<?xml version="1.0"?>
<!DOCTYPE alphabets SYSTEM "alphabet.dtd">
<?xml-stylesheet type="text/xsl" href="alphabet-nest.xsl"?>
<alphabets>
<alphabet name="Bengali / &#x09ac;&#x09be;&#x0982;&#x09b2;&#x09be;   with punctuation and numerals">
<!--
 Bangla is the second most commonly spoken language in India.
 The same script is used by Assamese.
1.  Problem when using this alphabet? 
  In the phrase "world peace" a CONJUNCT (called Yuktakhars / "juktakhars" in Assamese?)
  glyph is required that consists of two consonants
  SHA and BA, combined with one above the other. It was not clear whether this alphabet can achieve this combination.
  I think this webpage
  http://www.ekushey.org/projects/shadhinota/technical/phonetic_bangla_typing.htm
  might contain the answer.  - use a combining Virama? - I tried it and it
  didn't work.
   See also http://www.maruf42.com/archive/ct/unicode.pdf
   According to another webpage,
    'there are no specific codes for conjunct consonants. Unicode Bengali fonts and Unicode text editors will theoretically handle the display of conjunct consonants.'
    http://tlt.psu.edu/suggestions/international/bylanguage/bengalichart.html
    So this is perhaps just a defect of the textbox.
    The Unicode Standard book, page 214, seems to agree.
    There is no unicode prescription to specify a conjunct. It should happen
    automatically, with virama's being used sometimes. ZWNJ is used
    to prevent conjunctions.
2.  The rule that one should enter unicode characters in phonetic sequence (CV,CV,...)
  may prove confusing to users because in glyph-writing
  the sequence is often VC,VC,...
  We can refer to the above Phonetic Typing page to instruct users.
3. For future reference, there is a free list of words in Bangla.
  Dr. Avijit Das ( adrab@users.sf.net ) has a bangla word list that he
 set up for his bengaliwriter program (
 http://sourceforge.net/projects/bengaliwriter/ )
 The list is released under the GNU GPL. The list has around 112,900 words
 (ghose) converted the original list in ISCII format (with a hack to handle
khondo-ta) to unicode. The word list is broken up into different files
depending on the first letter of the words contained. This unicode list
can be obtained at (
http://prdownloads.sourceforge.net/lekho/dict_v1.15.tgz?download )
  -->
<orientation type="LR"/>
<encoding type="Western"/>
<palette>European/Asian</palette>
<train>training_bengali_BD.txt</train>
<space d="&#x25a1;" t=" " b="9" note="box" />
<paragraph d="&#182;" b="9"/>
<control d="Control" t="" b="8"/>
<group name="bengali letters" b="0">
<group name="bengali vowels" b="0">
<s d="&#x0985;" t="&#x0985;"  note=" LETTER A  "  />
<s d="&#x0986;" t="&#x0986;"  note=" LETTER AA  "  />
<s d="&#x0987;" t="&#x0987;"  note=" LETTER I  "  />
<s d="&#x0988;" t="&#x0988;"  note=" LETTER II  "  />
<s d="&#x0989;" t="&#x0989;"  note=" LETTER U  "  />
<s d="&#x098A;" t="&#x098A;"  note=" LETTER UU  "  />
<s d="&#x098B;" t="&#x098B;"  note=" LETTER VOCALIC R  "  />
<s d="&#x09E0;" t="&#x09E0;"  note=" LETTER VOCALIC RR  "  />
