why do I have to <C-u> before calling align?

why do I have to <C-u> before calling align?

[Le Wang]

[-- Attachment #1: Type: text/plain, Size: 206 bytes --]

Hi all,

I'm scratching my head over this one.  In Emacs 23.2.1, calling `align'
directly never works, I always have to  pass the universal arg.  It looks
like the function only accepts the region.

-- 
Le

[-- Attachment #2: Type: text/html, Size: 270 bytes --]

RE: why do I have to <C-u> before calling align?

[Drew Adams]

> I'm scratching my head over this one.  In Emacs 23.2.1,
> calling `align' directly never works, I always have to
> pass the universal arg.  It looks like the function only
> accepts the region.

Recipe please. Otherwise you won't be the only one scratching your head. ;-) How
are you calling it? etc.  How are you "passing" the universal arg?

There seems to be no direct reference to the universal arg or
`current-prefix-arg' in the `align' code.  (`align-regexp' is a different
story.)

However, I do see this in the user option `align-rules-list', which is used by
`align'.  Some of the `run-if' functions examine the prefix arg:

 `((lisp-second-arg
     (regexp   . "\\(^\\s-+[^( \t\n]\\|(\\(\\S-+\\)\\s-+\\)\\S-+\\(\\s-+\\)")
     (group    . 3)
     (modes    . align-lisp-modes)
     (run-if   . ,(function (lambda () current-prefix-arg))))
   ...
   (c-comma-delimiter
     (regexp   . ",\\(\\s-*\\)[^/ \t\n]")
     (repeat   . t)
     (modes    . align-c++-modes)
     (run-if   . ,(function (lambda () current-prefix-arg))))
   ...
   ;; With a prefix argument, comma delimiter will be aligned.  Since
   ;; perl-mode doesn't give us enough syntactic information (and we
   ;; don't do our own parsing yet), this rule is too destructive to
   ;; run normally.
   (basic-comma-delimiter
     (regexp   . ",\\(\\s-*\\)[^# \t\n]")
     (repeat   . t)
     (modes    . (append align-perl-modes '(python-mode)))
     (run-if   . ,(function (lambda () current-prefix-arg))))
   ...
   ;; With a numeric prefix argument, or C-u, space delimited text
   ;; tables will be aligned.
   (text-column
     (regexp   . "\\(^\\|\\S-\\)\\([ \t]+\\)\\(\\S-\\|$\\)")
     (group    . 2)
     (modes    . align-text-modes)
     (repeat   . t)
     (run-if   . ,(function
		   (lambda ()
		     (and current-prefix-arg
			  (not (eq '- current-prefix-arg)))))))
   ;; With a negative prefix argument, lists of dollar figures will
   ;; be aligned.
   (text-dollar-figure
     (regexp   . "\\$?\\(\\s-+[0-9]+\\)\\.")
     (modes    . align-text-modes)
     (justify  . t)
     (run-if   . ,(function
		   (lambda ()
		     (eq '- current-prefix-arg)))))

So maybe that's where it comes in.  Again, recipe please.

(Disclaimer: I know nothing about `align' and have never used it.)

Re: why do I have to <C-u> before calling align?

[Le Wang]

[-- Attachment #1: Type: text/plain, Size: 443 bytes --]

On Fri, Dec 31, 2010 at 7:31 AM, Drew Adams <drew.adams@oracle.com> wrote:

> However, I do see this in the user option `align-rules-list', which is used
> by
> `align'.  Some of the `run-if' functions examine the prefix arg:
>
>
This thread fell off my radar.

Thanks for the helpful reply Drew.  The magic is in align-rules-list.  I
guess some of the useful rules I rely on are considered too aggressive to be
turned on by default.


-- 
Le

[-- Attachment #2: Type: text/html, Size: 789 bytes --]