[1.42.03 release]

Allow .txt release for *_rules and *_list files.
Rules lookup, consider multi-byte characters when calculating matchign points.
Fix bug in Roman Number translation.


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@232 d46cf337-b52f-0410-862d-fd96e6ae7743

dictsource/af_list

 antwerpen   antv&rp@n
 babilon     babilOn
 bangladesj  baNglad'ES
+barcelona   bars@l'o@na
 barkly      _^_EN
 bellevue    b&lvju:
 bermuda     $2
 bethlehem   bEt_liEm
 boedapest   budapEst
 boekarest   bukarEst
+bogotá      bOgOt'A:
 bolivië     bul'Ivi:@
 bonnievale  _^_EN
 bordeaux    bOrd'@U
+boston      _^_EN
 brittanje   $2
 brooklyn    _^_EN
+(buenos aires) bwEnOs_'aIrEs
 burundi     bur'undi
 caledon     kalidOn
 carolina    karul'ina
 constantia  $2
 christiana  kr@stij'A:na
 damaskus    d@maskWs
+denemarke   de@n@mark@
 djakarta    dZak'arta
 doebaai     $2
 duitswes    dYytsv'Es
 franschhoek frA~nsh'uk
 gadara      x2@dA:ra
 galilea     x2al@l'i:a
+genève      Z@nE:v
 germiston   _^_EN
 getsémané   x2Ets'e@mani
 gonubie     g@nu:bi
 israel      IsraI:l
 italië      it'A:li;@
 jeffreysbaai dZEfrisb'AI
+junín       x2un'in
 kalafong    kalaf'ON
 kanada      kanad%a
 kanaän      kA:nA:an
 europa      Y@r'o@pa
 nigel       _^_EN
 nineve      nIn@fi
+oklahoma    @Ukl@h'@Uma
 outeniekwa  @Ut@n'ikwa
 oxford      _^_EN
+palermo     pal'&rmu
 pelindaba   p&l@nd'A:ba
+perú        p@ru
+piketberg   p@kEtb&rx2
 portugal    pOrtyx2al
 potchefstroom pOtSIfstr'o@m
+rhône       _^_FR
 richardsbaai ritS@dsb'AI
 robertson   _^_EN
 sahara      sah'A:ra
+salvador    salvadO:r
+santiago    santi;'A:gu
 secunda     s@kunda
 senekal     se@n@kal
 seychelle   seIS'&l@
 wellington  wEliNt@n
 wolseley    wulsli
 uganda      ug'anda
+uruguay     urugwaI
 zimbabwe    zI:mb'ab_wE
 zulu        zulu // Kwazulu-Natal
 
 André       'andreI
 annette     an'Et
 antoinette  antun'Et
+archimedes  arx2im'e@dEs
 aristoteles arist'o@t@lEs
+arthur      _^_EN
 badenhorst  bA:d@nhOrst
 barbara     bA:b@ra
 basson      $2
 bruwer      bruw'&:r
 caesar      siz@r
 carl        kA:r@L
+carlyle     _^_EN
 carol       _^_EN
 casanova    kasan'o@va
 cecil       sEs@l
 deventer d'e@v@nt@r
 diana       dij'A:na
 donaldson   _^_EN
+dostojewski dOstOj'Evski
 du		$u
+duncan      daNk@-n
 (du pisanie) du||pis'A:ni
 (du plessis) du||pl@s'i
 (du plooy)   d@||ploI
 ferreira    f@r&:r@
 filemon     f'il@mOn
 fischer     fiS@r
+flaubert    _^_FR
 floris      $1
 fouché      fuS'e@
 fourie      fur'i
 gertruida   $2
 gilbert     _^_EN
 gloria      glO:ria
+gordon      _^_EN
 grange      x2rA:nsi
 grové       x2ruv'e@
+gulliver      _^_EN
 gustav      gustaf
 guy         _^_EN
 hamlet      _^_EN
 herklaas    $1
 herkules    $1
 hilde       hild@
+ibsen       ibs@n
 ilse 'ilz@
 ingrid      iNgr@d
 isabel      is@b&l
 joris       $1
 josef       jo@s@f
 josephina   jo@s@f'ina
+josé          _^_ES
+juan        x2wan
 juanita     juan'ita
 julia       _^_EN
 	khayyam     kaj'am
 liesbet     lisbEt
 louis       luI
 louise      luw'i:z
+lukas       lukas
 mackintosh  _^_EN
 madelene	mad@l'e@n
 mbeki       mbE:ki
 mathilda    mat'Ilda
 matteus     mat'i:Ws
 matthee     mat'e@
+mattheüs    mat'i:Ws
 matthéüs    mat'i:Ws
 maureen     _^_EN
 mefistofeles mEfist'o@f@l@s
 petrus      pe@trWs
 phoebe      fi:bi
 picasso     pik'asu
-piketberg	pik'Etb&rx2
 pinokkio    pin'Oki;u
 pretorius   pr@to@riWs
+pythagoras  pit'A:x2o@ras
+quijote     kix2'OtE
 quixote     kix2'OtE
 rachmaninoff rax2m'aninOf
 rademeyer   rA:d@meI@r
 sophie      _^_EN
 stephan     ste@fan
 stephanus   st@fA:nWs
+stevenson   _^_EN
 strauss     straUs
 stockenström stOk@nstro@m
 suzanne     suz'A:n
 trudie      trudi
 versfeld    f&rsf&lt
 veronica    _^_EN
+victor      _^_EN
 victoria    _^_EN
 viljoen     $2
 villiers    vIli@rs
 porsche     pOrS@
 port        pOrt $1
 rover       _^_EN
+sorbonne    _^_FR
 sotho       su:tu
 toyota      $2
 venda       vEnda
 eks           Eks
 ekself %&ks'&lf
 elders        &ld@rs
+elementale    El@mEnt'A:l@
 elite         il'i:t
 encore        A~NkO:r
 enige         e@nIx2@
 eone          $2
 era           e@ra
 erger         &rg@r
+ergering      &rg@r@N
 espeak        i:;spi:k
 (et cetera)   Ets'Et@ra
 euro          Y@ru
 genade        $2
 genesis       x2e@n@s@s 
 //gesamentlik   x2@sA:m@ntl@k
+gentleman     _^_EN
 gewellys      x2e@v@lleIs
 gladiator     x2ladi'A:tOr
 globale       x2lo@b'A:l@
 inagneming    @nax2ne@m@N
 indien        Indin $verb
 induna        ind'u:na
+infame        @nfA:m@
 ingevolge     $3
 inkatha       iNk'A:ta
 inkluis       $2
 londense      lOnd@ns@
 lord          _^_EN
 luidop        lYyt_Op
+lykwaak       leIkvA:k
 
