (cvs-examine default-directory t))
 
-(defun cvs-query-directory (msg)
-  ;; last-command-char = ?\r hints that the command was run via M-x
+(defun cvs-query-directory (prompt)
+  "Read directory name, prompting with PROMPT.
+If in a *cvs* buffer, don't prompt unless a prefix argument is given."
   (if (and (cvs-buffer-p)
-	   (not current-prefix-arg)
-	   (not (eq last-command-char ?\r)))
+	   (not current-prefix-arg))
       default-directory
-    (read-directory-name msg nil default-directory nil)))
+    (read-directory-name prompt nil default-directory nil)))
 
 ;;;###autoload
 (defun cvs-quickdir (dir &optional flags noshow)
--8<---------------cut here---------------end--------------->8---

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

Re: PCL-CVS: M-x cvs-update vs. key binding in *cvs* buffer

[Stefan Monnier]

> Like this (maybe someone could suggest a nicer doc string)?

Looks good, yes.

> Should it go to go to Emacs 22.2 as well?

I have no opinion on that.


        Stefan

Re: PCL-CVS: M-x cvs-update vs. key binding in *cvs* buffer

[Reiner Steib]

On Wed, Aug 22 2007, Stefan Monnier wrote:

>> Like this (maybe someone could suggest a nicer doc string)?
>
> Looks good, yes.
>
>> Should it go to go to Emacs 22.2 as well?
>
> I have no opinion on that.

Installed on EMACS_22_BASE and TRUNK.

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