Re: [m17n-list] m17n Based Keyboard Layout for Bengali

[Sayak Sarkar]

Hi,

My apologies for the above mix-up. The initial time I posted the mail to the wrong list and it had bounced back, thereafter I wasn't quite sure about which list to post to. Hope This time its the correct one.

Also, sorry for the top posting, I did get a bit too anxious to fully understand the wijesekera layout first hand before implementing a new logic. However, I did a bit of experimentation with the m17n input layouts in I-bus and get a rough idea of what has been done and for the last couple of days I've been going through the unicode charts and had been mapping and classifying the input characters based on their UTF-8 hex codes and what type of characters they are (i.e. consonants, dependent/independent vowels, special, etc.).

Now from what I've done in the last few days, I've kind-of got a basic idea of how the layout system works, but also it has generated quite a lot of queries in mind. I'll list them as follows:-

1. In the wijesekera input system, there are input mappings defined with a combination of A-'x' where x as far I gather is maybe a consonant, similarly there are such combinations of G and S also. However when I tried using the Wijesekera input layout to try out what these combinations do, all that I got was the consonants themselves.This is something that I'm still unable to figure out.
2. While doing the mapping and classification of various Bengali characters using Utf-8 mapping, I couldn't quite find the Hex code for the character Khanda-ta - ৎ, it maybe because it has been newly defined or so, but since the character is already present in the Probhat layout I gather that it must have a Hex code in UTF-8. Could please let me know the Hex code for the same?
3. Similar to the above point there are some Bengali characters like Nukta (Unicode Hex 09BC) and Ishhar (Unicode Hex 09FA) that I cannot find in the present Probhat layout. How should I go about implementing these characters? Or should I just skip them.
4. Also another problem I'm facing is that I'm unable to test whether the logical conditions that I'm writing are properly functioning or not as cannot directly run the mim file. Could anyone please suggest me a way by which I can run the mim file. I tried directly placing the mim file in the Ibus directory along with the other mim files, but it does not work that way.
5. The main reason behind my anxiety being that I want to have a basic running local instance in my machine as soon as possible. I want to be able to check the logical conditions especially during defining states and for writing behaviours for the dependent vowels and special characters. I've have an understanding of the what logic is to implemented and if I'm able to actively test what conditions I'm writing I think I can have at least an implementation of the dependent vowels running in a week.

As far as the time constraints are concerned I want to be have a basic running instance within the 25th of June, so that I can finish work on the rest of the layout logic within June-end.

Currently I've the following mapping set defined which I think might be correct to the best of my knowledge. However the dependent vowels and special character would be needing some logical mapping as well, which would be dependent on the states as well. I'm currently working on this conditional logic behind the states and special characters, but without the ability to run them, I'm quite unsure what I'm doing, hence I haven't included them below:-

;; bn-disha.mim -- Bengali input method with probhat method

(input-method bn disha (version "0.0.1"))

(description "Bengali input method based on the Probhat input layout mapping.")

(title "কি")

(map
 (starter
  ("A") ("B") ("C") ("D") ("E") ("F") ("G") ("H") ("I") ("J") ("K") ("L") ("M")
  ("N") ("O") ("P") ("Q") ("R") ("S") ("T") ("U") ("V") ("W") ("X") ("Y") ("Z") 
  ("a") ("b") ("c") ("d") ("e") ("f") ("g") ("h") ("i") ("j") ("k") ("l") ("m")
  ("n") ("o") ("p") ("q") ("r") ("s") ("t") ("u") ("v") ("w") ("x") ("y") ("z") 
  ("`") ("~") ("\[") ("{") ("\]") ("}") ("\\") ("|") (";") (":") ("'") ("\"")
  (",") ("<") (".") (">")
  ((BackSpace)) ((Delete)))