 macaroni      makar'o@ni
 madame        mad'A:m
 omrede        $2
 omtrent       Omtr'Ent
 onbegrip      $1
+onbruik       $1
 onderrig      On@r'Ix2 $verb
 ondersoek     $3 $verb
 onderweg      $3
 sektore       $2
 sinagoge      s@n@x2o@x2@
 senatore      s@nat'o@r@
+siesta        si;'Esta
 singenot      sInx2@nOt
 sir           _^_EN
 skepsele      skEps@l@
 verafgeleë    f&rafx2@le@@
 verby         f@rbeI
 verdere       f&rd@r@
-vererger      f@r&rg@r
 ver           f'&:r
 vereduifie    fe@r@dYyfi
 verg          f&rx2

dictsource/dict_phonemes

 *    :    ;    b    C    C2   d    D    
 dZ   f    g    h    j    k    l    m    
 n    N    p    pF   r    s    S    t    
-tS   ts   v    x    z    Z    
+tS   ts   v    w    x    z    Z    
 
 
 Dictionary el_dict
 b    c    d    f    g    h    j    k    
 l    m    n    N    p    r    s    t    
 tS   v    w    z    
+
+
+Dictionary ur_dict
+
+@    a    a:   aI   aU   E    e:   i    
+I    i:   O    o:   U    u:   
+
+:    b    bh   c    ch   d    D    d.   
+dh   dh.  dZ   f    g    gh   H    j    
+J    Jh   k    kh   l    m    n    N    
+p    ph   Q    q    R    r.   s    S    
+s.   t    T    t.   th   th.  v    x    
+z    Z    z.   

dictsource/en_list

 antihero          $1
 antilog           $1
 antimatter        $1
-antinomy         ant'In@mI
+antimony          $2
 antioch           $1
 antipathy        ant'Ip@TI
 antiphonal       ant'If@n@L
 jasmine          dZazmIn
 jesus            dZi:z@s
 joky             dZoUkI
+jour             ZU@  // french
 jukebox          dZu:kb0ks
 july             dZu:l'aI
 
 Dominic		d0mI2nIk
 Donaghy		d0n'hi:
 Doris		d0rIs
+Dorothy		dO@r@Ti
 Dougall		du:g@L
 Dylan		dIl@n
 Edith		i:dIT
 Elizabeth	I2lIz@b@T
+Emily		EmI2lI
 Emma		Em@
 Emmy		EmI
 Etheridge	ET@-rIdZ

dictsource/en_rules

          r) ouge         u:Z
          _) ough         O:
             ough (t      O:
-            ough (*      OI
             ough         oU
          b) ough         aU
          b) ough (t      O:
          _) re (emP2     r%i:
          _) re (enP2     r%i:
          _) re (eqP2     r%i:
+         _) re (esP2     r%i:
          _) re (exP2     r%i:
             referend     rEf@r'End
             re (fug      rE
 ?5      ss) wor (d       wVR
             wor (th_     w3:
 ?5          wor (th_     wVR
-         s) w (ick
+        As) w (ick
 
 .group x
          _) x (C         Eks
          _) y (n@        V
          _) y (s@        V
          C) y (abl       I2
-            y (y
             y (C         I
          C) y            %I
             y (ar        j

dictsource/hu_list

 velünk	$unstressend
 veletek	$unstressend
 velük	$unstressend
+közt	$unstressend
+hiszik	$unstressend
+hiszitek	$unstressend
+önt	$unstressend
+benn	$unstressend
+kinn	$unstressend
+kés	$unstressend
+néz	$unstressend
+vont	$unstressend
+fiút	$unstressend
+von	$unstressend
+nézett	$unstressend
+rám	$unstressend
+rád	$unstressend
+mód	$unstressend
+győz	$unstressend
+repült	$unstressend
+ül	$unstressend
+gyújt	$unstressend
+hajt	$unstressend
+köp	$unstressend
+látni	$unstressend
+sült	$unstressend
+roncs	$unstressend
+is	$unstressend
+fut	$unstressend
+ló	$unstressend
+véljük	$unstressend
+vélik	$unstressend
+vélem	$unstressend
+véled	$unstressend
+véli	$unstressend
+vélitek	$unstressend
 
