all messages for Emacs-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
* flypell and check-comments.
@ 2007-03-07 14:36 Michaël Cadilhac
  2007-03-07 18:28 ` martin rudalics
  0 siblings, 1 reply; 15+ messages in thread
From: Michaël Cadilhac @ 2007-03-07 14:36 UTC (permalink / raw)
  To: emacs-devel


[-- Attachment #1.1.1: Type: text/plain, Size: 238 bytes --]

Hi!

The ispell package has an interesting feature governed by
``ispell-check-comments'', this feature behaves weirdly with flyspell.

In fact, ispell-check-comments is not at all managed by flyspell.

I propose the following to do that:

[-- Attachment #1.1.2: flyspell.patch --]
[-- Type: text/x-patch, Size: 3958 bytes --]

Index: lisp/textmodes/flyspell.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/textmodes/flyspell.el,v
retrieving revision 1.115
diff -B -w -c -r1.115 flyspell.el
*** lisp/textmodes/flyspell.el	21 Jan 2007 03:53:09 -0000	1.115
--- lisp/textmodes/flyspell.el	7 Mar 2007 13:40:11 -0000
***************
*** 369,382 ****
    '(font-lock-string-face font-lock-comment-face font-lock-doc-face)
    "Faces corresponding to text in programming-mode buffers.")
  
  (defun flyspell-generic-progmode-verify ()
    "Used for `flyspell-generic-check-word-predicate' in programming modes."
!   (let ((f (get-text-property (point) 'face)))
!     (memq f flyspell-prog-text-faces)))
  
  ;;;###autoload
  (defun flyspell-prog-mode ()
!   "Turn on `flyspell-mode' for comments and strings."
    (interactive)
    (setq flyspell-generic-check-word-predicate
          'flyspell-generic-progmode-verify)
--- 369,391 ----
    '(font-lock-string-face font-lock-comment-face font-lock-doc-face)
    "Faces corresponding to text in programming-mode buffers.")
  
+ (defun flyspell-face-at-point-in-p (point list)
+   "Check if the face at POINT is in LIST."
+   (let ((f (get-text-property point 'face)))
+     (if (listp f)
+ 	(catch 'found
+ 	  (dolist (face f)
+ 	    (when (memq face list) (throw 'found t))))
+       (memq f list))))
+ 
  (defun flyspell-generic-progmode-verify ()
    "Used for `flyspell-generic-check-word-predicate' in programming modes."
!   (flyspell-face-at-point-in-p (point) flyspell-prog-text-faces))
  
  ;;;###autoload
  (defun flyspell-prog-mode ()
!   "Turn on `flyspell-mode' for comments and strings.
! Note that comment checking is also affected by `ispell-check-comments'."
    (interactive)
    (setq flyspell-generic-check-word-predicate
          'flyspell-generic-progmode-verify)
***************
*** 1008,1014 ****
             start end poss word ispell-filter)
        (if (or (eq flyspell-word nil)
   	      (and (fboundp flyspell-generic-check-word-predicate)
!  		   (not (funcall flyspell-generic-check-word-predicate))))
  	  t
  	(progn
  	  ;; destructure return flyspell-word info list.
--- 1017,1026 ----
             start end poss word ispell-filter)
        (if (or (eq flyspell-word nil)
   	      (and (fboundp flyspell-generic-check-word-predicate)
!  		   (not (funcall flyspell-generic-check-word-predicate)))
! 	      (and (not ispell-check-comments)
! 		   (flyspell-face-at-point-in-p (car (cdr flyspell-word))
! 						'(font-lock-comment-face))))
  	  t
  	(progn
  	  ;; destructure return flyspell-word info list.
***************
*** 1561,1566 ****
--- 1573,1581 ----
  	  (let ((old beg))
  	    (setq beg end)
  	    (setq end old)))
+       ;; Fontify the part checked, so that font-lock based tests are
+       ;; fine.
+       (font-lock-fontify-region beg end)
        (if (and flyspell-large-region (> (- end beg) flyspell-large-region))
  	  (flyspell-large-region beg end)
  	(flyspell-small-region beg end)))))
Index: lisp/ChangeLog
===================================================================
RCS file: /sources/emacs/emacs/lisp/ChangeLog,v
retrieving revision 1.10783
diff -C0 -r1.10783 ChangeLog
*** lisp/ChangeLog	7 Mar 2007 12:50:23 -0000	1.10783
--- lisp/ChangeLog	7 Mar 2007 13:43:33 -0000
***************
*** 0 ****
--- 1,11 ----
+ 2007-03-07  Michaël Cadilhac  <michael@cadilhac.name>
+ 
+ 	* textmodes/flyspell.el (flyspell-face-at-point-in-p): New.  Check
+ 	if the face at a specified point is in a face list.
+ 	(flyspell-generic-progmode-verify): Use it.
+ 	(flyspell-prog-mode): Talk about `ispell-check-comments' in the
+ 	docstring.
+ 	(flyspell-word): Consider `ispell-check-comments': if nil, don't
+ 	check text with face `font-lock-comment-face'.
+ 	(flyspell-region): Fontify the region before processing it.
+ 

