Sorry to refer to the wrong code point of arabic digit.
It would be U+0660-U+0669.
The bidi paper I refered to is "Unicode Technical Report #9" which is
"Unicode 3.0 Bidirectional Algorithm".
It's at http://www.unicode.org/unicode/reports/tr9/tr9-5.html.
Table 3-7. Bidirectional Character Types
Description Right-to-Left Arabic
Scope Arabic, ..... (0600-07BF,.....)
] Date: Thu, 28 Oct 1999 13:01:22 -0700
] From: email@example.com (Yung-Fong Tang)
] Subject: Re: arabic number in bidi algorithm
] To: Chookij Vanatham <chookij.vanatham@eng.Sun.COM>
] Cc: Unicode List <firstname.lastname@example.org>
] MIME-version: 1.0
] X-Accept-Language: de,ja,el,fr,en-GB,ca
] hum ? Where it said it is "strong type, RIGHT-TO-LEFT" ? Which book ? Which
] ? Which page ? Or which file on-line ?
] Unicode 2.0 said the following...
] 3.11 Bidirectional Behavior [start 3-14]
] Table-3-5 Bidirectional Character Types [page 3-17]
] EN Weak European Number European digits, Eastern Arabic-Indic digits, ...
] AN Weak Arabic Number Arabic-Indic digits, Arabic decimal & thousands
] Chookij Vanatham wrote:
] > Hello,
] > According to the Bidi algorithm, the arabic number (U+0671 - U+0679) has the
] > strong type, RIGHT-TO-LEFT.
] U+671-U+679 is NOT arabic digit. U+0671-U+0679 is exteneded Arabic letters
] WASLA - TTEH
] U+0660 - U+0669 is arabic digit.
] > So, regardless of RLO and RLE, can I say that
] > the arabic number (U+0671 - U+0679) would be always written from RIGHT to
] > This is confused me due to I see that, in miscrosoft word (arabic version),
] > the arabic number is written from LEFT to RIGHT.
] > May be someone can let me know which one is right or I miss anything ?
] > Thanks a lot.
] > Chookij V.
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:54 EDT