 // word pairs
 

dictsource/hu_rules

 
         biz (A     biz,      // bizalmas  etc.
         biedermeier bi:dER2ma:jER2
-_) batthiány bAc:a:n^i
+     _) batthiány  bAc:a:n^i
 
 .group c
         c          ts
   rövi) d (zárlat  d
  sarja) d (zik     d
  nyála) d (z       d
-ren) dsz (erint d|s
+   ren) dsz (erint d|s
+ szaba) d (szájú   d
 
 .group dz
   fogó) dz (kod    ts
      e) gy (sínű   J
      e) gy (sor    J
     né) gy (sor    J
+     e) gy (es-egyedül J     //need one J phoneme with this word, not J: phoneme
 
 .group h
         h          h
 .group í
         í          i:
         íts        i:tS:
-ítsd i:tSd
-  
+        ítsd       i:tSd
+
      t) í (zen     i
      t) í (zet     i
      v) í (zen     i
   pero) n (jegy    n
      ö) n (járó    n
 tulajdo) n (jog    n
-ko) nn (ektor n //connector word, we spokening one n letter, not two n letter
-
-
+    ko) nn (ektor  n         //connector word, we spokening one n letter, not two n letter
+    me) nj (_S2    n^
+    be) nn (szülött n
+   éle) n (jár     n
+  magá) n (jelleg  n
 
 .group o
         o          o
      _) richa (rd  R2iCa:
      _) richá (rd  R2iCa:
         reichsmarschall R2EjsmAR2SAl:
-?!2  _) roose (velt R2u:z@
+?!2  _) roose (velt R2u:z
 
 
 .group s
     vi) ssz (hang  s
 felelő) ss (ég     S
 égzengé) ssz (erű  Ss
+világo) ssz (ürke  Ss
+motoro) ssz (emüveg Ss
+  bőgé) ssz (erű   Ss
+kavará) ssz (erű   Ss
+   _sa) ssz (eg    Ss
+
 
 .group sz
         sz         s
  épüle) t (sarok   t
   mene) t (sor     t
    skó) t (sapka   t
-lé) t (jogosultsága t
-bérle) t (jegy t
-csapa) t (játék t
-ese) t (jog t
-fejeze) t (jel t
-ké) t (jegyű t
-krike) tt (játék t:
-mene) t (jegy t
-szüne) t (jel t
-ú) t (jelző t
-ú) t (jog t
-gondola) t (jel t
-helyze) t (jel t
-
+    lé) t (jogosultsága t
+ bérle) t (jegy    t
+ csapa) t (játék   t
+   ese) t (jog     t
+fejeze) t (jel     t
+    ké) t (jegyű   t
+ krike) tt (játék  t:
+  mene) t (jegy    t
+ szüne) t (jel     t
+     ú) t (jelző   t
+     ú) t (jog     t
+gondola) t (jel    t
+helyze) t (jel     t
+folyama) t (sáv    t
+   ada) t (szolgáltat t
+    _ú) t (sáv     t
+négyze) t (centi   t
 
 .group ts
         ts (_S2    tS        //general rule with word end of ts letters
     já) tsz        ts:
     já) tssz       ts:
     já) tsz (m     ts
+    já) tsz (hat   ts
         tsz (_S3   ts:
         tsz (é     ts:
         tsz (o     ts:
 
 .group ú
         ú          u:
+     h) ú (szan    u
 
 .group ü
         ü          y
 
 .group ű
         ű          y:
+     t) ű (zijáték y
 
 .group v
         v          v
   nehé) z (súly    z
    raj) z (szeg    z
    raj) z (szög    z
-pén) z (t s
-rende) z (ked s
-kocká) z (tat s
-hibá) z (tat s
-különbö) z (tet s
-válto) z (tat s
-válto) z (tass s
-tájéko) z (tat s
-tájéko) z (tass s 
-
+   pén) z (t       s
+ rende) z (ked     s
+ kocká) z (tat     s
+  hibá) z (tat     s
+különbö) z (tet    s
+ válto) z (tat     s
+ válto) z (tass    s
+tájéko) z (tat     s
+tájéko) z (tass    s
+    mé) z (szín    z
+    bi) z (tat     s
+    rá) z (kód     s
 
 
 .group

dictsource/ta_list

 போனஸ்	$alt
 போகி	$alt
 பாபா 	ba:ba:
-புத்த	buddV
+//புத்த	buddV
 
 
 	// initial த as [d]
 கட	$alt
 கோ	$alt
 கதி	$alt
+கோபி 	$alt
+கச்சாமி	$alt
+குரு	$alt
 
 
 பப	pVpV

dictsource/ta_rules

 	க்க (B	kk
      ற்) க	_kV
      ற்) க (B	_k
+     ட்) க	kV
+     ட்) க (B	k
+	க (ிரு	k
 
