bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Jambunathan K]

;; Choose Devanagari
C-x RET l D e v <tab> <return> 

;; toggle input method
C-\ 

Debugger entered--Lisp error: (error "Can't activate input method `dev-aiba'")
  signal(error ("Can't activate input method `dev-aiba'"))
  error("Can't activate input method `%s'" "dev-aiba")
  activate-input-method("dev-aiba")
  toggle-input-method(nil 1)
  call-interactively(toggle-input-method nil nil)


In GNU Emacs 24.0.94.1 (i386-mingw-nt5.1.2600)
 of 2012-03-19 on MARVIN
Windowing system distributor `Microsoft Corp.', version 5.1.2600
Configured using:
 `configure --with-gcc (4.6) --no-opt --enable-checking --cflags
 -ID:/devel/emacs/libs/libXpm-3.5.8/include
 -ID:/devel/emacs/libs/libXpm-3.5.8/src
 -ID:/devel/emacs/libs/libpng-dev_1.4.3-1/include
 -ID:/devel/emacs/libs/zlib-dev_1.2.5-2/include
 -ID:/devel/emacs/libs/giflib-4.1.4-1/include
 -ID:/devel/emacs/libs/jpeg-6b-4/include
 -ID:/devel/emacs/libs/tiff-3.8.2-1/include
 -ID:/devel/emacs/libs/gnutls-3.0.9/include'

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: ENG
  value of $XMODIFIERS: nil
  locale-coding-system: cp1252
  default enable-multibyte-characters: t

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Glenn Morris]

Please try this:

*** lisp/language/indian.el	2012-01-19 07:21:25 +0000
--- lisp/language/indian.el	2012-03-27 20:35:25 +0000
***************
*** 46,52 ****
   "Devanagari" '((charset unicode)
  		(coding-system utf-8)
  		(coding-priority utf-8)
! 		(input-method . "dev-aiba")
  		(documentation . "\
  Such languages using Devanagari script as Hindi and Marathi
  are supported in this language environment."))
--- 46,52 ----
   "Devanagari" '((charset unicode)
  		(coding-system utf-8)
  		(coding-priority utf-8)
! 		(input-method . "devanagari-aiba")
  		(documentation . "\
  Such languages using Devanagari script as Hindi and Marathi
  are supported in this language environment."))

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Glenn Morris]

BTW, though that fix seems obviously correct to me, I won't commit it
unless someone who actually uses Indian input methods tests it.
For all I know, something somewhere else assumes "dev-aiba".
(It's broken all the way back to Emacs 22.1.)

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Jambunathan K]

On 3/30/12, Glenn Morris <rgm@gnu.org> wrote:
>
> BTW, though that fix seems obviously correct to me, I won't commit it
> unless someone who actually uses Indian input methods tests it.
> For all I know, something somewhere else assumes "dev-aiba".
> (It's broken all the way back to Emacs 22.1.)
>

Glenn

(Refer http://debbugs.gnu.org/cgi/bugreport.cgi?bug=11103)

I am copying Rustom who has more understanding of devanagari. He can
suggest the default input method to go with that particular language
environment and also test your earlier patch.

C-h L Devanagari actually gives the following output

Devanagari language environment

Such languages using Devanagari script as Hindi and Marathi
are supported in this language environment.

Input methods:
  devanagari-inscript ("DevIS" in mode line)
  devanagari-aiba ("DevAB" in mode line)
  devanagari-kyoto-harvard ("DevKH" in mode line)
  devanagari-itrans ("DevIT" in mode line)

Character sets:
  unicode: Unicode (ISO10646)

Coding systems:
  utf-8 (`U' in mode line):
	UTF-8 (no signature (BOM))
	(alias: utf-8 mule-utf-8)

[back]

My 2 cents will be that the current defaults of devanagari-aiba need
to be changed to devanagari-itrans. I don't speak the language that
uses devanagari script.

Jambunathan K.

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Rustom Mody]

[-- Attachment #1: Type: text/plain, Size: 1736 bytes --]

Sorry I dont get the context.
On my emacs 23.3.1 I can activate devanagari-aiba input method.
Is there something else you want me to check?
I dont know this input method (and google does not help) but looking at
describe-input-method and trying out a bit it seems very close
(identical??) to itrans.

