arabxetex update (4mar10)

git-svn-id: svn://tug.org/texlive/trunk@17320 c570f23f-e606-0410-a88d-b1316a301751

1 files changed, 476 insertions, 0 deletions

diff --git a/Master/texmf-dist/fonts/misc/xetex/fontmapping/arabxetex/arabtex-fdf2alif-maghribi-voc.map b/Master/texmf-dist/fonts/misc/xetex/fontmapping/arabxetex/arabtex-fdf2alif-maghribi-voc.map
new file mode 100644
index 00000000000..c85d129e046
--- /dev/null
+++ b/Master/texmf-dist/fonts/misc/xetex/fontmapping/arabxetex/arabtex-fdf2alif-maghribi-voc.map
@@ -0,0 +1,476 @@
+LHSName  "ArabTeX-maghribi-vocalized"
+RHSName  "UNICODE"
+
+Define a U+0061  ;a
+Define b U+0062  ;b
+Define c U+0063  ;c
+Define d U+0064  ;d
+Define e U+0065  ;e
+Define f U+0066  ;f
+Define g U+0067  ;g
+Define h U+0068  ;h
+Define i U+0069  ;i
+Define j U+006A  ;j
+Define k U+006B  ;k
+Define l U+006C  ;l
+Define m U+006D  ;m
+Define n U+006E  ;n
+Define o U+006F  ;o
+Define p U+0070  ;p
+Define q U+0071  ;q
+Define r U+0072  ;r
+Define s U+0073  ;s
+Define t U+0074  ;t
+Define u U+0075  ;u
+Define v U+0076  ;v
+Define w U+0077  ;w
+Define x U+0078  ;x
+Define y U+0079  ;y
+Define z U+007A  ;z
+Define A U+0041  ;A
+Define B U+0042  ;B
+Define E U+0045  ;E
+Define G U+0047  ;G
+Define I U+0049  ;I
+Define N U+004E  ;N
+Define O U+004F  ;O
+Define T U+0054  ;T
+Define U U+0055  ;U
+Define W U+0057  ;W
+Define Y U+0059  ;Y
+Define C U+005E  ;^  << CARET
+Define UL U+005F ;_  << UNDERLINE 
+Define P U+002E  ;.  << DOT
+Define V U+002C  ;,  << COMMA (_V_irgule)
+Define CO U+003A ;:  << COLON
+Define SC U+003B ;; << SEMICOLON
+Define Ain U+0060 ;` 
+Define H U+0027  ;'  << HAMZA 
+Define Q U+0022  ;"  << QUOTE
+Define Z U+002D  ;-  << HYPHEN
+Define BAR U+007C ;|
+Define AN  U+0061 U+004E ; aN
+Define LLAH U+006C U+006C U+005F U+0061 U+0068 
+;Define NUL U+00A6;
+Define Ha  U+0623
+Define Hi  U+0625
+Define Hy  U+0626
+Define Hw  U+0624
+Define Hh  U+06C0 ;; = U+0647 U+0654  
+Define HA  U+0622
+Define HB  U+0640 U+0654
+Define HBAR U+0621 ; '| 
+
+pass(Unicode)
+
+U+E000 > ; eliminate special code for capitalization in transliterations
+
+pass(Unicode)
+
+Class [VWL]         = (U+0061  U+0065  U+0069  U+006F  U+0075 ) ;(a e i o u) 
+Class [AVWL]    = (U+064B..U+0650);
+Class [LTR]	    = (U+0061..U+007A U+0041 U+0045 U+0047 U+0049 U+004F U+0055 U+0042 U+004E U+0054 U+0059 U+005E U+002E U+005F U+0022 U+0027 U+0060) ;a..z A E I O U B N T Y ^ . _ " "" `