 	க (ீத		g
 	க (ஜ		gV
      _) க (ாந்த	g
      _) க (ுண்ட	g
      _) க (ோபுர	g
-     _) க (ுரு		g
      _) க (ுகன		g
      _) க (ணபதி	gV
      _) க (ணீர		gV
      _) க (தியில்       gV
      _) க (டிகார       gV
     வி) க்ட (ோரியா	kt.	// victoria
-	க (ோப		g
 	க (ாரன		k
 	க (ாரர		k
-     _) க (ோபம	k
      _) கோப (ித்து	ko:b
      _) க (ோபப்		k
      _) க (ெட்டி	g
+     _) க (வண		gV
+     _) க (ோகுல	g
+        கோப (ால	go:p
+     _) க (ோபிய	g
+     _) க (ோபிநா	g
+     _) க (ோபிசெட்	g
+     _) க (ூடலூர	g
+     _) க (ுடியாத்த	g
+     _) க (வன		gV
+     _) க (ுருவின	g
+     _) க (ுருவிட	g
+     _) க (ுருக்கள	g
+     _) க (ுருவோட	g
+     _) க (ுருவுடை	g
+     _) க (ுருரா	g
+     _) க (ுருநா	g
+     _) க (ுருவாய	g
 
 
 .group ங
 
      _) ச (ிட்	tS
      _) ச (ின்	tS
+     _) ச (ேரி	tS
 
      _) ச (ப்பாத்தி	tSV
 
+     _) சிரு (ஷ்டி	sri
+	ச (ாந்தி		s.
+	ச (ண்மு	s.V
 
 .group ஜ
 	ஜ	dZV
      _)	த (ுஷ்		d
      _) த (க்ஷிணா	dV
      _) த (ண்ட		dV
+     _) த (ண்டவாள	tV
      _) த (ருமி		dV
      _) த (ர்பை		dV
      _) த (ாவா		d
      _) த (ினுச	d
      _) த (ாராவி	d
      _) த (ர்க		dV
-     _) த (ான		d
+//     _) த (ான		d
      _) த (ேக		d
      _) த (ேக்கு	t
      _) த (ேக்கிலை	t
      _) த (ன்ராஜ	dV
      _) த (ாராபுர	d
      _) த (டி_மாட	dV
-
+     _) த (ேவன	d
+     _) த (ேவி		d
+     _) த (ேவர்க	d
+     _) த (ேவதை	d
+     _) த (ேவாம்ச	d
+     _) த (ிவ்ய		d
+     _) த (ிண்டிவன	d
+     _) த (ுறை		d
+     _) த (ிண்டுக்கல	d
+     _) த (மயந்தி	dV
+     _) த (ுரோண	d
+     _) த (ுரியோதனன	d
+     _) த (ுச்சாதனன	d
+     _) த (ிரௌபதி	d
+
+     யு) த்த		ddV
+     யு) த்த(B		dd
      பு) த்த (ி		dd
      பு) த்த (ியி	dd
      பு) த்த (ர		ddV
      பு) த்த (ன		ddV
 
+  _சாந்) த (ி		t
+   _சும) த (ி		t
+ _வசந்) த (ி		t
+ _சுகந்) த (ி		t
+ _வனி) த (ா		t
+  _சுனி) த (ா		t
+    _ல) த (ா		t
+பிரேமல) த (ா		t
+  _ரேவ) த (ி		t
+ _வினி) த (ா		t
+ _வசந்) த		tV
+ _பார்வ) த (ி		t
+பிரபாவ) த (ி		t
+_கோம) த (ி		t
+   _பிர) த (ீ		t
+
+     _) திரு (ஷ்டி	dri
+     _) திரு (திரா	dri
+     _) துர (ோண	tr
+     _) திர (ௌபதி	tr
+
 
 .group ந
 	ந	nV
     த) ப (ா	p
     த) ப (B	b
 
-	பத (ி	pVt
+     @) பத (ி	pVt
 	ப (ுர	p
         ப (ால	p
         ப (ாள	p
     ம்) ப (ா	b
 
 
-     _) ப (ால	b
-     _) ப (ால்	p
+//     _) ப (ால		b
+     _) ப (ால்		p
      _) ப (ாலின்		p
      _) ப (ாலை		p
-     _) ப (ாலில்		p
+//     _) ப (ாலில்		p
      _) ப (ாலுடன் 	p
-     _) ப (ாலினுடைய 	p
+//     _) ப (ாலினுடைய 	p
 
