Detecting arabic characters with regex.

regex-arabic.md

Detect all Arabic Characters:

/[\u0600-\u06ff]|[\u0750-\u077f]|[\ufb50-\ufbc1]|[\ufbd3-\ufd3f]|[\ufd50-\ufd8f]|[\ufd92-\ufdc7]|[\ufe70-\ufefc]|[\uFDF0-\uFDFD]/

Summary:

  Arabic (0600—06FF, 225 characters)

  Arabic Supplement (0750—077F, 48 characters)

  Arabic Extended-A (08A0—08FF, 39 characters)

  Arabic Presentation Forms-A (FB50—FDFF, 608 characters)

  Arabic Presentation Forms-B (FE70—FEFF, 140 characters)

  Rumi Numeral Symbols (10E60—10E7F, 31 characters)

  Arabic Mathematical Alphabetic Symbols (1EE00—1EEFF, 143 characters)

For more info check this Wiki link to see arabic letters in Unicode:

https://en.wikipedia.org/wiki/Arabic_(Unicode_block)

References:

http://stackoverflow.com/questions/11323596/regular-expression-for-arabic-language

Very useful, thanks !

I used it in a regex instruction to keep arabic and numeric chars and remove the rest:

...
t = re.sub(r'[^0-9\u0600-\u06ff\u0750-\u077f\ufb50-\ufbc1\ufbd3-\ufd3f\ufd50-\ufd8f\ufd50-\ufd8f\ufe70-\ufefc\uFDF0-\uFDFD]+', ' ', text)
...

do you have an example in golang?

in PHP

$arabic_regex = '/[\x{0600}-\x{06FF}|\x{0750}-\x{077f}|\x{fb50}-\x{fbc1}|\x{fbd3}-\x{fd3f}|\x{fd50}-\x{fd8f}|\x{fd92}-\x{fdc7}|\x{fe70}-\x{fefc}|\x{FDF0}-\x{FDFD}]/u';