On Sat, Mar 31, 2012 at 12:17 AM, Jambunathan K <kjambunathan@gmail.com>wrote:

> On 3/30/12, Glenn Morris <rgm@gnu.org> wrote:
> >
> > BTW, though that fix seems obviously correct to me, I won't commit it
> > unless someone who actually uses Indian input methods tests it.
> > For all I know, something somewhere else assumes "dev-aiba".
> > (It's broken all the way back to Emacs 22.1.)
> >
>
> Glenn
>
> (Refer http://debbugs.gnu.org/cgi/bugreport.cgi?bug=11103)
>
> I am copying Rustom who has more understanding of devanagari. He can
> suggest the default input method to go with that particular language
> environment and also test your earlier patch.
>
> C-h L Devanagari actually gives the following output
>
> Devanagari language environment
>
> Such languages using Devanagari script as Hindi and Marathi
> are supported in this language environment.
>
> Input methods:
>  devanagari-inscript ("DevIS" in mode line)
>  devanagari-aiba ("DevAB" in mode line)
>  devanagari-kyoto-harvard ("DevKH" in mode line)
>  devanagari-itrans ("DevIT" in mode line)
>
> Character sets:
>  unicode: Unicode (ISO10646)
>
> Coding systems:
>  utf-8 (`U' in mode line):
>        UTF-8 (no signature (BOM))
>        (alias: utf-8 mule-utf-8)
>
> [back]
>
> My 2 cents will be that the current defaults of devanagari-aiba need
> to be changed to devanagari-itrans. I don't speak the language that
> uses devanagari script.
>
> Jambunathan K.
>

[-- Attachment #2: Type: text/html, Size: 2394 bytes --]

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Jambunathan K]

[-- Attachment #1: Type: text/plain, Size: 3723 bytes --]


Glenn

You can ignore this mail, if you want.

Rustom,

Here is the context -

> Sorry I dont get the context.
> On my emacs 23.3.1 I can activate devanagari-aiba input method.
> Is there something else you want me to check?

Are you activating the input method like this - 

1. C-x C-m l Devanagari
2. C-\

Do you see a crash like this -

Debugger entered--Lisp error: (error "Can't activate input method `dev-aiba'")
  signal(error ("Can't activate input method `dev-aiba'"))
  error("Can't activate input method `%s'" "dev-aiba")
  activate-input-method("dev-aiba")
  toggle-input-method(nil 1)
  call-interactively(toggle-input-method nil nil)

> I dont know this input method (and google does not help) 

See [1]. They apparently has something to do with diacritic marks
(whatever that means).

> but looking at describe-input-method and trying out a bit it seems
> very close (identical??) to itrans.

See [2].  The table that compares two methods and is attached here for
future reference.  There are 12 entries in that table where the aiba and
itrans input methods differ.

I was wondering, as someone who is more familiar with Devanagari which
of these would you prefer as the input method.  If you think that you
have no particular opinion, I think we can leave the defaults as such.

Here is an entry for SHA.

,---- DEVANAGARI LETTER SHA with devanagari-itrans is mapped to "sha"
|              position: 9395 of 15548 (60%), column: 31
|             character: श (displayed as श) (codepoint 2358, #o4466, #x936)
|     preferred charset: unicode (Unicode (ISO10646))
| code point in charset: 0x0936
|                syntax: w 	which means: word
|              category: .:Base, L:Left-to-right (strong), i:Indian
|              to input: type "sha" with devanagari-itrans
|           buffer code: #xE0 #xA4 #xB6
|             file code: #xE0 #xA4 #xB6 (encoded by coding system utf-8-dos)
|               display: by this font (glyph code)
|     uniscribe:-outline-Mangal-normal-normal-normal-*-20-*-*-*-p-*-iso10646-1 (#x9E)
| 
| Character code properties: customize what to show
|   name: DEVANAGARI LETTER SHA
|   general-category: Lo (Letter, Other)
|   decomposition: (2358) ('श')
| 
| [back]
`----

,---- DEVANAGARI LETTER SHA with devanagari-itrans is mapped to "^sa"
|              position: 2855 of 5721 (50%), column: 22
|             character: श (displayed as श) (codepoint 2358, #o4466, #x936)
|     preferred charset: unicode (Unicode (ISO10646))
| code point in charset: 0x0936
|                syntax: w 	which means: word
|              category: .:Base, L:Left-to-right (strong), i:Indian
|              to input: type "^sa" with devanagari-aiba
|           buffer code: #xE0 #xA4 #xB6
|             file code: #xE0 #xA4 #xB6 (encoded by coding system utf-8-dos)
|               display: by this font (glyph code)
|     uniscribe:-outline-Mangal-normal-normal-normal-*-20-*-*-*-p-*-iso10646-1 (#x9E)
| 
| Character code properties: customize what to show
|   name: DEVANAGARI LETTER SHA
|   general-category: Lo (Letter, Other)
|   decomposition: (2358) ('श')
| 
| [back]
`----

Footnotes: 

[1] http://texa.human.is.tohoku.ac.jp/aiba/codes/table/draft/r02/html/

Document has following particulars.