[-- Attachment #1.1.3: Type: text/plain, Size: 608 bytes --]


Note the last part of the patch, the fontification. This not only for
ispell-check-comments : other flyspell features uses font-locking
based tests, and this modification is more like a bugfix than a part
of the patch (and it may be better to install it separately).

TIA!

-- 
 |   Michaël `Micha' Cadilhac       |  «Tu aimeras ton prochain.»            |
 |   http://michael.cadilhac.name   |    D'abord, Dieu ou pas,               |
 |   JID/MSN:                       |       j'ai horreur qu'on me tutoie.    |
 `----  michael.cadilhac@gmail.com  |           -- P. Desproges         -  --'

[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-07 14:36 flypell and check-comments Michaël Cadilhac
@ 2007-03-07 18:28 ` martin rudalics
  2007-03-07 18:42   ` Michaël Cadilhac
  2007-03-07 20:31   ` Stefan Monnier
  0 siblings, 2 replies; 15+ messages in thread
From: martin rudalics @ 2007-03-07 18:28 UTC (permalink / raw)
  To: Michaël Cadilhac; +Cc: emacs-devel

> +       ;; Fontify the part checked, so that font-lock based tests are
> +       ;; fine.
> +       (font-lock-fontify-region beg end)

I'd use `jit-lock-fontify-now' here.

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-07 18:28 ` martin rudalics
@ 2007-03-07 18:42   ` Michaël Cadilhac
  2007-03-07 20:31   ` Stefan Monnier
  1 sibling, 0 replies; 15+ messages in thread
From: Michaël Cadilhac @ 2007-03-07 18:42 UTC (permalink / raw)
  To: martin rudalics; +Cc: emacs-devel


[-- Attachment #1.1: Type: text/plain, Size: 635 bytes --]

martin rudalics <rudalics@gmx.at> writes:

>> +       ;; Fontify the part checked, so that font-lock based tests are
>> +       ;; fine.
>> +       (font-lock-fontify-region beg end)
>
> I'd use `jit-lock-fontify-now' here.

You're probably right, I'm not at all in the font-lock thing :-)
Thanks!

-- 
 |   Michaël `Micha' Cadilhac       |  To be portable,                       |
 |   http://michael.cadilhac.name   |    Just stay on Windows.               |
 |   JID/MSN:                       |                                        |
 `----  michael.cadilhac@gmail.com  |          -- A Microsoft Guy       -  --'

[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-07 18:28 ` martin rudalics
  2007-03-07 18:42   ` Michaël Cadilhac
@ 2007-03-07 20:31   ` Stefan Monnier
  2007-03-07 21:55     ` martin rudalics
  2007-03-11  4:24     ` Richard Stallman
  1 sibling, 2 replies; 15+ messages in thread
From: Stefan Monnier @ 2007-03-07 20:31 UTC (permalink / raw)
  To: martin rudalics; +Cc: Micha?l Cadilhac, emacs-devel

>> +       ;; Fontify the part checked, so that font-lock based tests are
>> +       ;; fine.
>> +       (font-lock-fontify-region beg end)

> I'd use `jit-lock-fontify-now' here.