+Class [BRACKETS]    = (U+003C U+003E U+00AB U+00BB U+0028 U+0029 U+005B U+005D U+2018 U+2019 U+201C U+201D); < > () « » [ ] ‘ ’ “ ”
+Class [PUNCT] = (V SC P CO)
+Class [Iy]	    = (U+0049 U+0079) ;(I y)
+Class [Uw]          = (U+0055 U+0077) ;(U w)
+Class [Uwu]          = (U+0055 U+0077 U+0075) ;(U w u)
+;Class [ae]	    = (U+0061 U+0065) ;(a e)
+;Class [ou]          = (U+0075 U+006F) ;(u o)
+Class [Iyi]	    = (U+0069 U+0079 U+0049) 
+Class [UI]          = (U+0049 U+0055) ; U I
+Class [Digits] = (U+0030..U+0039)
+
+Define BEG  (#|Z|[BRACKETS])
+Define END  (#|[BRACKETS]|[PUNCT] U+0020|[PUNCT] #)
+Define ENDZ (#|[BRACKETS]|[PUNCT] U+0020|[PUNCT] #|Z)
+
+; special ligatures
+;; 1. Standard ArabTeX input
+;; WARNING! there are, alas, two classes of Arabic fonts:
+;; a) Those that are Unicode-compliant and include the initial alif in glyph U+FDF2 
+;;    For these the ligature llah can be accessed by the sequence lam-lam-ha
+;; b) Those that do not have the initial alif in U+FDF2
+;; The BAR is used in the DMG transliteration to suppress the initial A
+(a|A) BAR? l Z LLAH > U+FDF2 ;; 'al-ll_ah' or 'Al-ll_ah'
+LLAH > U+0644 U+0644 U+0647  ;; 'll_ah' -> 'l l h'
+;; UPPERCASE mnemonics for various ligatures or special glyphs
+;; 2a. custom ArabXeTeX input for ligatures not defined in Unicode
+U+004C U+004C U+0048 > U+0644 U+0644 U+0647  ;; 'LLH' -> 'l l h'
+U+0046 U+0041 U+004C U+0049 U+004C U+004C U+0041 U+0048 > U+0641 U+064E U+0644 U+0650 U+0644 U+0647 ;; 'FALILLAH' : ligature in Adobe Arabic
+;; 2b. Ligatures and glyphs defined in Unicode:
+U+0041 U+004C U+004C U+0041 U+0048 > U+FDF2 ; 'ALLAH'
+U+004C U+004C U+0041 U+0048 > U+0644 U+0644 U+0647 ; 'LLAH'
+
+U+0053 U+004C U+004D > U+FDFA ; 'SLM' : .sallY al-ll_ah `alayhi wa-sallam
+U+0028 U+0028 > U+FD3F ;; '((' : ornate right par
+U+0029 U+0029 > U+FD3E ;; '))' : ornate left par
+U+0052 U+0049 U+0059 U+0041 U+004C > U+FDFC ;; 'RIYAL' : Saudi currency sign
+;; most fonts won't have the following ones:
+U+0053 U+0041 U+004C U+004C U+0041 U+0053 U+0054 U+004F U+0050 > U+FDF0 ;; 'SALLASTOP' (.sallY with "ye-barree", Quranic stop sign)
+U+0051 U+0041 U+004C U+0041 > U+FDF1 ;; 'QALA' : qalY with ye-barree, Quranic stop sign
+U+0041 U+004B U+0042 U+0041 U+0052 > U+FDF3 ;; 'AKBAR'
+U+004D U+0055 U+0048 U+0041 U+004D U+004D U+0041 U+0044 > U+FDF4 ;; 'MUHAMMAD'
