Implementation of the Unicode BiDi algorithm
PHP Extension for GNU FriBidi, an implementation of the Unicode Bidirectional Algorithm (bidi)
This extension is basically a wrapper for the GNU FriBidi implementation
of the Unicode Bidirectional algorithm. The need for such an algorithm
rises from the bidirectional language usage done by applications.
Arabic/Hebrew embedded within English is such a case.
The only function used is fribidi_log2vis (logical [non-reversed] to visual [reversed]).
Input:
1) The Logical string.
2) Base direction of application. Possible values:
* FRIBIDI_AUTO (autodetect the base direction)
* FRIBIDI_LTR (left to right
* FRIBIDI_RTL (right to left)
* FRIBIDI_WLTR (weak left to right)
* FRIBIDI_WRTL (weak right to left)
3) The char code being used, which can be one of the following constants:
* FRIBIDI_CHARSET_UTF8 (Unicode)
* FRIBIDI_CHARSET_8859_6 (ISO-8859-6 - Arabic)
* FRIBIDI_CHARSET_8859_8 (ISO-8859-8 - Hebrew)
* FRIBIDI_CHARSET_CP1255 (MS Codepage 1255 - Hebrew/Yiddish)
* FRIBIDI_CHARSET_CP1256 (MS Codepage 1256 - Arabic)
* FRIBIDI_CHARSET_CAP_RTL (Used for test purposes, will treat CAPS as non-English letters)
Note: Currently, GNU FriBidi supports the above Character Codes alone.
Output: The visual string, on success, false, otherwise.
Static - ./configure --with-fribidi=base-directory-of-FriBidi-installation-path
(defaults to /usr/local).
dl - same, just add shared, to get … --with-fribidi=shared,base…
The function fribidi_log2vis computes three more arrays which are currently not passed back as output.
These arrays are:
1) mapping from the logical to the visual string.
2) mapping from the visual to the logical string.
3) embedding level of characters as figured out by the bidi algorithm.
The extension needs further implementation to support this.
p.s. - If you don’t understand this, you probably don’t need it.