Actually it should be neither:
- font-lock-fontify-region will unnecessarily
  refontify the region if it was already fontified.
- jit-lock-fontify-now is wrong if the user is not using jit-lock.
- if the user decided not to use font-lock...

There's a need for a new function `font-lock-ensure-fontified' which would
do nothing if font-lock doesn't use any helper thingy like lazy-lock or
font-lock and which would call jit-lock-fontify-now (or the equivalent for
lazy-lock) otherwise.

ps-print currently uses ps-print-ensure-fontified to work around the lack of
font-lock-ensure-fontified.  And I'm pretty sure ps-print is not the only
such package.


        Stefan

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-07 20:31   ` Stefan Monnier
@ 2007-03-07 21:55     ` martin rudalics
  2007-03-11  4:24     ` Richard Stallman
  1 sibling, 0 replies; 15+ messages in thread
From: martin rudalics @ 2007-03-07 21:55 UTC (permalink / raw)
  To: Stefan Monnier; +Cc: Micha?l Cadilhac, emacs-devel

 > Actually it should be neither:
 > - font-lock-fontify-region will unnecessarily
 >   refontify the region if it was already fontified.

That was my concern here ...

 > - jit-lock-fontify-now is wrong if the user is not using jit-lock.

... `occur-mode' has ...

	    (if (fboundp 'jit-lock-fontify-now)
		(jit-lock-fontify-now beg end)))

 > - if the user decided not to use font-lock...

... it would be best (and cheaper) to use `syntax-ppss' and
`parse-partial-sexp'.

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-07 20:31   ` Stefan Monnier
  2007-03-07 21:55     ` martin rudalics
@ 2007-03-11  4:24     ` Richard Stallman
  2007-03-11 12:59       ` Michaël Cadilhac
  1 sibling, 1 reply; 15+ messages in thread
From: Richard Stallman @ 2007-03-11  4:24 UTC (permalink / raw)
  To: Stefan Monnier; +Cc: rudalics, michael, emacs-devel

Has a fix for this problem been installed?

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-11  4:24     ` Richard Stallman
@ 2007-03-11 12:59       ` Michaël Cadilhac
  2007-03-11 14:38         ` martin rudalics
                           ` (2 more replies)
  0 siblings, 3 replies; 15+ messages in thread
From: Michaël Cadilhac @ 2007-03-11 12:59 UTC (permalink / raw)
  To: rms; +Cc: rudalics, Stefan Monnier, emacs-devel


[-- Attachment #1.1.1: Type: text/plain, Size: 189 bytes --]

Richard Stallman <rms@gnu.org> writes:

> Has a fix for this problem been installed?

Not yet.  If I follow Stefan's and Martin's advice (well, I hope
I really did), I have the following:


[-- Attachment #1.1.2: flyspell.patch --]
[-- Type: text/x-patch, Size: 5281 bytes --]

Index: lisp/font-lock.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/font-lock.el,v
retrieving revision 1.313
diff -c -r1.313 font-lock.el
*** lisp/font-lock.el	18 Feb 2007 20:36:06 -0000	1.313
--- lisp/font-lock.el	11 Mar 2007 11:49:31 -0000
***************
*** 1004,1009 ****
--- 1004,1020 ----
    (font-lock-set-defaults)
    (funcall font-lock-fontify-region-function beg end loudly))
  