+U+0053 U+0041 U+004C U+0041 U+004D > U+FDF5 ;; 'SALAM'
+U+0052 U+0041 U+0053 U+004F ? U+0055 U+004C > U+FDF6 ;; 'RASUL' (also unicode 'RASOUL')
+U+0041 U+004C U+0041 U+0059 U+0048 (U+0049|U+0045) > U+FDF7 ;; 'ALAYHI' (also unicode 'ALAYHE')
+U+0057 U+0041 U+0053 U+0041 U+004C U+004C U+0041 U+004D > U+FDF8 ;; 'WASALLAM' 
+U+0053 U+0041 U+004C U+004C U+0041 > U+FDF9 ;; 'SALLA'
+U+004A U+0041 U+004C U+004C U+0041 > U+FDFB ;; 'JALLA' : "jalla jalAluhu"
+U+0042 U+0041 U+0053 U+004D U+0041 U+004C U+0041 > U+FDFD ;; 'BASMALA' : very few fonts have this ligature afaik
+
+U+005E U+0053 U+004C U+0059 > U+06D6 ;; ^SLY
+U+005E U+0051 U+004C U+0059 > U+06D7 ;; ^QLY
+U+005E U+004D U+0049 U+004D > U+06D8 ;; ^MIM-
+U+005E U+004C U+0041 > U+06D9 ;; ^LA
+U+005E U+004A U+0049 U+004D > U+06DA ;; ^JIM
+U+005E U+0044 U+004F U+0054 U+0053 > U+06DB ;; ^DOTS
+U+005E U+0053 U+0049 U+004E > U+06DC ;; ^SIN
+U+0048 U+0049 U+005A U+0042 > U+06DE ;; HIZB
+U+0043 U+0049 U+0052 U+0043 U+005A U+0045 U+0052 U+004F > U+06DF ;; CIRCZERO
+U+0052 U+0045 U+0043 U+0054 U+005A U+0045 U+0052 U+004F > U+06E0 ;; RECTZERO
+U+005E U+004A U+0041 U+005A U+004D > U+06E1 ;; ^JAZM
+U+005E U+004D U+0049 U+004D > U+06E2 ;; ^MIM
+U+005F U+0053 U+0049 U+004E > U+06E3 ;; _SIN
+U+005E U+004D U+0041 U+0044 U+0044 U+0041 > U+06E4 ;; ^MADDA
+U+0057 U+0041 U+0057 > U+06E5 ;; WAW
+U+0059 U+0045 U+0048 > U+06E6 ;; YEH
+U+005E U+0059 U+0045 U+0048 > U+06E7 ;; ^YEH
+U+005E U+004E U+0055 U+004E > U+06E8 ;; ^NUN
+U+0053 U+0041 U+004A U+0044 U+0041 > U+06E9 ;; SAJDA
+U+005F U+0053 U+0054 U+004F U+0050 > U+06EA ;; _STOP
+U+005E U+0053 U+0054 U+004F U+0050 > U+06EB ;; ^STOP
+U+005E U+0052 U+0053 U+0054 U+004F U+0050 > U+06EC ;; ^RSTOP
+U+005F U+004D U+0049 U+004D > U+06ED  ;; _MIM
+
+U+0044 U+004F U+0054 U+0053 > U+061E ;; DOTS  (also possible by typing "::")
+
+;; This is to enclose digits within glyph U+06DD, see http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=ArabicFonts
+U+005B U+005B ([Digits]+)=dig U+005D U+005D > U+202E U+06DD @dig U+202C ; [[digits]]
+
+
+; vowels
+u u > U 
+i i > I 
+a a > A
+CO O > U+06FC ;
+a e > a y
+a o > a w
+CO U > U+06C7 ;
+e   > i ;; but not pashto > U+0659 and kashmiri > y + U+0658     (U+06CE is kurdish)
+o   > u 
+E > I 
+O  > U 
+
+
+
+;; initial characters ...
+a l Z / BEG _ > U+0627 l Z
+H l Z / BEG _ > U+0627 l Z
+;; wasla on initial alif : NEW! NOT TESTED!