<s d="&#x098C;" t="&#x098C;"  note=" LETTER VOCALIC L  "  />
<s d="&#x09E1;" t="&#x09E1;"  note=" LETTER VOCALIC LL  "  />
<s d="&#x098F;" t="&#x098F;"  note=" LETTER E  "  />
<s d="&#x0990;" t="&#x0990;"  note=" LETTER AI  "  />
<s d="&#x0993;" t="&#x0993;"  note=" LETTER O  "  />
<s d="&#x0994;" t="&#x0994;"  note=" LETTER AU  "  />
</group>
<group name="bengali consonants" b="0">
<s d="&#x0995;" t="&#x0995;"  note=" LETTER KA  "  />
<s d="&#x0996;" t="&#x0996;"  note=" LETTER KHA  "  />
<s d="&#x0997;" t="&#x0997;"  note=" LETTER GA  "  />
<s d="&#x0998;" t="&#x0998;"  note=" LETTER GHA  "  />
<s d="&#x0999;" t="&#x0999;"  note=" LETTER NGA  "  />
<s d="&#x099A;" t="&#x099A;"  note=" LETTER CA  "  />
<s d="&#x099B;" t="&#x099B;"  note=" LETTER CHA  "  />
<s d="&#x099C;" t="&#x099C;"  note=" LETTER JA  "  />
<s d="&#x099D;" t="&#x099D;"  note=" LETTER JHA  "  />
<s d="&#x099E;" t="&#x099E;"  note=" LETTER NYA  "  />
<s d="&#x099F;" t="&#x099F;"  note=" LETTER TTA  "  />
<s d="&#x09A0;" t="&#x09A0;"  note=" LETTER TTHA  "  />
<s d="&#x09A1;" t="&#x09A1;"  note=" LETTER DDA  "  />
<s d="&#x09DC;" t="&#x09DC;"  note=" LETTER RRA  class(s)"  />
<s d="&#x09A2;" t="&#x09A2;"  note=" LETTER DDHA  "  />
<s d="&#x09DD;" t="&#x09DD;"  note=" LETTER RHA  class(s)"  />
<s d="&#x09A3;" t="&#x09A3;"  note=" LETTER NNA  "  />
<s d="&#x09A4;" t="&#x09A4;"  note=" LETTER TA  "  />
<s d="&#x09A5;" t="&#x09A5;"  note=" LETTER THA  "  />
<s d="&#x09A6;" t="&#x09A6;"  note=" LETTER DA  "  />
<s d="&#x09A7;" t="&#x09A7;"  note=" LETTER DHA  "  />
<s d="&#x09A8;" t="&#x09A8;"  note=" LETTER NA  "  />
<s d="&#x09AA;" t="&#x09AA;"  note=" LETTER PA  "  />
<s d="&#x09AB;" t="&#x09AB;"  note=" LETTER PHA  "  />
<s d="&#x09AC;" t="&#x09AC;"  note=" LETTER BA  "  />
<s d="&#x09AD;" t="&#x09AD;"  note=" LETTER BHA  "  />
<s d="&#x09AE;" t="&#x09AE;"  note=" LETTER MA  "  />
<s d="&#x09AF;" t="&#x09AF;"  note=" LETTER YA  "  />
<s d="&#x09DF;" t="&#x09DF;"  note=" LETTER YYA  class(s)"  />
<s d="&#x09B0;" t="&#x09B0;"  note=" LETTER RA  "  />
<s d="&#x09F0;" t="&#x09F0;"  note=" LETTER RA WITH MIDDLE DIAGONAL  "  />
<s d="&#x09B2;" t="&#x09B2;"  note=" LETTER LA  "  />
<s d="&#x09F1;" t="&#x09F1;"  note=" LETTER RA WITH LOWER DIAGONAL  "  />
<s d="&#x09B6;" t="&#x09B6;"  note=" LETTER SHA  "  />
<s d="&#x09B7;" t="&#x09B7;"  note=" LETTER SSA  "  />
<s d="&#x09B8;" t="&#x09B8;"  note=" LETTER SA  "  />
<s d="&#x09B9;" t="&#x09B9;"  note=" LETTER HA  "  />
<s d="&#x09BD;" t="&#x09BD;"  note=" SIGN AVAGRAHA  "  />
</group>			    
</group>			    
<group name="Bengali combining vowels" b="52" >	    
<s d="&#x25cc;&#x09BE;" t="&#x09BE;"  note=" VOWEL SIGN AA  "  />
<s d="&#x25cc;&#x09BF;" t="&#x09BF;"  note=" VOWEL SIGN I  "  />
<s d="&#x25cc;&#x09C0;" t="&#x09C0;"  note=" VOWEL SIGN II  "  />
<s d="&#x25cc;&#x09C1;" t="&#x09C1;"  note=" VOWEL SIGN U  "  />
<s d="&#x25cc;&#x09C2;" t="&#x09C2;"  note=" VOWEL SIGN UU  "  />
<s d="&#x25cc;&#x09C3;" t="&#x09C3;"  note=" VOWEL SIGN VOCALIC R  "  />