      _) ப (ாலாபிஷேக	p
-     _) ப (ாலு	b
-     _) ப (ானு	b
-     _) ப (ீம	b
+     _) ப (ாலு		b
+     _) ப (ானு		b
+     _) ப (ீம		b
      _) ப (காசுர	bV
-     _) ப (ோஜ	b
-     _) ப (தில	bV
-     _) ப (ஜ	bV
+     _) ப (ோஜ		b
+     _) ப (தில		bV
+     _) ப (ஜ		bV
      _) ப (ைரவ		b
-     _) ப (யம	bV
+     _) ப (யம		bV
      _) ப (ார்கவி	b
      _) ப (ூதேவி	b
      _) ப (ம்பாய		bV
      _) ப (ாதுஷா	b
-     _) ப (க்த	bV
-     _) ப (லம	bV
+     _) ப (க்த		bV
+     _) ப (லம		bV
      _) ப (லவான	bV
      _) ப (ிராமண	b
      _) ப (ிரம்ம		b
-     _) ப (ம்பர	bV
-     _) ப (லூன	bV
-     _) ப (ூமி	b
+     _) ப (ம்பர		bV
+     _) ப (லூன		bV
+     _) ப (ூமி		b
      _) ப (ூலோக	b
      _) ப (ூகோல	b
-     _) ப (ூத	b
+     _) ப (ூத		b
      _) ப (ுதன		b
      _) ப (லி		bV
      _) ப (கவான	bV
      _) ப (ோபால	b
      _) ப (ெங்களூர	b
      _) ப (ூபாள	b
-     _) ப (ுத்த		b
+     _) புத்த(K		buddV
+     _) புத் (தJ_புத+	put
      _) ப (னியன	bV
   கோ) ப (ி		p
+     _) ப (ாலச		b
+     _) ப (ாலக		b
+     _) ப (ாலமு		b
+     _) ப (ாலதண்	b
+     _) ப (ூவுலக	b
+     _) ப (ிரகஸ்பதி	b
+     _) ப (ிரகதீ		b
+     _) ப (ொம்மை	b
+     _) ப (ொம்மல	b
+     _) ப (ாக்கி		b
+     _) ப (ஸ		bV
+     _) ப (ாஷை		b
+     _) ப (ிரம்மை	b
+     _) ப (ாரம		b
+     _) ப (ங்களா	bV
+     _) ப (ங்கம		bV
+     _) ப (ுட்டி		b
+
+     _) பிர (ம்மன	brV
+     _) பிர (ாமணன	br
+     _) பிர (கஸ்பதி	brV
+     _) பிர (கதீ		brV
+     _) பிர (கலாதன	prV
+     _) பிர (காஷ	prV
+     _) பிர (காச	prV
+     _) பிர (சாத	prV
+     _) பிர (ே		pr
+     _) பிரு (கன்நளை	bri
+     _) பிர (சாந்தி	prV
+     _) பிர (ம்மை	brV
+     _) பிர (சன்ன	prV
+     _) பிர (ச்சனை	prV
+     _) பிர (ப		prV
+     _) பிர (தீ		prV
 
 
 	// musical notes

platforms/windows/windows_dll/src/speak_lib.h

 /* This is the header file for the library version of espeak */
 /*                                                           */
 /*************************************************************/
+#define ESPEAK_API __declspec(dllexport)
 
 #include <stdio.h>
 
 #ifdef __cplusplus
 extern "C"
 #endif
-int espeak_Initialize(espeak_AUDIO_OUTPUT output, int buflength, const char *path, int options);
+ESPEAK_API int espeak_Initialize(espeak_AUDIO_OUTPUT output, int buflength, const char *path, int options);
 /* Must be called before any synthesis functions are called.
    output: the audio data can either be played by eSpeak or passed back by the SynthCallback function.
 
 #ifdef __cplusplus
 extern "C"
 #endif
-void espeak_SetSynthCallback(t_espeak_callback* SynthCallback);
+ESPEAK_API void espeak_SetSynthCallback(t_espeak_callback* SynthCallback);
 /* Must be called before any synthesis functions are called.
    This specifies a function in the calling program which is called when a buffer of
    speech sound data has been produced. 
 #ifdef __cplusplus
 extern "C"
 #endif
-void espeak_SetUriCallback(int (*UriCallback)(int, const char*, const char*));
+ESPEAK_API void espeak_SetUriCallback(int (*UriCallback)(int, const char*, const char*));
 /* This function may be called before synthesis functions are used, in order to deal with
    <audio> tags.  It specifies a callback function which is called when an <audio> element is
    encountered and allows the calling program to indicate whether the sound file which
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_Synth(const void *text,
+ESPEAK_API espeak_ERROR espeak_Synth(const void *text,
 	size_t size,
 	unsigned int position,
 	espeak_POSITION_TYPE position_type,
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_Synth_Mark(const void *text,
+ESPEAK_API espeak_ERROR espeak_Synth_Mark(const void *text,
 	size_t size,
 	const char *index_mark,
 	unsigned int end_position,
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_Key(const char *key_name);
+ESPEAK_API espeak_ERROR espeak_Key(const char *key_name);
 /* Speak the name of a keyboard key.
    If key_name is a single character, it speaks the name of the character.
    Otherwise, it speaks key_name as a text string.
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_Char(wchar_t character);
+ESPEAK_API espeak_ERROR espeak_Char(wchar_t character);
 /* Speak the name of the given character 
 
    Return: EE_OK: operation achieved 
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_SetParameter(espeak_PARAMETER parameter, int value, int relative);
+ESPEAK_API espeak_ERROR espeak_SetParameter(espeak_PARAMETER parameter, int value, int relative);
 /* Sets the value of the specified parameter.
    relative=0   Sets the absolute value of the parameter.
    relative=1   Sets a relative value of the parameter.
 #ifdef __cplusplus
 extern "C"
 #endif
-int espeak_GetParameter(espeak_PARAMETER parameter, int current);
+ESPEAK_API int espeak_GetParameter(espeak_PARAMETER parameter, int current);
 /* current=0  Returns the default value of the specified parameter.
    current=1  Returns the current value of the specified parameter, as set by SetParameter()
 */
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_SetPunctuationList(const wchar_t *punctlist);
+ESPEAK_API espeak_ERROR espeak_SetPunctuationList(const wchar_t *punctlist);
 /* Specified a list of punctuation characters whose names are to be spoken when the
    value of the Punctuation parameter is set to "some".
 