+H / BEG _ l l > U+0671 ; e.g. alla_dI -> 'lla_dI 
+[VWL]=v1 / BEG _ > U+0627 @v1
+A / BEG _ > U+0627 a
+Q?=q1 U / BEG _ > U+0627 @q1 u w ; -U
+; Q?=q1 I / BEG _ > U+0627 @q1 i y ; -I
+Q?=q1 I / BEG _ > U+0627 @q1 i y
+
+
+;; Words with anomalous orthography:
+
+; mi'aT -> ماية  
+;;; i H / m Q ? _ Q ? a ? (T|t) > i A Hy 
+m Q i H Q a / _ (T|t)   > m U+0627 Hy 
+m Q i H a   / _ (T|t)   > m U+0627 Hy a
+m i H Q a   / _ (T|t)   > m i U+0627 Hy 
+m i H a     / _ (T|t)   > m i U+0627 Hy a
+
+
+;; contextual analysis of hamzas
+;; THESE DO NOT APPLY FOR SINDHI
+
+; initial hamzas 
+H  / (#|^[LTR]) _ Q? [Iyi] >   Hi 
+H Q A / (#|^[LTR]) _       >   HA ; alif madda
+H  / (#|^[LTR]) _ Q? [Uwu] >   Ha 
+H  / (#|^[LTR]) _ Q? a     >   Ha  
+Q H / (#|^[LTR]) _         >   HB ; quoted isolated hamza at beginning of word
+
+; quoted hamzas in contextual mode, equivalent to verbatim mode
+a Q H >   Ha  
+i Q H >   Hi 
+y Q H >   Hy 
+w Q H >   Hw 
+h Q H >   Hh 
+A Q H >   HA 
+B Q H >   HB 
+BAR Q H > HBAR 
+
+; hamza + fathatan
+H / a _ Q? AN  > Ha
+;;H / [Iyi] _ Q? AN  > Hy 
+H / _ Q? AN > Hy
+
+;final hamzas
+H / A _ Q? [VWL]? N? (#|^[LTR]) > HBAR
+H / [Uw] _ Q? [VWL]? N? (#|^[LTR]) > HBAR
+H / [Iy] _ Q? (i N?|u N?)? (#|^[LTR]) > HBAR ; not AN ...
+H / a _ Q? (#|^[LTR]) > Ha ; not AN... 
+H / a _ Q? (u|i) N? (#|^[LTR]) > Ha ; not AN... 
+H / i _ Q? [VWL]? N? (#|^[LTR]) > Hy
+H / u _ Q? [VWL]? N? (#|^[LTR]) > Hw
+
+;;and more hamzas ...
+
+; n°1: kasra/ya
+H / [LTR] _ Q? [Iyi] ^N > Hy
+H / [Iyi] _ > Hy
+
+; n°2: damma/waw 
+H / [LTR] _ Q? (U|u) ^N > Hw
+H / (U|u) _ Q? ^[Iyi] > Hw
+H / U _ > HBAR
+
+; n°3: fatha/alif
+H Q? A > HA
+H / A _ Q? a > HBAR
+H / ^[UI] _ Q? a > Ha
+H / a _ Q? ^[UI] > Ha
+
+
+pass(Unicode)
+
+Class [Digits] = (U+0030..U+0039)
+Class [ArDigits] = (U+0660..U+0669)
+Class [ArEasternDigits] = (U+06F0..U+06F9)
+Class [VWL]         = (U+0061  U+0065  U+0069  U+006F  U+0075 ) ;(a e i o u) 
+Class [LTR]	    = (U+0061..U+007A U+0041 U+0045 U+0049 U+004F U+0055 U+0042 U+004E U+0054 U+0059 U+005E U+002E U+005F U+0022 U+0027 U+0060) ;(a..z A E I O U B N T Y ^ . _ " "" `)
+Class [Iy]	    = (U+0049 U+0079) ;(I y)
+Class [Uw]          = (U+0055 U+0077) ;(U w)
+
+
+
+
+;normal code
+
+
+(P z|U+1E93) (P z|U+1E93)  > U+0638 U+0651  ;;SHADDA
+(P z|U+1E93)  > U+0638 
+C z C z	  > U+0698 U+0651  ;;SHADDA
+C z	  > U+0698 
+V z V z   > U+0696 U+0651 ;;SHADDA
+V z       > U+0696 ;pashto
