completing-read in pcvs.el

completing-read in pcvs.el

[kimitto]

(emacs-version)
"GNU Emacs 22.1.1 (i386-mingw-nt5.1.2600)
 of 2007-07-08 on CUBE"

completing-read()'s second arg requires alist (or hash-table,
obarray),
but in pcvs.el uses a list.

Calling cvs-mode-set-flags() makes error
if uning a tiny original completion elisp.

Fix it, please.


Best regards,
Kimito Yada


diff -u c\:/meadow/lisp/pcvs.el c\:/meadow/lisp/pcvs.mod.el
--- c:/meadow/lisp/pcvs.el	2007-07-08 14:28:00.001000000 +0900
+++ c:/meadow/lisp/pcvs.mod.el	2008-07-24 23:52:23.020236800 +0900
@@ -2217,8 +2217,8 @@
   (interactive
    (list (completing-read
 	  "Which flag: "
-	  '("cvs" "diff" "update" "status" "log" "tag" ;"rtag"
-	    "commit" "remove" "undo" "checkout")
+	  '(("cvs") ("diff") ("update") ("status") ("log") ("tag") ;("rtag")
+	    ("commit") ("remove") ("undo") ("checkout"))
 	  nil t)))
   (let* ((sym (intern (concat "cvs-" flag "-flags"))))
     (let ((current-prefix-arg '(16)))

bug#894: completing-read in pcvs.el

[Stefan Monnier]

> completing-read()'s second arg requires alist (or hash-table,
> obarray), but in pcvs.el uses a list.

Emacs also accepts a list (at least since Emacs-22, maybe even
Emacs-21), so that's not a problem.

> Calling cvs-mode-set-flags() makes error
> if uning a tiny original completion elisp.

Could you describe in which circumstance the problem manifests itself?


        Stefan

bug#939: completing-read in pcvs.el

[kimito yada]

> Emacs also accepts a list (at least since Emacs-22, maybe even
> Emacs-21), so that's not a problem.
Sorry, I couldn't catch the deescription for try-completion in info elisp-3.

"In particular, a list of strings or symbols is allowed, even though
     we usually do not think of such lists as alists."

And I understood it.


But I think, using completion-read should unite,
pcvs.el:1280 uses alist, but 2247 list.


Thank you,
Kimito Yada

bug#945: completing-read in pcvs.el

[Stefan Monnier]

>> Emacs also accepts a list (at least since Emacs-22, maybe even
>> Emacs-21), so that's not a problem.
> Sorry, I couldn't catch the deescription for try-completion in info elisp-3.

> "In particular, a list of strings or symbols is allowed, even though
>      we usually do not think of such lists as alists."

> And I understood it.

> But I think, using completion-read should unite,
> pcvs.el:1280 uses alist, but 2247 list.

I do not understnad "should unite".  Rather than talk about the code,
please tell me which behavior of PCL-CVS is affected.  I.e. give
a recipe exhibiting undesirable behavior.


        Stefan

bug#948: completing-read in pcvs.el

[kimito yada]

> please tell me which behavior of PCL-CVS is affected.  I.e. give
> a recipe exhibiting undesirable behavior.

I'm using modified cycle-mini.el.  (original http://joereiss.net/misc/cycle-mini.el)

1. start emacs with -q option
2. load cycle-mini.el
3. M-x cvs-examine and select <working dir>.
4. M-x cvs-mode-set-flags
5. --> Wrong type argument: listp, "cvs"

I noticed earlier that cycle-mini-reset() must be corrected.


Sorry for my poor English.

Kimito Yada