 #ifdef __cplusplus
 extern "C"
 #endif
-void espeak_SetPhonemeTrace(int value, FILE *stream);
+ESPEAK_API void espeak_SetPhonemeTrace(int value, FILE *stream);
 /* Controls the output of phoneme symbols for the text
    value=0  No phoneme output (default)
    value=1  Output the translated phoneme symbols for the text
 #ifdef __cplusplus
 extern "C"
 #endif
-void espeak_CompileDictionary(const char *path, FILE *log, int flags);
+ESPEAK_API void espeak_CompileDictionary(const char *path, FILE *log, int flags);
 /* Compile pronunciation dictionary for a language which corresponds to the currently
    selected voice.  The required voice should be selected before calling this function.
 
 #ifdef __cplusplus
 extern "C"
 #endif
-const espeak_VOICE **espeak_ListVoices(espeak_VOICE *voice_spec);
+ESPEAK_API const espeak_VOICE **espeak_ListVoices(espeak_VOICE *voice_spec);
 /* Reads the voice files from espeak-data/voices and creates an array of espeak_VOICE pointers.
    The list is terminated by a NULL pointer
 
    If voice_spec is NULL then all voices are listed.
-   If voice spec is given, then only the voices which are compatible with the voice_spec
+   If voice spec is give, then only the voices which are compatible with the voice_spec
    are listed, and they are listed in preference order.
 */
 
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_SetVoiceByName(const char *name);
+ESPEAK_API espeak_ERROR espeak_SetVoiceByName(const char *name);
 /* Searches for a voice with a matching "name" field.  Language is not considered.
    "name" is a UTF8 string.
 
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_SetVoiceByProperties(espeak_VOICE *voice_spec);
+ESPEAK_API espeak_ERROR espeak_SetVoiceByProperties(espeak_VOICE *voice_spec);
 /* An espeak_VOICE structure is used to pass criteria to select a voice.  Any of the following
    fields may be set:
 
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_VOICE *espeak_GetCurrentVoice(void);
+ESPEAK_API espeak_VOICE *espeak_GetCurrentVoice(void);
 /* Returns the espeak_VOICE data for the currently selected voice.
    This is not affected by temporary voice changes caused by SSML elements such as <voice> and <s>
 */
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_Cancel(void);
+ESPEAK_API espeak_ERROR espeak_Cancel(void);
 /* Stop immediately synthesis and audio output of the current text. When this
    function returns, the audio output is fully stopped and the synthesizer is ready to
    synthesize a new message.
 #ifdef __cplusplus
 extern "C"
 #endif
-int espeak_IsPlaying(void);
+ESPEAK_API int espeak_IsPlaying(void);
 /* Returns 1 if audio is played, 0 otherwise.
 */
 
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_Synchronize(void);
+ESPEAK_API espeak_ERROR espeak_Synchronize(void);
 /* This function returns when all data have been spoken.
    Return: EE_OK: operation achieved 
 	   EE_INTERNAL_ERROR.
 #ifdef __cplusplus
 extern "C"
 #endif
-espeak_ERROR espeak_Terminate(void);
+ESPEAK_API espeak_ERROR espeak_Terminate(void);
 /* last function to be called.
    Return: EE_OK: operation achieved 
 	   EE_INTERNAL_ERROR.
 #ifdef __cplusplus
 extern "C"
 #endif
-const char *espeak_Info(void* ptr);
+ESPEAK_API const char *espeak_Info(void* ptr);
 /* Returns the version number string.
    The parameter is for future use, and should be set to NULL
 */

src/compiledata.cpp

 	wxString filename;
 	wxFileName fname;
 	wxString dictstr;
+	wxString filetype;
 	wxString report = _T("");
 	int err;
 	int errors = 0;
 		fprintf(f_phused,"Phonemes which are used in the *_rules and *_list files\n");
 	}
 
-	bool cont = dir.GetFirst(&filename, _T("*_rules"), wxDIR_FILES);
+	bool cont = dir.GetFirst(&filename, _T("*_rules*"), wxDIR_FILES);
 	while ( cont )
 	{
 		fname = wxFileName(filename);
+		filetype = fname.GetName().AfterLast('_');
+		if((filetype != _T("rules")) && (filetype != _T("rules.txt")))
+		{
+			cont = dir.GetNext(&filename);
+			continue;
+		}
+
 		dictstr = fname.GetName().BeforeLast('_');
 		strcpy(dictname,dictstr.mb_str(wxConvLocal));
+
 		dict_count++;
 		strcpy(voicename,dictname);
 
 		// read the *_rules file to see if a phoneme table is specified though a voice name
-		sprintf(path,"%s%s_rules",path_dsource,dictname);
-		if((f_in = fopen(path,"r")) != NULL)
+		sprintf(path,"%s%s_rules.txt",path_dsource,dictname);
+		if((f_in = fopen(path,"r")) == NULL)
+		{
+			sprintf(path,"%s%s_rules",path_dsource,dictname);
+			f_in = fopen(path,"r");
+		}
+
+		if(f_in != NULL)
 		{
 			unsigned int ix;
 			unsigned int c;

src/compiledict.cpp

 	
 	text_mode = 0;
 