+z z	  > U+0632 U+0651  ;;SHADDA
+z	  > U+0632 
+Q Y Y  > U+0649 U+0651  ;;SHADDA
+Q Y	  > U+0649 
+Y Y	  > U+064E U+0649 U+0651  ;;SHADDA
+Y	  > U+064E U+0649 
+P I P I / _ # > U+0650 U+06CC U+0651 ;; SHADDA
+P I / _ # > U+0650 U+06CC ;; workaround to have ya without dots in Arabic
+y y  > U+064A U+0651  ;;SHADDA
+y	  > U+064A 
+v v > U+06A5 U+0651 ;;SHADDA 
+v > U+06A5 ;; maghribi
+W 	  > U+0648 U+0627 ; waw + alif 
+C U       > U+064F U+0648 U+0653; historical spelling : madda on waw
+C w C w > U+06C9 U+0651 ;;SHADDA
+C w > U+06C9
+CO w CO w > U+06CA U+0651 ;,SHADDA
+CO w > U+06CA
+U A / _ # > U+064F U+0648 U+0627 
+Q U [Uw]  > U+0648 U+0651  ;;SHADDA
+Q U       > U+0648 
+U [Uw]    > U+064F U+0648 U+0651  ;;SHADDA
+U         > U+064F U+0648 
+w w	  > U+0648 U+0651  ;;SHADDA
+w	  > U+0648 
+Q u N?    > 
+u N	  > U+064C 
+u	  > U+064F 
+V t V t   > U+0679 U+0651 ;;SHADDA
+V t 	  > U+0679 ;urdu
+UL (t|s) UL (t|s) > U+062B U+0651  ;;SHADDA
+UL (t|s)  > U+062B 
+(P t|U+1E6D) (P t|U+1E6D)  > U+0637 U+0651  ;;SHADDA
+(P t|U+1E6D)	  > U+0637 
+t t	  > U+062A U+0651  ;;SHADDA
+t	  > U+062A 
+T T Q AN  > U+0629 U+0651   ;;SHADDA
+T Q AN    > U+0629   
+T AN      > U+0629 U+064B
+T T	  > U+0629 U+0651  ;;SHADDA
+T	  > U+0629 
+C s C s   > U+0634 U+0651  ;;SHADDA
+C s	  > U+0634 
+(P s|U+1E63) (P s|U+1E63)  > U+0635 U+0651  ;;SHADDA
+(P s|U+1E63)  > U+0635 
+V s V s   > U+069A U+0651 ;;SHADDA
+V s       > U+069A ;Pashto
+CO s CO s > U+069B U+0651 ;;SHADDA
+CO s      > U+069B ; sin with 3 dots below
+s s	  > U+0633 U+0651  ;;SHADDA
+s	  > U+0633 
+C r C r	  > U+06EF U+0651 ;;SHADDA ; missing in ArabTeX  <<<
+C r	  > U+06EF 
+CO r CO r > U+0697 U+0651 ;;SHADDA ; missing in ArabTeX <<<
+CO r      > U+0697 ; Dargwa language, Dagestan
+P r P r	  > U+0694 U+0651 ;;SHADDA
+P r	  > U+0694 ; r with dot below; in Kurdish .r -> 0694 
+V r V r  > U+0691 U+0651 ;;SHADDA
+V r 	  > U+0691 ;Urdu 
+r r	  > U+0631 U+0651  ;;SHADDA
+r	  > U+0631 
+P q P q	  > U+066F U+0651  ;;SHADDA
+P q	  > U+066F 
+q q > U+06A7 U+0651 ;; maghribi
+q > U+06A7 ;; maghribi
+p p	  > U+067E U+0651  ;;SHADDA
+p	  > U+067E 
+CO n CO n > U+06B1 U+0651 ;;SHADDA
+CO n > U+06B1
+C n C n  > U+06AD U+0651  ;;SHADDA
+C n      > U+06AD 
+P n P n   > U+06BA U+0651  ;;SHADDA
+P n 	  > U+06BA 	   ; urdu
+V n V n   > U+06BC U+0651 ;;SHADDA 
+V n       > U+06BC ; Pashto
+n n	  > U+0646 U+0651  ;;SHADDA
+n	  > U+0646 
+m m	  > U+0645 U+0651  ;;SHADDA