+ ;; Avoid requiring 'jit-lock and 'lazy-lock.
+ (defalias 'font-lock--jitify 'jit-lock-fontify-now)
+ (defalias 'font-lock--lazify 'lazy-lock-fontify-region)
+ 
+ (defun font-lock-ensure-fontified (beg end)
+   "Ensure the region BEG..END is fontified, if needed."
+   (cond ((and (boundp 'jit-lock-mode) (symbol-value 'jit-lock-mode))
+ 	 (font-lock--jitify beg end))
+ 	((and (boundp 'lazy-lock-mode) (symbol-value 'lazy-lock-mode))
+ 	 (font-lock--lazify beg end))))
+ 
  (defun font-lock-unfontify-region (beg end)
    (save-buffer-state nil
      (funcall font-lock-unfontify-region-function beg end)))
Index: lisp/textmodes/flyspell.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/textmodes/flyspell.el,v
retrieving revision 1.115
diff -c -r1.115 flyspell.el
*** lisp/textmodes/flyspell.el	21 Jan 2007 03:53:09 -0000	1.115
--- lisp/textmodes/flyspell.el	11 Mar 2007 11:49:33 -0000
***************
*** 369,382 ****
    '(font-lock-string-face font-lock-comment-face font-lock-doc-face)
    "Faces corresponding to text in programming-mode buffers.")
  
  (defun flyspell-generic-progmode-verify ()
    "Used for `flyspell-generic-check-word-predicate' in programming modes."
!   (let ((f (get-text-property (point) 'face)))
!     (memq f flyspell-prog-text-faces)))
  
  ;;;###autoload
  (defun flyspell-prog-mode ()
!   "Turn on `flyspell-mode' for comments and strings."
    (interactive)
    (setq flyspell-generic-check-word-predicate
          'flyspell-generic-progmode-verify)
--- 369,391 ----
    '(font-lock-string-face font-lock-comment-face font-lock-doc-face)
    "Faces corresponding to text in programming-mode buffers.")
  
+ (defun flyspell-face-at-point-in-p (list)
+   "Check if the face at point is in LIST."
+   (let ((f (get-text-property (point) 'face)))
+     (if (listp f)
+ 	(catch 'found
+ 	  (dolist (face f)
+ 	    (when (memq face list) (throw 'found t))))
+       (memq f list))))
+ 
  (defun flyspell-generic-progmode-verify ()
    "Used for `flyspell-generic-check-word-predicate' in programming modes."
!   (flyspell-face-at-point-in-p flyspell-prog-text-faces))
  
  ;;;###autoload
  (defun flyspell-prog-mode ()
!   "Turn on `flyspell-mode' for comments and strings.
! Note that comment checking is also affected by `ispell-check-comments'."
    (interactive)
    (setq flyspell-generic-check-word-predicate
          'flyspell-generic-progmode-verify)
***************
*** 1008,1014 ****
             start end poss word ispell-filter)
        (if (or (eq flyspell-word nil)
   	      (and (fboundp flyspell-generic-check-word-predicate)
!  		   (not (funcall flyspell-generic-check-word-predicate))))
  	  t
  	(progn
  	  ;; destructure return flyspell-word info list.
--- 1017,1025 ----
             start end poss word ispell-filter)
        (if (or (eq flyspell-word nil)
   	      (and (fboundp flyspell-generic-check-word-predicate)
!  		   (not (funcall flyspell-generic-check-word-predicate)))
! 	      (and (not ispell-check-comments)
! 		   (nth 4 (syntax-ppss (car (cdr flyspell-word))))))
  	  t
  	(progn
  	  ;; destructure return flyspell-word info list.
***************
*** 1561,1566 ****
--- 1572,1580 ----
  	  (let ((old beg))
  	    (setq beg end)
  	    (setq end old)))
+       ;; Fontify the part checked, so that font-lock based tests are
+       ;; fine.
+       (font-lock-ensure-fontified beg end)
        (if (and flyspell-large-region (> (- end beg) flyspell-large-region))
  	  (flyspell-large-region beg end)
  	(flyspell-small-region beg end)))))
Index: lisp/ChangeLog
===================================================================
RCS file: /sources/emacs/emacs/lisp/ChangeLog,v
retrieving revision 1.10783
diff -C0 -r1.10783 ChangeLog
*** lisp/ChangeLog	7 Mar 2007 12:50:23 -0000	1.10783
--- lisp/ChangeLog	11 Mar 2007 11:49:50 -0000
***************
*** 0 ****
--- 1,17 ----
+ 2007-03-11  Michaël Cadilhac  <michael@cadilhac.name>
+ 
+ 	* textmodes/flyspell.el (flyspell-face-at-point-in-p): New function.
+ 	Check if the face at point is in a given list.
+ 	(flyspell-generic-progmode-verify): Use it.
+ 	(flyspell-prog-mode): Mention that `ispell-check-comments' affects
+ 	the way comments are detected.
+ 	(flyspell-word): If `ispell-check-comments' is nil, check if we are
+ 	in a comment, if so, don't check the word.
+ 	(flyspell-region): Ensure fontification before flyspelling the region,
+ 	so that font-lock based tests work.
+ 
+ 	* font-lock.el (font-lock--jitify, font-lock--lazify)
+ 	(font-lock-ensure-fontified): New aliases and function.  Ensure that
+ 	a region is fontified.  Taken from `ps-print.el'.
+ 	Suggested by Stefan Monnier.
+