,---- 
| Word Processing in Tibetan and Sanskrit
| Toru AIBA
| 
| Graduate School of Information Sciences, Tohoku University
| March 2000
| 
| Handout for my presentation at "Fourth International Symposium on
| Multilingual Information Processing" at Tsukuba in March 26, 2000.
`----

[2] Image captioned "Table 3: Transliteration Schemes in Sanskrit
(Substitutional)", 


[-- Attachment #2: itrans-aiba-comparison.gif --]
[-- Type: image/gif, Size: 7568 bytes --]

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Rustom Mody]

[-- Attachment #1: Type: text/plain, Size: 4287 bytes --]

On Sat, Mar 31, 2012 at 1:28 PM, Jambunathan K <kjambunathan@gmail.com>wrote:

>
> Glenn
>
> You can ignore this mail, if you want.
>
> Rustom,
>
> Here is the context -
>
> > Sorry I dont get the context.
> > On my emacs 23.3.1 I can activate devanagari-aiba input method.
> > Is there something else you want me to check?
>
> Are you activating the input method like this -
>
> 1. C-x C-m l Devanagari
> 2. C-\
>
> Do you see a crash like this -
>

If I do it like you say I only see
Can't activate input method `dev-aiba'


No backtrace.

If I do what I usually do ie C-x RET C-\
and then give devanagari-aiba (not dev-aiba) it seems to work for me

My preferences later (I'll look more carefully at both)

Rusi



> Debugger entered--Lisp error: (error "Can't activate input method
> `dev-aiba'")
>  signal(error ("Can't activate input method `dev-aiba'"))
>  error("Can't activate input method `%s'" "dev-aiba")
>  activate-input-method("dev-aiba")
>  toggle-input-method(nil 1)
>  call-interactively(toggle-input-method nil nil)
>
> > I dont know this input method (and google does not help)
>
> See [1]. They apparently has something to do with diacritic marks
> (whatever that means).
>
> > but looking at describe-input-method and trying out a bit it seems
> > very close (identical??) to itrans.
>
> See [2].  The table that compares two methods and is attached here for
> future reference.  There are 12 entries in that table where the aiba and
> itrans input methods differ.
>
> I was wondering, as someone who is more familiar with Devanagari which
> of these would you prefer as the input method.  If you think that you
> have no particular opinion, I think we can leave the defaults as such.
>
> Here is an entry for SHA.
>
> ,---- DEVANAGARI LETTER SHA with devanagari-itrans is mapped to "sha"
> |              position: 9395 of 15548 (60%), column: 31
> |             character: श (displayed as श) (codepoint 2358, #o4466, #x936)
> |     preferred charset: unicode (Unicode (ISO10646))
> | code point in charset: 0x0936
> |                syntax: w      which means: word
> |              category: .:Base, L:Left-to-right (strong), i:Indian
> |              to input: type "sha" with devanagari-itrans
> |           buffer code: #xE0 #xA4 #xB6
> |             file code: #xE0 #xA4 #xB6 (encoded by coding system
> utf-8-dos)
> |               display: by this font (glyph code)
> |
> uniscribe:-outline-Mangal-normal-normal-normal-*-20-*-*-*-p-*-iso10646-1
> (#x9E)
> |
> | Character code properties: customize what to show
> |   name: DEVANAGARI LETTER SHA
> |   general-category: Lo (Letter, Other)
> |   decomposition: (2358) ('श')
> |
> | [back]
> `----
>
> ,---- DEVANAGARI LETTER SHA with devanagari-itrans is mapped to "^sa"
> |              position: 2855 of 5721 (50%), column: 22
> |             character: श (displayed as श) (codepoint 2358, #o4466, #x936)
> |     preferred charset: unicode (Unicode (ISO10646))
> | code point in charset: 0x0936
> |                syntax: w      which means: word
> |              category: .:Base, L:Left-to-right (strong), i:Indian
> |              to input: type "^sa" with devanagari-aiba
> |           buffer code: #xE0 #xA4 #xB6
> |             file code: #xE0 #xA4 #xB6 (encoded by coding system
> utf-8-dos)
> |               display: by this font (glyph code)
> |
> uniscribe:-outline-Mangal-normal-normal-normal-*-20-*-*-*-p-*-iso10646-1
> (#x9E)
> |
> | Character code properties: customize what to show
> |   name: DEVANAGARI LETTER SHA
> |   general-category: Lo (Letter, Other)
> |   decomposition: (2358) ('श')
> |
> | [back]
> `----
>
> Footnotes:
>
> [1] http://texa.human.is.tohoku.ac.jp/aiba/codes/table/draft/r02/html/
>
> Document has following particulars.
>
> ,----
> | Word Processing in Tibetan and Sanskrit
> | Toru AIBA
> |
> | Graduate School of Information Sciences, Tohoku University
> | March 2000
> |
> | Handout for my presentation at "Fourth International Symposium on
> | Multilingual Information Processing" at Tsukuba in March 26, 2000.
> `----
>
> [2] Image captioned "Table 3: Transliteration Schemes in Sanskrit
> (Substitutional)",
>
>

[-- Attachment #2: Type: text/html, Size: 5402 bytes --]

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Jambunathan K]

I applied this patch. 