+m	  > U+0645 
+P l P l	  > U+06B6 U+0651 ; 
+P l 	  > U+06B6 ; ADDED <<< 
+C l C l	  > U+06B5 U+0651 ;;SHADDA
+C l	  > U+06B5 ;kurdish 
+l l	  > U+0644 U+0651  ;;SHADDA
+l	  > U+0644 
+P k P k > U+06A9 U+0651 ;;SHADDA 
+P k > U+06A9 ; pashto urdu 
+UL k UL k > U+063A U+0651 ;;SHADDA
+UL k  > U+063A
+k k	  > U+0643 U+0651  ;;SHADDA
+k	  > U+0643 
+C g C g   > U+062C U+0651  ;;SHADDA
+C g	  > U+062C 
+CO j CO j > U+0684 U+0651 ;;SHADDA
+CO j > U+0684
+j j	  > U+062C U+0651  ;;SHADDA
+j	  > U+062C 
+Q I       > U+064A
+I         > U+0650 U+064A
+C I       > U+0650 U+064A U+0653 ; historical spelling : madda on ya
+UL I 	  >  ; suggestion of O. Smrz (arabtex-plus project) URL = ###
+Q i N?    >  ;
+i N       > U+064D 
+i         > U+0650 
+x x	  > U+062E U+0651  ;;SHADDA
+x	  > U+062E 
+UL h UL h > U+062E U+0651  ;;SHADDA
+UL h	  > U+062E 
+(P h|U+1E25) (P h|U+1E25)  > U+062D U+0651  ;;SHADDA
+(P h|U+1E25) > U+062D 
+V h V h   > U+06C1 U+0651 ;;SHADDA
+V h       > U+06C1
+h h  > U+0647 U+0651  ;;SHADDA
+h	  > U+0647 
+CO g CO g > U+06B3 U+0651 ;;SHADDA
+CO g > U+06B3
+V g V g > U+06AC U+0651 ;;SHADDA
+V g > U+06AC ; kaf with dot above
+P g P g	  > U+063A U+0651  ;;SHADDA
+P g	  > U+063A 
+g g	  > U+06AF U+0651  ;;SHADDA
+g	  > U+06AF 
+G G       > U+06AB U+0651  ;;SHADDA
+G         > U+06AB ;pashto
+P f P f	  > U+06A1 U+0651  ;;SHADDA
+P f	  > U+06A1 
+f f > U+06A2 U+0651 ;; SHADDA
+f > U+06A2 ;; maghribi
+V d V d > U+068A U+0651 ;;SHADDA
+V d > U+068A
+P V d P V d > U+068B U+0651 ;;SHADDA
+P V d > U+068B ;; (like U+0688 with dot below, for Lahnda = Western Punjabi) 
+C d C d > U+06EE U+0651 ;;SHADDA
+C d > U+06EE ; not defined in ArabTeX
+CO d CO d > U+068F U+0651 ;;SHADDA
+CO d > U+068F
+UL (d|z) UL (d|z) > U+0630 U+0651  ;;SHADDA
+UL (d|z)  > U+0630 
+(P d|U+1E0D) (P d|U+1E0D)  > U+0636 U+0651  ;;SHADDA
+(P d|U+1E0D)  > U+0636 
+d d	  > U+062F U+0651  ;;SHADDA
+d	  > U+062F 
+V c V c   > U+0685 U+0651 ;;SHADDA
+V c	  > U+0685 ;pashto 
+C c C c	  > U+0686 U+0651  ;;SHADDA
+C c	  > U+0686
+P C c P C c > U+06BF U+0651 ;;SHADDA
+P C c > U+06BF ;; cheh with dot above <<<
+c c  > U+0681 U+0651 ;;SHADDA
+c	  > U+0681 ; ح with hamza above
+CO c CO c > U+0682 U+0651 ;;SHADDA
+CO c > U+0682 ; old pashto 
+CO b CO b > U+067B U+0651 ;;SHADDA
+CO b > U+067B
+P b P b   > U+066E U+0651  ;;SHADDA
+P b       > U+066E 
+b b	  > U+0628 U+0651  ;;SHADDA
+b	  > U+0628 
+H A       > U+0622 
