# Transliteration encoding for the Quran 
# Last edited on 2004-01-26 03:47:37 by stolfi
# 
# JSAR is J. Stolfi's ad-hoc remapping of HAR (HTML Arabic
# transcription alphabet) to a more compact encoding, using only one
# byte for lowercase HAR "letters" and caret "^" for uppercase.
# 
# JSAR was a hacked-up solution to an immediate problem. It is now
# obsolete since Unicode versions of the Quran are becoming available
# (although they are still quite buggy).
# 
#                  Lowercase              Uppercase              
#                  ---------------------  ---------------------
#   UC Phon Name      HAR              JSAR  HAR              JSAR
#   -- ---- --------- ---------------  ----  ---------------  ----
#   21 h_u  h_ua'     kh               ©     KH               ^©  
#   24 ?    hamza     /                '     (/)              ^'  
#   27 a    'alif     a                a     A                ^a  
#   28 b    ba'       b                b     B                ^b  
#   2A t    ta'       t                t     T                ^t  
#   2B t_-  t_-a'     th               þ     TH               ^þ  
#   2C g^v  g^vim     j                j     J                ^j  
#   2D h_.  h_.a'     <U>h</U>         µ     <U>H</U>         ^µ  
#   2F d    dal       d                d     D                ^d  
#   30 d_-  d_-al     <U>th</U>        £     <U>Th</U>        ^£  
#   31 r    ra'       r                r     R                ^r  
#   32 z    zai       z                z     Z                ^z  
#   33 s    sin       s                s     S                ^s  
#   34 s^v  s^vin     sh               x     SH               ^x  
#   35 s_.  s_.ad     <U>s</U>         ß     <U>S</U>         ^ß  
#   36 d_.  d_.ad     <U>d</U>         ð     <U>D</U>         ^ð  
#   37 t_.  t_.a'     <U>t</U>         ±     <U>T</U>         ^±  
#   38 z_.  z_.a'     <i><u>th</u></i> ç     <i><u>Th</u></i> ^ç  
#   39 `    `ayn      AA               ä     (AA)             ^ä  
#   3A g^.  g^.yn     gh               ¤     GH               ^¤  
#   41 f    fa'       f                f     F                ^f  
#   42 q    qaf       q                q     Q                ^q  
#   43 k    kaf       k                k     K                ^k  
#   44 l    lam       l                l     L                ^l  
#   45 m    mim       m                m     M                ^m  
#   46 n    nun       n                n     N                ^n  
#   47 h    ha'       h                h     H                ^h  
#   48 w    waw       w                w     W                ^w  
#   4A y    ya'       y                y     Y                ^y  
#
# Vowel marks
#
#   UC Phon  Name     HAR              JSAR  HAR              JSAR
#   -- ----- -------- ---------------  ----  ---------------  ----
#   4E a     al.md.   <U>a</U>         â     <U>A</U>         ^â  
#   50 i              i                i     I                ^i  
#   4F u              u                u     U                ^u  
#
# Long vowels
#
#   50+4A             ee               ë     EE               ^ë  
#   27+4F             o                o     O                ^o  
#   4F+48             oo               ö     OO               ^ö  
# 
# Unknown:
# 
#   UC Phon  Name     HAR              JSAR  HAR              JSAR
#   -- ----- -------- ---------------  ----  ---------------  ----
#   ?? ?              <U>n</U>         ñ     <U>N</U>         ^ñ  
#   ?? ?              <U>l</U>         ¬     <U>L</U>         ^¬  
# 
# The left column is the low-order byte (in hex) of the corresponding
# Unicode point, in the Arabic range (u0600 - u06FF).
# 
# # In the "Phon" and "Name" columns, "^." is dot-above, "^v" is
# hacheck, "_-" is macron-below, "_." is dot-below, "_u" is
# brevis-below.
# 
# Note that the Arabic script itself has no case distinction, and
# anyway the HAR codes "AA" and "/" have no uppercase equivalent. 
# 
# The JSAR codes "ñ" (HAR underlined "n") and "¬" (HAR underlined "l")
# are used only once in the whole text (respectively, in 5:107
# "faâ©arañi", and 74:9 "fa£â¬ika"). They were not defined in the HAR
# table. I am assuming that may be errors in the original files.
# Perhaps ¬ is lam-'alif?
#