-	sprintf(fname,"%s%s",path,filename);
+	// try with and without '.txt' extension
+	sprintf(fname,"%s%s.txt",path,filename);
 	if((f_in = fopen(fname,"r")) == NULL)
-		return(-1);
+	{
+		sprintf(fname,"%s%s",path,filename);
+		if((f_in = fopen(fname,"r")) == NULL)
+			return(-1);
+	}
 
 	fprintf(f_log,"Compiling: '%s'\n",fname);
 
 				case 'X':
 					c = RULE_NOVOWELS;
 					break;
+				case 'J':
+					c = RULE_SKIPCHARS;
+					break;
 				case 'L':
 					// expect two digits
 					c = *p++ - '0';
 	if(f_log == NULL)
 		f_log = stderr;
 
+	// try with and without '.txt' extension
 	sprintf(path,"%s%s_",dsource,dict_name);
-	sprintf(fname_in,"%srules",path);
-	f_in = fopen_log(fname_in,"r");
-	if(f_in == NULL)
+	sprintf(fname_in,"%srules.txt",path);
+	if((f_in = fopen(fname_in,"r")) == NULL)
 	{
-		if(fname_err)
-			strcpy(fname_err,fname_in);
-		return(-1);
+		sprintf(fname_in,"%srules",path);
+		if((f_in = fopen_log(fname_in,"r")) == NULL)
+		{
+			if(fname_err)
+				strcpy(fname_err,fname_in);
+			return(-1);
+		}
 	}
 
 	sprintf(fname_out,"%s%c%s_dict",path_home,PATHSEP,dict_name);

src/dictionary.cpp

 	static char output[60];
 
 	static char symbols[] = {' ',' ',' ',' ',' ',' ',' ',' ',' ',
-			'@','&','%','+','#','S','D','Z','A','L',' ',' ',' ',' ',' ','N','K','V',' ','T','X','?','W'};
+			'@','&','%','+','#','S','D','Z','A','L','!',' ','?','?','J','N','K','V','?','T','X','?','W'};
 
 	static char symbols_lg[] = {'A','B','C','H','F','G','Y'};
 
 
 				if((letter == rb) || ((letter==(unsigned char)REPLACED_E) && (rb=='e')))
 				{
-					add_points = 21;
+					if((letter & 0xc0) != 0x80)
+						add_points = 21;    // don't add point for non-initial UTF-8 bytes
 					consumed++;
 				}
 				else
 					}
 					break;
 
+				case RULE_SKIPCHARS:
+					{
+						// Used for lang=Tamil, used to match on the next word after an unknown word ending
+						// only look until the end of the word (including the end-of-word marker)
+						// Jx  means 'skip characters until x', where 'x' may be '_' for 'end of word'
+						char *p = post_ptr + letter_xbytes;
+						char *p2 = p;
+						int rule_w;            // skip characters until this
+						utf8_in(&rule_w,rule);
+						while((letter_w != rule_w) && (letter_w != RULE_SPACE))
+						{
+							p2 = p;
+							p += utf8_in(&letter_w,p);
+						}
+						if(letter_w == rule_w)
+						{
+							post_ptr = p2;
+						}
+					}
+					break;
+
 				case RULE_INC_SCORE:
 					add_points = 20;      // force an increase in points
 					break;
 				default:
 					if(letter == rb)
 					{
-						if(letter == RULE_SPACE)
-							add_points = (21-distance_right);
-						else
+						if((letter & 0xc0) != 0x80)
+						{
+							// not for non-initial UTF-8 bytes
 							add_points = (21-distance_right);
+						}
 					}
 					else
 						failed = 1;
 						if(letter == RULE_SPACE)
 							add_points = 4;
 						else
-							add_points = (21-distance_left);
+						{
+							if((letter & 0xc0) != 0x80)
+							{
+								// not for non-initial UTF-8 bytes
+								add_points = (21-distance_left);
+							}
+						}
 					}
 					else
 						failed = 1;

src/extras.cpp

 	const int tab2 = 18;
 	const int tab3 = 28;
 
-	sprintf(fname_in,"%s/%s_rules",path_dsource,dictname);
-	sprintf(fname_out,"%s_formatted",fname_in);
-
+	// try with and without '.txt' extension
+	sprintf(fname_in,"%s/%s_rules.txt",path_dsource,dictname);
 	if((f_in = fopen(fname_in,"r")) == NULL)
 	{
-		wxLogError(_T("Can't open rules file: ") + wxString(fname_in,wxConvLocal));
-		return;
+		sprintf(fname_in,"%s/%s_rules",path_dsource,dictname);
+		if((f_in = fopen(fname_in,"r")) == NULL)
+		{
+			wxLogError(_T("Can't open rules file: ") + wxString(fname_in,wxConvLocal));
+			return;
+		}
 	}
+
+	sprintf(fname_out,"%s/%s_rules_formatted",path_dsource,dictname);
 	if((f_out = fopen(fname_out,"w")) == NULL)
 	{
 		wxLogError(_T("Can't write to file: ") + wxString(fname_out,wxConvLocal));