+C A       > U+064E U+0622 ; historical spelling
+UL A      > U+064E U+0649 
+UL a	  > U+0670
+UL u   > U+0657 ; inverted damma
+UL i   > U+0656 ; subscript alif
+Q AN / (A H|Ha) _ > ; 
+AN / (A H|Ha) _ > U+064B
+Q AN   > U+0627 
+AN 	  > U+0627 U+064B 
+Q a	  > 
+a 	  > U+064E 
+Q A	  > U+0627 
+A	  > U+064E U+0627 
+Ain Ain	  > U+0639 U+0651  ;;SHADDA
+Ain	  > U+0639 
+H H	  > U+0621 U+0651 ;;SHADDA
+H	  > U+0621 
+U+002C / [Digits] _ [Digits] > U+002C  ; comma in numerical context not an Arabic comma
+Z Z / [Digits] _ [Digits] > U+2013  ; two hyphens in numerical context -> endash
+(Z|U+005D|U+005B|U+0028|U+0029)=xx / [Digits] _ [Digits] > U+202D @xx U+202C ; hyphen or brackets in numerical context: surrounded by LRO & PDF marks
+[Digits]  > [ArDigits]
+Z Z Z / # _ #    > U+2014 
+Z Z / # _ #    > U+2013 
+Z Z       > U+0640
+B         > U+0640
+Z / # _   > U+200D  ; -x  > force initial form with "zero-width joiner"
+Z / _ #   > U+200D  ;  x- > force final form    
+U+003C U+003C  > U+00BB ;<<
+U+003E U+003E  > U+00AB ;>>
+U+00AB    > U+00BB 
+U+00BB    > U+00AB
+UL        > U+0640  ;_
+U+003F    > U+061F  ;?
+U+003B    > U+061B  ;;
+U+003A U+003A > U+061E ;; '::' ligature 'DOTS' also defined above...
+U+002C    > U+060C  ;,
+U+005D / [LTR] _ [LTR]   > U+200D U+005D U+200D ; keep shaping when inserting ...[...]... ... 
+U+005B / [LTR] _ [LTR]   > U+200D U+005B U+200D ; (but of course this screws up contextual 
+                                                ; analysis, so quoting of hamzas etc may be necessary
+;U+005D    > U+005B  ;]  these are automatically mirrored, following Unicode rules
+;U+005B    > U+005D  ;[
+Q [VWL]   >   ;stripoff quoted vowels
+Q 	  > U+0652
+U+00B0	  > U+0652 ; °
+N         >   ;
+C         >   ;^
+BAR       >   ;| 
+
+pass(Unicode)
+ 
+Class [CONS] = (U+0621 U+0623 U+0624 U+0626 U+0628 U+062A..U+063A U+0641..U+0647 U+066E U+066F U+0672 U+0675 U+0676 U+0678..U+06BF U+06EE U+06EF U+06FA..U+06FF U+0750..U+076D) ; to be improved: not sure that all glyphs listed (esp for Urdu,Sindhi,etc are really "consonants" in the sense that they can carry a sukun, not to mention that some of these languages may not have a sukun at all ;-)
+Class [VWLX] = (U+0627 U+0648..U+065E U+0670) ; long vowels, vowel signs, shadda, sukun etc.
+Class [VWL]  = (U+0627 U+0648..U+0650 U+0670);
+U+0627 [VWL]?  / [VWL] Z _ > U+0627 ; e.g. fa-isti_hraj -> initial alif has no vowel 
+Z > ;