<s d="&#x25cc;&#x09C4;" t="&#x09C4;"  note=" VOWEL SIGN VOCALIC RR  "  />
<s d="&#x25cc;&#x09E2;" t="&#x09E2;"  note=" VOWEL SIGN VOCALIC L  "  />
<s d="&#x25cc;&#x09E3;" t="&#x09E3;"  note=" VOWEL SIGN VOCALIC LL  "  />
<s d="&#x25cc;&#x09C7;" t="&#x09C7;"  note=" VOWEL SIGN E  "  />
<s d="&#x25cc;&#x09C8;" t="&#x09C8;"  note=" VOWEL SIGN AI  "  />
<s d="&#x25cc;&#x09CB;" t="&#x09CB;"  note=" VOWEL SIGN O  "  />
<s d="&#x25cc;&#x09CC;" t="&#x09CC;"  note=" VOWEL SIGN AU  "  />
<s d="&#x25cc;&#x09CD;" t="&#x09CD;"  note=" SIGN VIRAMA  "  />
<s d="&#x25cc;&#x09D7;" t="&#x09D7;"  note=" AU LENGTH MARK  "  />
<!-- possibly missing characters:  x0981, x0982, x0983  (3 bengali signs) -->
</group>
<group name="Bengali numbers" b="113">
<s d="&#x09E7;" t="&#x09E7;"  note="BENGALI DIGIT ONE"/>
<s d="&#x09E8;" t="&#x09E8;"  note="BENGALI DIGIT TWO"/>
<s d="&#x09E9;" t="&#x09E9;"  note="BENGALI DIGIT THREE"/>
<s d="&#x09EA;" t="&#x09EA;"  note="BENGALI DIGIT FOUR"/>
<s d="&#x09EB;" t="&#x09EB;"  note="BENGALI DIGIT FIVE"/>
<s d="&#x09EC;" t="&#x09EC;"  note="BENGALI DIGIT SIX"/>
<s d="&#x09ED;" t="&#x09ED;"  note="BENGALI DIGIT SEVEN"/>
<s d="&#x09EE;" t="&#x09EE;"  note="BENGALI DIGIT EIGHT"/>
<s d="&#x09EF;" t="&#x09EF;"  note="BENGALI DIGIT NINE"/>
<s d="&#x09E6;" t="&#x09E6;"  note="BENGALI DIGIT ZERO"/>
</group>
<group name="Numbers" b="114">
<s d="1" t="1"/>
<s d="2" t="2"/>
<s d="3" t="3"/>
<s d="4" t="4"/>
<s d="5" t="5"/>
<s d="6" t="6"/>
<s d="7" t="7"/>
<s d="8" t="8"/>
<s d="9" t="9"/>
<s d="0" t="0"/>
</group>
<group name="Punctuation" b="112">
<group name="Bengali Punctuation" b="113">
<s d="&#x09F2;" t="&#x09F2;" note="RUPEE MARK"/>
<s d="&#x09F3;" t="&#x09F3;" note="RUPEE SIGN"/>
<s d="&#x09F4;" t="&#x09F4;" note="CURRENCY NUMERATOR ONE	"/>
<s d="&#x09F5;" t="&#x09F5;" note="CURRENCY NUMERATOR TWO	"/>
<s d="&#x09F6;" t="&#x09F6;" note="CURRENCY NUMERATOR THREE"/>
<s d="&#x09F7;" t="&#x09F7;" note="CURRENCY NUMERATOR FOUR	"/>
<s d="&#x09F8;" t="&#x09F8;" note="CURRENCY NUMERATOR ONE LESS THAN THE DENOMINATOR"/>
<s d="&#x09F9;" t="&#x09F9;" note="CURRENCY DENOMINATOR SIXTEEN"/>
<s d="&#x09FA;" t="&#x09FA;" note="ISSHAR"/>
</group>
<s d="%" t="%"/>
<s d="*" t="*"/>
<s d="+" t="+"/>
<s d="=" t="="/>
<s d="/" t="/"/>
<s d="#" t="#"/>
<s d="$" t="$"/>
<s d="|" t="|"/>
<s d="\" t="\"/>
<s d="~" t="~"/>
<s d="^" t="^"/>
<s d="_" t="_"/>
<s d="&amp;" t="&amp;"/>
<s d="@" t="@"/>
<s d="[" t="["/>
<s d="]" t="]"/>
<s d="{" t="{"/>
<s d="}" t="}"/>
<s d="&lt;" t="&lt;"/>
<s d="&gt;" t="&gt;"/>
<s d="(" t="("/>
<s d=")" t=")"/>
<s d="&quot;" t="&quot;"/>
<s d="`" t="`"/>
<s d="&apos;" t="&apos;"/>
<s b="9"  d="&#x25CA;" t="&#x9;" note="TAB (represented by a diamond)"/>
<s d="-" t="-"/>
<s d=":" t=":"/>
<s d=";" t=";"/>
<s d="?" t="?"/>
<s d="!" t="!"/>
<s d="," t=","/>
<s d="." t="."/>
</group>
<group name="Joiners" b="111">
<s b="73" t="&#x200c;" d="&#x2225;" note="Zero width non-joiner"/>
<s b="74" t="&#x200d;" d="&#x2195;" note="Zero width joiner"/>
</group>
</alphabet>
</alphabets>