src/numbers.cpp

 	prev = 0;
 	subtract = 0x7fff;
 	ph_out[0] = 0;
+	flags[0] = 0;
+	flags[1] = 0;
 
 	if((tr->langopts.numbers & NUM_ROMAN_CAPITALS) && !(wtab[0].flags & FLAG_ALL_UPPER))
 		return(0);

src/readclause.cpp

   0x055e,  // Armenian question
   0x055b,  // Armenian emphasis mark
 
+  0x0b1b,  // Arabic ;
+  0x061f,  // Arabic ?
+
   0x1362,  // Ethiopic period
   0x1363,
   0x1364,
   CLAUSE_QUESTION + PUNCT_IN_WORD,  // Armenian question
   CLAUSE_PERIOD + PUNCT_IN_WORD,  // Armenian emphasis mark
 
+  CLAUSE_SEMICOLON,  // Arabic ;
+  CLAUSE_QUESTION,   // Arabic question mark
+
   CLAUSE_PERIOD,     // Ethiopic period
   CLAUSE_COMMA,      // Ethiopic comma
   CLAUSE_SEMICOLON,  // Ethiopic semicolon
 	char phonemes2[55];
 	char *string1 = (char *)string;
 
+	flags[0] = flags[1] = 0;
 	if(LookupDictList(tr,&string1,phonemes,flags,0,NULL))
 	{
 		SetWordStress(tr, phonemes, flags, -1, 0);
 				// Because of an xml tag, we are waiting for the
 				// next non-blank character to decide whether to end the clause
 				// i.e. is dot followed by an upper-case letter?
-				if(c1 == '\n')
-				{
-//					end_clause_after_tag &= ~CLAUSE_DOT;
-				}
 				
 				if(!iswspace(c1))
 				{
-					if(iswdigit(c1) || (IsAlpha(c1) && !iswlower(c1)))
+					if(!IsAlpha(c1) || !iswlower(c1))
+//					if(iswdigit(c1) || (IsAlpha(c1) && !iswlower(c1)))
 					{
 						UngetC(c2);
 						ungot_char2 = c1;

src/synthdata.cpp

 #include "translate.h"
 #include "wave.h"
 
-const char *version_string = "1.42.02  30.Nov.09";
+const char *version_string = "1.42.03  08.Dec.09";
 const int version_phdata  = 0x014200;
 
 int option_device_number = -1;

src/tr_languages.cpp

 		break;
 
 	case L('a','r'):   // Arabic
-	case L('u','r'):   // Urdu
 			tr->letter_bits_offset = OFFSET_ARABIC;
 			tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1;   // disable check for unpronouncable words
 		break;
 			tr->langopts.stress_rule = 6;      // stress on last heaviest syllable, excluding final syllable
 			tr->langopts.stress_flags =  0x10004;   // use 'diminished' for unstressed final syllable
 			tr->langopts.numbers = NUM_SWAP_TENS;
-			tr->langopts.break_numbers = 0x24924aa8;  // for languages which have numbers for 100,000 and 100,00,000, eg Hindi
+			tr->langopts.break_numbers = 0x14aa8;  // for languages which have numbers for 100,000 and 100,00,000, eg Hindi
 			tr->letter_bits_offset = OFFSET_DEVANAGARI;
 
 			if(name2 == L('p','a'))
 		}
 		break;
 
+	case L('u','r'):   // Urdu
+			tr->letter_bits_offset = OFFSET_ARABIC;
+			tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1;   // disable check for unpronouncable words
+			tr->langopts.numbers = NUM_SWAP_TENS;
+			tr->langopts.break_numbers = 0x52a8;  // for languages which have numbers for 100,000 and 100,00,000, eg Hindi
+		break;
+
 	case L('v','i'):  // Vietnamese
 		{
 			static const short stress_lengths_vi[8] = {150, 150,  180, 180,  210, 230,  230, 240};

src/translate.h

 #define RULE_LINENUM		8  // next 2 bytes give a line number, for debugging purposes
 
 #define RULE_SPACE		32   // ascii space
-#define RULE_SYLLABLE	9
-#define RULE_STRESSED	10
-#define RULE_DOUBLE		11
-#define RULE_INC_SCORE	12
-#define RULE_DEL_FWD		13
-#define RULE_ENDING		14
+#define RULE_SYLLABLE	9    // @
+#define RULE_STRESSED	10   // &
+#define RULE_DOUBLE		11   // %
+#define RULE_INC_SCORE	12   // +
+#define RULE_DEL_FWD		13   // #
+#define RULE_ENDING		14   // S
 #define RULE_DIGIT		15   // D digit
 #define RULE_NONALPHA	16   // Z non-alpha
 #define RULE_LETTERGP   17   // A B C H F G Y   letter group number
 #define RULE_LETTERGP2  18   // L + letter group number
-#define RULE_CAPITAL    19   //    word starts with a capital letter
+#define RULE_CAPITAL    19   // !   word starts with a capital letter
 #define RULE_REPLACEMENTS 20  // section for character replacements
+#define RULE_SKIPCHARS  23   // J
 #define RULE_NO_SUFFIX  24   // N
 #define RULE_NOTVOWEL   25   // K
 #define RULE_IFVERB     26   // V