[-- Attachment #1.1.3: Type: text/plain, Size: 511 bytes --]


After what we can safely change ps-print to use
font-lock-ensure-fontified.

Stefan, Martin, is the code here what you meant (it may not be,
font-locking is not my cup of tea)?

-- 
 |   Michaël `Micha' Cadilhac       |  Fenouil                               |
 |   http://michael.cadilhac.name   |                                        |
 |   JID/MSN:                       |    -- Poème de François Le Lionnais    |
 `----  michael.cadilhac@gmail.com  |                                   -  --'

[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-11 12:59       ` Michaël Cadilhac
@ 2007-03-11 14:38         ` martin rudalics
  2007-03-11 17:25           ` Michaël Cadilhac
  2007-03-11 21:58         ` Stefan Monnier
  2007-03-24 16:55         ` Chong Yidong
  2 siblings, 1 reply; 15+ messages in thread
From: martin rudalics @ 2007-03-11 14:38 UTC (permalink / raw)
  To: Michaël Cadilhac; +Cc: emacs-devel, rms, Stefan Monnier

 > Stefan, Martin, is the code here what you meant (it may not be,
 > font-locking is not my cup of tea)?

Maybe the following should be reconsidered

 > ! 	      (and (not ispell-check-comments)
 > ! 		   (nth 4 (syntax-ppss (car (cdr flyspell-word))))))

since on the one hand you don't check for `ispell-check-comments' being
eq to 'exclusive (hence you don't emulate ispell faithfully - I don't
know whether this is even desirable) while on the other hand you don't
check whether you're in a string (hence you don't fully emulate the
default of `flyspell-prog-text-faces' either).  But maybe I'm wrong.

Also `syntax-ppss' may get expensive for `flyspell-large-region'.  It's
generally better to use it for the first position only, save position
and parse-state, and then do `parse-partial-sexp' from the last saved
position and parse-state.

By no means I want(ed) to express here that syntax _should_ be parsed.
What I meant was that if a user does not want to use font-lock _and_
wants to spell-check comments or strings specially, parsing syntax is
the only practical way.  Hence, IMHO ignoring `ispell-check-comments'
here is a viable alternative as well.

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-11 14:38         ` martin rudalics
@ 2007-03-11 17:25           ` Michaël Cadilhac
  2007-03-11 19:52             ` martin rudalics
  0 siblings, 1 reply; 15+ messages in thread
From: Michaël Cadilhac @ 2007-03-11 17:25 UTC (permalink / raw)
  To: martin rudalics; +Cc: emacs-devel, rms, Stefan Monnier


[-- Attachment #1.1: Type: text/plain, Size: 1762 bytes --]

martin rudalics <rudalics@gmx.at> writes:

>> Stefan, Martin, is the code here what you meant (it may not be,
>> font-locking is not my cup of tea)?
>
> Maybe the following should be reconsidered
>
>> ! 	      (and (not ispell-check-comments)
>> ! 		   (nth 4 (syntax-ppss (car (cdr flyspell-word))))))
>
> since on the one hand you don't check for `ispell-check-comments' being
> eq to 'exclusive

I don't think it matters. At least, if it has to be fixed, it's not
here, but maybe in `flyspell-generic-progmode-verify'. Well, I don't
really know.

