Version 8 of Easy input of Romanised Sanskrit

Updated 2016-01-03 13:16:24 by scud

WJG (8th June, 2005) I recently found details of a better way to input romanised Sanskrit and Pali using a unicode compliant font. So, here's the revised version of this code. Rather than using unique codes, it creates compounds by stacking elements. So the long semi-vowel R would be input as "r <Alt>. <Alt>-".

 # -------------------------------------------------------------
 # skt_input_0.1b.tcl
 # Written by William J Giddings, 2005
 # -------------------------------------------------------------
 # Usage:
 # skt_input <widget>
 # 
 # Purpose:
 # --------
 # To provide text editing application with an easy access to the
 # specialist characters required in romanising the Indic languages.
 #
 # User operation:
 # ---------------
 # Use combination of "Letter + Alt-Key" to obtain letters with diacritics.
 #
 # Notes:
 # ------
 # This version now uses the character stacking extensions used in Gentium.
 #
 # References:
 # ----------
 # Obtain suitable fonts from:
 # http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=Gentium
 # 
 # Extra information and resources including the excellent BabelMap Unicode viewer:
 # http://www.babelstone.co.uk
 # 
 # -------------------------------------------------------------

 # Reminders
 # What input method is most convenient?

 # -------------------------------------------------------------
 # setup custome bindings to handle input of Sanksrit
 # -------------------------------------------------------------
 proc {skt_input} {txt} {

        bind $txt <Alt-Key-.> {
                puts "%k %K"
        if {[regexp [%W get insert-1char insert] {S s T t D d R r L l M m N n H h} ]} {
            %W insert insert \u0323
                }
    }

        bind $txt <Alt-Key-~> {
        if {[regexp [%W get insert-1char insert] {N n} ]} {
            %W insert insert \u0303
                }
    }

    # it's a lot easier just to press the hash key
        bind $txt <Alt-Key-#> {
        if {[regexp [%W get insert-1char insert] {N n} ]} {
            %W insert insert \u0303
                }
    }


        bind $txt <Alt-Key-'> {
        if {[regexp [%W get insert-1char insert] {S s} ]} {
            %W insert insert \u0301 
        }
    }

        bind $txt <Alt-Key--> {
        if {[regexp [%W get insert-1char insert] {A a I i O o U u} ]} {
            %W insert insert \u0304
            return
                }       
        if {[regexp [%W get insert-2char insert-1char] {l L r R} ]} {
            %W insert insert \u0304
                }
    }
 }

 # -------------------------------------------------------------
 # test the code
 # -------------------------------------------------------------
 proc demo  {} {
    package require Tk
    text .txt -font {{Gentium} 14}
    pack .txt -fill both -expand 1 
    skt_input .txt
 }

 demo

WJG (04/02/06) Just added a couple of fixes. There was no equivalent for visarga (ITRANS H) and its always a fiddle to press the shift key to get the tilde. So, added a binding for the hash to place a tilde over n (ITRANS ~N).


scud - 2016-01-03 13:16:24

Scud here. I feel your font is too small, especially for older people. I couldnt really read you home page, and with two sets of reading glasses. I've so far found nothing about Romanised Sanskrit input but I'll look again.