bug#4069: 23.1.50; whitespace.el

bug#4069: 23.1.50; whitespace.el

[Craig Falls]

There is a problem with whitespace-cleanup, when called before
whitespace-turn-on.  The buffer-local variable
whitespace-indentation-regexp doesn't get updated to correspond to the
buffer-local value of indent-tabs-mode.  This means that even if
indent-tabs-mode is nil, a line with text following a single tab
character will not be converted into a line of text following 8 space
characters when whitespace-cleanup is called.  There is a simple
work-around -- just turn whitespace on and off again to set the
buffer-local variables:

(defun my-whitespace-cleanup ()
  "Fixes a bug in whitespace-cleanup in which buffer-local
indent-tabs-mode is not respected."
  (interactive)
  (whitespace-turn-on) ;; sets buffer-local variables
  (whitespace-turn-off) ;; don't actually want the mode to be on
  (whitespace-cleanup))

Here are instructions to replicate the problem:

Start up with no .emacs.  Set indent-tabs-mode to nil.  Make sure
'indentation is in whitespace-style.  Insert a tab character into a
buffer with no spaces after it, e.g. the line could be "\tfoo".  Run
whitespace-cleanup in the buffer.  Note that the tab character is still
there.  Now run my-whitespace-cleanup.  Note that the tab character has
been replaced by spaces.

The fact that indent-tabs-mode is buffer-local is important.  For
example, makefile-mode sets it to true since tabs are a necessary part
of the syntax of make files.

In GNU Emacs 23.1.50.8 (x86_64-unknown-linux-gnu, GTK+ Version 2.10.4)
 of 2009-08-03 on nyc-qws-005
Windowing system distributor `The X.Org Foundation', version 11.0.70101000
Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: en_US.UTF-8
  value of $XMODIFIERS: nil
  locale-coding-system: utf-8-unix
  default-enable-multibyte-characters: t

Major mode: GNUmakefile

Minor modes in effect:
  shell-dirtrack-mode: t
  flyspell-mode: t
  desktop-save-mode: t
  diff-auto-refine-mode: t
  partial-completion-mode: t
  dynamic-completion-mode: t
  show-paren-mode: t
  recentf-mode: t
  iswitchb-mode: t
  global-auto-revert-mode: t
  tooltip-mode: t
  mouse-wheel-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  global-auto-composition-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  column-number-mode: t
  line-number-mode: t
  transient-mark-mode: t

Load-path shadows:
/mnt/global/dev/lib/elisp/inf-caml hides /mnt/global/dev/lib/elisp/caml-mode/inf-caml
/mnt/global/dev/lib/elisp/tuareg-mode/camldebug hides /mnt/global/dev/lib/elisp/caml-mode/camldebug

bug#4069: 23.1.50; whitespace.el

[Chong Yidong]

Hi Vinicius,

Could you take a look at this bug report?  Thanks:


"Craig Falls" <cfalls@janestcapital.com> wrote:

> There is a problem with whitespace-cleanup, when called before
> whitespace-turn-on.  The buffer-local variable
> whitespace-indentation-regexp doesn't get updated to correspond to the
> buffer-local value of indent-tabs-mode.  This means that even if
> indent-tabs-mode is nil, a line with text following a single tab
> character will not be converted into a line of text following 8 space
> characters when whitespace-cleanup is called.  There is a simple
> work-around -- just turn whitespace on and off again to set the
> buffer-local variables:

> (defun my-whitespace-cleanup ()
>   "Fixes a bug in whitespace-cleanup in which buffer-local
> indent-tabs-mode is not respected."
>   (interactive)
>   (whitespace-turn-on) ;; sets buffer-local variables
>   (whitespace-turn-off) ;; don't actually want the mode to be on
>   (whitespace-cleanup))

> Here are instructions to replicate the problem:

> Start up with no .emacs.  Set indent-tabs-mode to nil.  Make sure
> 'indentation is in whitespace-style.  Insert a tab character into a
> buffer with no spaces after it, e.g. the line could be "\tfoo".  Run
> whitespace-cleanup in the buffer.  Note that the tab character is still
> there.  Now run my-whitespace-cleanup.  Note that the tab character has
> been replaced by spaces.

> The fact that indent-tabs-mode is buffer-local is important.  For
> example, makefile-mode sets it to true since tabs are a necessary part
> of the syntax of make files.

bug#4069: 23.1.50; whitespace.el

[Vinicius Jose Latorre]

Hi Chong,


> Could you take a look at this bug report?  Thanks:
>
>
> "Craig Falls" <cfalls@janestcapital.com> wrote:
>
>   
>> There is a problem with whitespace-cleanup, when called before
>> whitespace-turn-on.  The buffer-local variable
>> whitespace-indentation-regexp doesn't get updated to correspond to the
>> buffer-local value of indent-tabs-mode.  This means that even if
>> indent-tabs-mode is nil, a line with text following a single tab
>> character will not be converted into a line of text following 8 space
>> characters when whitespace-cleanup is called.  There is a simple
>> work-around -- just turn whitespace on and off again to set the
>> buffer-local variables:
>>     
>
>   
>> (defun my-whitespace-cleanup ()
>>   "Fixes a bug in whitespace-cleanup in which buffer-local
>> indent-tabs-mode is not respected."
>>   (interactive)
>>   (whitespace-turn-on) ;; sets buffer-local variables
>>   (whitespace-turn-off) ;; don't actually want the mode to be on
>>   (whitespace-cleanup))
>>     
>
>   
>> Here are instructions to replicate the problem:
>>     
>
>   
>> Start up with no .emacs.  Set indent-tabs-mode to nil.  Make sure
>> 'indentation is in whitespace-style.  Insert a tab character into a
>> buffer with no spaces after it, e.g. the line could be "\tfoo".  Run
>> whitespace-cleanup in the buffer.  Note that the tab character is still
>> there.  Now run my-whitespace-cleanup.  Note that the tab character has
>> been replaced by spaces.
>>     
>
>   
>> The fact that indent-tabs-mode is buffer-local is important.  For
>> example, makefile-mode sets it to true since tabs are a necessary part
>> of the syntax of make files.
>>     


I already sent back an answer to Craig.

This is neither a bug nor a problem, Craig had implemented a wrong 
function to do what he wanted.

Should I send to you and emacsbugs list all the emails about this 
conversation?


Regards,


Vinicius