> while on the other hand you don't check whether you're in a string
> (hence you don't fully emulate the default of
> `flyspell-prog-text-faces' either).

I don't understand, why should I check that I'm in a string?

> Also `syntax-ppss' may get expensive for `flyspell-large-region'.  It's
> generally better to use it for the first position only, save position
> and parse-state, and then do `parse-partial-sexp' from the last saved
> position and parse-state.

I probably agree :) However, I don't really see how this can be
implemented. If ispell has to be mimicked, maybe the test should be
made using comment-start -- comment-end, but not sure.

> Hence, IMHO ignoring `ispell-check-comments' here is a viable
> alternative as well.

Well, this is not just for the fun to make a patch, I really want to
use `ispell-check-comments' (to nil).

-- 
 |   Michaël `Micha' Cadilhac       |    This .sig has been generated        |
 |   http://michael.cadilhac.name   |         by Outlook Express 98          |
 |   JID/MSN:                       |    and triple-checked on               |
 `----  michael.cadilhac@gmail.com  |         Windows Mail Vista.       -  --'

[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-11 17:25           ` Michaël Cadilhac
@ 2007-03-11 19:52             ` martin rudalics
  2007-03-12 13:01               ` Michaël Cadilhac
  0 siblings, 1 reply; 15+ messages in thread
From: martin rudalics @ 2007-03-11 19:52 UTC (permalink / raw)
  To: Michaël Cadilhac; +Cc: emacs-devel, rms, Stefan Monnier

 >>since on the one hand you don't check for `ispell-check-comments' being
 >>eq to 'exclusive
 >
 >
 > I don't think it matters. At least, if it has to be fixed, it's not
 > here, but maybe in `flyspell-generic-progmode-verify'. Well, I don't
 > really know.

As far as I understand this, `ispell-check-comments' t means check
comments too, 'exclusive means check comments only, somehow inverting
the sense of nil.  When using faces, nil means don't check when the face
is font-lock-comment-face, 'exclusive means check iff the face is
font-lock-comment-face, t means faces don't matter.  When using
parse-state it's equally simple: nil means don't check for (nth 4
parse-state), exclusive means check iff (nth 4 parse-state), and t means
parse-state doesn't matter.  Trivial.

 >>while on the other hand you don't check whether you're in a string
 >>(hence you don't fully emulate the default of
 >>`flyspell-prog-text-faces' either).
 >
 >
 > I don't understand, why should I check that I'm in a string?

Because the default value of `flyspell-prog-text-faces' has
`font-lock-string-face', `font-lock-comment-face' and
`font-lock-doc-face'.  To emulate this you would have to spell-check iff
either (nth 3 parse-state) or (nth 4 parse-state) hold.  If a user
wanted to spell-check doc-strings only this wouldn't work though.

 >>Also `syntax-ppss' may get expensive for `flyspell-large-region'.  It's
 >>generally better to use it for the first position only, save position
 >>and parse-state, and then do `parse-partial-sexp' from the last saved
 >>position and parse-state.
 >
 >
 > I probably agree :) However, I don't really see how this can be
 > implemented. If ispell has to be mimicked, maybe the test should be
 > made using comment-start -- comment-end, but not sure.

It would have to be done in `flyspell-external-point-words' before
`flyspell-word' is called.  Never mind.  If there are few spelling
errors `syntax-ppss' might be even faster.

 >>Hence, IMHO ignoring `ispell-check-comments' here is a viable
 >>alternative as well.
 >
 >
 > Well, this is not just for the fun to make a patch, I really want to
 > use `ispell-check-comments' (to nil).