  (consonant
    ("k","ক")    ; 0995
    ("K","খ")    ; 0996
    ("g","গ")    ; 0997
    ("G","ঘ")    ; 0998
    ("M","ঙ")    ; 0999
    ("c","চ")    ; 099A
    ("C","ছ")    ; 099B
    ("j","জ")    ; 099C
    ("J","ঝ")    ; 099D
    ("&","ঞ")    ; 099E
    ("t","ট")    ; 099F
    ("T","ঠ")    ; 09A0
    ("d","ড")    ; 09A1
    ("D","ঢ")    ; 09A2
    ("N"."ণ")    ; 09A3
    ("f","ত")    ; 09A4
    ("F","থ")    ; 09A5
    ("q","দ")    ; 09A6
    ("Q","ধ")    ; 09A7
    ("n","ন")    ; 09A8
    ("p","প")    ; 09AA
    ("P","ফ")    ; 09AB
    ("b","ব")    ; 09AC
    ("B","ভ")    ; 09AD
    ("m","ম")    ; 09AE
    ("Z","য")    ; 09AF
    ("r","র")    ; 09B0
    ("l","ল")    ; 09B2
    ("x","শ")    ; 09B6
    ("S","ষ")    ; 09B7
    ("s","স")    ; 09B8
    ("h","হ")    ; 09B9
    ("R","ড়")    ; 09BC
    ("X","ঢ়")    ; 09BD
    ("z","য়")    ; 09BF
    ("*","ৎ")
   )

  (special
    ("/", " ্")    ;09CD
    (">","ঁ")    ; 0981
    ("L","ং")    ;0982
    ("H","ঃ")    ;0983
  )

  (independent
    ("A","অ")    ; 0985
    ("v","আ")    ; 0986
    ("I","ই")    ; 0987
    ("E","ঈ")    ; 0988
    ("U","উ")    ; 0989
    ("W","ঊ")    ; 098A
    ("V","ঋ")    ; 098B
    ("y","এ")    ; 098F
    ("Y","ঐ")    ; 0990
    ("o","ও")    ; 0993
    ("O","ঔ")    ; 0994
   )

  (dependent
    ("a","া")    ;09BE
    ("i","ি")    ;09BF
    ("e","ী")    ;09C0
    ("u","ু")    ;09C1
    ("w","ূ")    ;09C2
    ("<"," ৃ")    ;09C3
    ("[","ে")    ;09C7
    ("{"," ৈ")    ;09C8
    ("]","ো")    ;09CB
    ("}","ৌ")    ;09CC
  )
)

I would be very grateful if you could provide me some help with the above, especially on how I could run the mim file to test the conditions.

Awaiting your reply,

Regards,
Sayak

On Mon, Jun 18, 2012 at 3:57 PM, Runa Bhattacharjee <address@hidden> wrote:

Hello Sayak,
Is there some reason why this entire discussion is not on an m17n mailing list,
which imho is a much more appropriate place both for archiving purposes and
eyeballs?

Kind regards
Runa

(apologies for top-posting)

On শনিবার 16 জুন 2012 09:06 অপরাহ্ণ, Sayak Sarkar wrote:

Hi,

I have been going through the si-wijesekera.mim input layout and have been
trying to understand how the various maps and states have been defined in
it. However, there are lots of areas which I am finding a bit difficult to
understand. I do understand the basic key mappings that have been done
through the maps "starter", "cnsonant", "misc", and to a certain extent
about the conditional logic implemented in the mapping of keys like "a", "d",
"E", "A". However in some places like where "macro", "backward", the variable
 "use-surrounding-text", also same with some of the states etc have been
defined I don't quite understand what exactly has been done. This may be a
bit due to my lack of understanding of the Sinhala language but also may be
due to my limited knowledge of the mim implementation.

I would be highly grateful if you could give me some idea about how I could
get a better understanding of these problems. May be some kind of
documentation of the si-wijesekera layout would be helpful for me.

Also, I'm not quite sure how to run a particular mim file that I create, so
that I get know if it is running properly or have I made a mistake
somewhere.

Awaiting your suggestions.

Regards, Sayak

On Sat, Jun 16, 2012 at 2:31 PM, TAKAHASHI Naoto <address@hidden

<mailto:address@hiddengo.jp>> wrote:

On 08/06/2012, Sayak Sarkar <address@hidden

<mailto:address@hiddencom>> wrote:

I'm currently working on small dummy layouts to get a grip of the system
and would be very grateful if anyone could please also give me some ideas
about how character classes; consonants & dependent vowels; syllable
classes syllable systems are implemented.

We may help you if your questions are more concrete.

-- TAKAHASHI Naoto




-- About Me:http://about.me/sayak_sarkar Twitter:
http://twitter.com/sayak_sarkar Blog: http://sayaksarkar.wordpress.com

<http://sayaksarkar.wordpress.com/>

--
blog: http://arrbee.wordpress.com
irc: arrbee or runa_b on Freenode
http://fedoraproject.org/wiki/User:Runab

-- 
About Me:http://about.me/sayak_sarkar
Twitter: http://twitter.com/sayak_sarkar

Blog: http://sayaksarkar.wordpress.com