I continue to get the orignal crash. 

Quick look suggests that language-info-alist (right after emacs -Q and
no intervening commands) continues to refer to "dev-aiba".

As a result when I do, C-x C-m l Devanagari, the `default-input-method'
picked up is "dev-aiba".

This results in a crash when I do C-\

Log:
====

(pp-eval-expression (quote language-info-alist))

=>
--8<---------------cut here---------------start------------->8---
 ("Devanagari"
  (documentation . "Such languages using Devanagari script as Hindi and Marathi\nare supported in this language environment.")
  (input-method . "dev-aiba")
  (coding-priority utf-8)
  (coding-system utf-8)
  (charset unicode))
--8<---------------cut here---------------end--------------->8---

(set-language-environment "Devanagari")
(eval-expression (quote default-input-method) nil)

=> "dev-aiba"





> Please try this:
>
> *** lisp/language/indian.el	2012-01-19 07:21:25 +0000
> --- lisp/language/indian.el	2012-03-27 20:35:25 +0000
> ***************
> *** 46,52 ****
>    "Devanagari" '((charset unicode)
>   		(coding-system utf-8)
>   		(coding-priority utf-8)
> ! 		(input-method . "dev-aiba")
>   		(documentation . "\
>   Such languages using Devanagari script as Hindi and Marathi
>   are supported in this language environment."))
> --- 46,52 ----
>    "Devanagari" '((charset unicode)
>   		(coding-system utf-8)
>   		(coding-priority utf-8)
> ! 		(input-method . "devanagari-aiba")
>   		(documentation . "\
>   Such languages using Devanagari script as Hindi and Marathi
>   are supported in this language environment."))

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Eli Zaretskii]

> From: Jambunathan K <kjambunathan@gmail.com>
> Date: Sat, 31 Mar 2012 14:16:32 +0530
> Cc: 11103@debbugs.gnu.org
> 
> 
> I applied this patch. 
> 
> I continue to get the orignal crash. 
> 
> Quick look suggests that language-info-alist (right after emacs -Q and
> no intervening commands) continues to refer to "dev-aiba".
> 
> As a result when I do, C-x C-m l Devanagari, the `default-input-method'
> picked up is "dev-aiba".
> 
> This results in a crash when I do C-\

Please don't call "a crash" something that is merely an error.

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Glenn Morris]

Jambunathan K wrote:

> I continue to get the orignal crash. 

I echo the comments about this not being a crash.

> Quick look suggests that language-info-alist (right after emacs -Q and
> no intervening commands) continues to refer to "dev-aiba".

Mine doesn't. You need to rebuild Emacs to pick up the change. Did you
do that? You need to get leim-list.el regenerated. I don't know the
minimum steps needed for that, so, as always, I suggest bootstrapping.

> As a result when I do, C-x C-m l Devanagari, the `default-input-method'
> picked up is "dev-aiba".

I get devanagari-aiba.

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Jambunathan K]

>> Quick look suggests that language-info-alist (right after emacs -Q and
>> no intervening commands) continues to refer to "dev-aiba".
>
> Mine doesn't. 

Ok.

> You need to rebuild Emacs to pick up the change. Did you do that? 

No. I wasn't aware of it. It's all new to me.

> You need to get leim-list.el regenerated. I don't know the minimum
> steps needed for that, so, as always, I suggest bootstrapping.

I rely only on binaries. I have not built Emacs at all and I don't want
to build one.

>> As a result when I do, C-x C-m l Devanagari, the `default-input-method'
>> picked up is "dev-aiba".
>
> I get devanagari-aiba.

Check your fix in. Once Christopher Scholtes makes available a new
binary build, I will update the status of this bug. Meanwhile Rustom
might have also made his opinion on whether he wants aiba or itrans by
default.

Jambunathan K.

bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'")

[Glenn Morris]

Version: 24.0.95

Fair enough, I checked it in.
If you think there should be actual changes to the input method, I
suggest making a new report with the details. Followup to this report if
there turns out to be some problem with the fix for the input method not
working at all.

bug#11103: closed (Re: bug#11103: 24.0.94; (error "Can't activate input method `dev-aiba'"))

[Jambunathan K]

> Your bug report
> #11103: 24.0.94; (error "Can't activate input method `dev-aiba'")
> which was filed against the emacs package, has been closed.

Fix verified with 
GNU Emacs 24.0.95.1 (i386-mingw-nt5.1.2600) of 2012-04-03 on MARVIN.