In that case you should probably implement the 'exclusive option as
well.  Otherwise you should insert a note about this.

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-11 12:59       ` Michaël Cadilhac
  2007-03-11 14:38         ` martin rudalics
@ 2007-03-11 21:58         ` Stefan Monnier
  2007-03-24 16:55         ` Chong Yidong
  2 siblings, 0 replies; 15+ messages in thread
From: Stefan Monnier @ 2007-03-11 21:58 UTC (permalink / raw)
  To: Michaël Cadilhac; +Cc: rudalics, rms, emacs-devel

> + ;; Avoid requiring 'jit-lock and 'lazy-lock.
> + (defalias 'font-lock--jitify 'jit-lock-fontify-now)
> + (defalias 'font-lock--lazify 'lazy-lock-fontify-region)
> + 
> + (defun font-lock-ensure-fontified (beg end)
> +   "Ensure the region BEG..END is fontified, if needed."
> +   (cond ((and (boundp 'jit-lock-mode) (symbol-value 'jit-lock-mode))
> + 	 (font-lock--jitify beg end))
> + 	((and (boundp 'lazy-lock-mode) (symbol-value 'lazy-lock-mode))
> + 	 (font-lock--lazify beg end))))
> + 

Actually, I'd rather you do it as

  (defvar font-lock-ensure-fontified-function 'ignore)

and then in jit-lock.el set it (buffer-locally) to jit-lock-fontify-now.


        Stefan

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-11 19:52             ` martin rudalics
@ 2007-03-12 13:01               ` Michaël Cadilhac
  2007-03-12 18:59                 ` martin rudalics
  0 siblings, 1 reply; 15+ messages in thread
From: Michaël Cadilhac @ 2007-03-12 13:01 UTC (permalink / raw)
  To: martin rudalics; +Cc: emacs-devel, rms, Stefan Monnier


[-- Attachment #1.1: Type: text/plain, Size: 2478 bytes --]

martin rudalics <rudalics@gmx.at> writes:

>>>since on the one hand you don't check for `ispell-check-comments' being
>>>eq to 'exclusive
>>
>> I don't think it matters. At least, if it has to be fixed, it's not
>> here, but maybe in `flyspell-generic-progmode-verify'. Well, I don't
>> really know.
>
> As far as I understand this, `ispell-check-comments' t means check
> comments too, 'exclusive means check comments only, somehow inverting
> the sense of nil.

That's my understanding too.

> When using faces, nil means don't check when the face is
> font-lock-comment-face, 'exclusive means check iff the face is
> font-lock-comment-face, t means faces don't matter. When using
> parse-state it's equally simple: nil means don't check for (nth
> 4 parse-state), exclusive means check iff (nth 4 parse-state), and
> t means parse-state doesn't matter. Trivial.

Yes, you're right.  Well, I was sick yesterday :-)

>>>while on the other hand you don't check whether you're in a string
>>>(hence you don't fully emulate the default of
>>>`flyspell-prog-text-faces' either).
>>
>> I don't understand, why should I check that I'm in a string?
>
> Because the default value of `flyspell-prog-text-faces' has
> `font-lock-string-face', `font-lock-comment-face' and
> `font-lock-doc-face'.  To emulate this you would have to spell-check iff
> either (nth 3 parse-state) or (nth 4 parse-state) hold.  If a user
> wanted to spell-check doc-strings only this wouldn't work though.

Sorry, I still don't understand :-/.  I don't want to emulate any of
existing features of flyspell, the point is just to add one.

> It would have to be done in `flyspell-external-point-words' before
> `flyspell-word' is called.  Never mind.  If there are few spelling
> errors `syntax-ppss' might be even faster.

You're the knowledgeable one for me :-)

>> Well, this is not just for the fun to make a patch, I really want to
>> use `ispell-check-comments' (to nil).
>
> In that case you should probably implement the 'exclusive option as
> well.  Otherwise you should insert a note about this.

You're right.

Thanks.

-- 
 |   Michaël `Micha' Cadilhac       |  <ESC>ape this <COLON> thing,          |
 |   http://michael.cadilhac.name   |           <Q>uit and                   |
 |   JID/MSN:                       |    do <NOT> <RET>urn.                  |
 `----  michael.cadilhac@gmail.com  |          -- VI                    -  --'

[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-12 13:01               ` Michaël Cadilhac
@ 2007-03-12 18:59                 ` martin rudalics
  0 siblings, 0 replies; 15+ messages in thread
From: martin rudalics @ 2007-03-12 18:59 UTC (permalink / raw)
  To: Michaël Cadilhac; +Cc: emacs-devel, rms, Stefan Monnier

> Sorry, I still don't understand :-/.  I don't want to emulate any of
> existing features of flyspell, the point is just to add one.

My point was that in absence of font-locking it would have been possible
to spell-check comments and/or strings only by looking at the 3rd and 4th
return values of the parse-state.  If someone really needs this it will
be easy to add, hence there's no need to think about it any further.

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-11 12:59       ` Michaël Cadilhac
  2007-03-11 14:38         ` martin rudalics
  2007-03-11 21:58         ` Stefan Monnier
@ 2007-03-24 16:55         ` Chong Yidong
  2007-03-25 19:48           ` Stefan Monnier
  2 siblings, 1 reply; 15+ messages in thread
From: Chong Yidong @ 2007-03-24 16:55 UTC (permalink / raw)
  To: emacs-devel

michael@cadilhac.name (Michaël Cadilhac) writes:

> Not yet.  If I follow Stefan's and Martin's advice (well, I hope
> I really did), I have the following:
>
> ...
>
> Stefan, Martin, is the code here what you meant (it may not be,
> font-locking is not my cup of tea)?

What's the situation with this patch?

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: flypell and check-comments.
  2007-03-24 16:55         ` Chong Yidong
@ 2007-03-25 19:48           ` Stefan Monnier
  0 siblings, 0 replies; 15+ messages in thread
From: Stefan Monnier @ 2007-03-25 19:48 UTC (permalink / raw)
  To: Chong Yidong; +Cc: emacs-devel

>> Not yet.  If I follow Stefan's and Martin's advice (well, I hope
>> I really did), I have the following:
>> 
>> ...
>> 
>> Stefan, Martin, is the code here what you meant (it may not be,
>> font-locking is not my cup of tea)?

> What's the situation with this patch?

Last I heard I just told him that I didn't want a font-lock-ensure-fontified
function which had just a big switch to handle the 3 known cases, but
instead one which just called font-lock-ensure-fontified-function and set
this function when turning on jit-lock-mode.


        Stefan

^ permalink raw reply	[flat|nested] 15+ messages in thread

end of thread, other threads:[~2007-03-25 19:48 UTC | newest]

Thread overview: 15+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-03-07 14:36 flypell and check-comments Michaël Cadilhac
2007-03-07 18:28 ` martin rudalics
2007-03-07 18:42   ` Michaël Cadilhac
2007-03-07 20:31   ` Stefan Monnier
2007-03-07 21:55     ` martin rudalics
2007-03-11  4:24     ` Richard Stallman
2007-03-11 12:59       ` Michaël Cadilhac
2007-03-11 14:38         ` martin rudalics
2007-03-11 17:25           ` Michaël Cadilhac
2007-03-11 19:52             ` martin rudalics
2007-03-12 13:01               ` Michaël Cadilhac
2007-03-12 18:59                 ` martin rudalics
2007-03-11 21:58         ` Stefan Monnier
2007-03-24 16:55         ` Chong Yidong
2007-03-25 19:48           ` Stefan Monnier

Code repositories for project(s) associated with this external index

	https://git.savannah.gnu.org/cgit/emacs.git
	https://git.savannah.gnu.org/cgit/emacs/org-mode.git

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.