Arabic virtual keyboard

RTL regex mode: Use this mode for writing right-to-left (RTL) regular expressions. All characters (even latin letters) will appear in RTL order: e.g., abc will appear as cba. Copying the regex to clipboard will remove the characters (U+061C (ARABIC LETTER MARK)) that force this behavior so that they do not impact the functioning of the regex.
Separate Arabic letters: Arabic letters will be displayed as separated (by inserting the separator character U+180E "MONGOLIAN VOWEL SEPARATOR"). This is especially useful for writing character disjunctions `[abc]` and ranges `[a-c]`. Copying the regex to clipboard will remove these separators so that they do not impact the functioning of the regex.

Need more special characters?

Add a key with a new character (and an optional description):


Regex Cheatsheet input (source)

Sample regexSample match
Any character except line breaka.cabc a.c
Escapes a special charactera\.cabc a.c
Character groups ∨
one Unicode digit in any scriptfile_\d_\dfile_9_ ٩
"word character": Unicode letter, ideogram, digit, or underscore\w-\w\w\w字-ま_۳
"whitespace character": any Unicode separatora\sb\scaaaa b cccdde
One character that is not a digit2\D\D\D12ABC34
One character that is not a word character\W\W\W\W\W*-+=)
One character that is not a whitespace character\S\S\S\SYo3) Y o3
More White-Space ∨
TabT\t\w\wA T ab
Carriage return charactersee below
Line feed charactersee below
Line separator on WindowsAB\r\nCDAB
CD
Quantifiers ∨
Once or noneplurals?plural and plurals
Zero or more times (greedy) aB*a bB*b aB*aBBa bb aBBBBB
One or more (greedy) Version \w-\w+Version A-b1_100
Exactly three times\D{3}ABC
Two to four times (greedy) \d{2,4}156
Three or more times (greedy) \w{4,}regex_
Lazy quantifiers (w/ question mark) ∨
Once or none (lazy)plurals?plural and plurals
One or more (lazy) Version \w-\w+?Version A-b1_100
Two to four times (lazy) \d{2,4}?156
Three or more times (lazy) \w{4,}?regex_
Zero or more times (lazy) aB*a b* aB*?aBBa bb aBBBBB
Logic ∨
Alternation / OR operand22|3322 + 11 = 33
Capturing groupA(nt|pple)An Ant on an Apple
Contents of Group 1r(\w)g\1xregex not rigex
Contents of Group 2(\d\d)\+(\d\d)=\2\+\112+65=65+12
Non-capturing groupA(?:nt|pple)An Ant on an Apple
Character Classes ∨
One of the characters in the bracketsT[ao]pTap and Top
One of the characters in the range from x to y[A-Z]+UPPERcase LEtterS
One character that is not x or y[^12]A1b2+1
One of the characters not in the range from x to y[^a-z]{3}abA1! cd
Anchors and Boundaries ∨
Start of line^abc [1-9]*abc 18 abc 19
End of lineabc [1-9]*$abc 18 abc 19
Word boundary: position where one side only is a letter, digit or underscore\bكتب\bكتب مكتبه كتبه
Not a word boundary\Bكتب\Bكتب مكتبه كتبه
Lookarounds ∨
Positive lookahead(?=\d{10})\d{5}0123456789
Positive lookbehind(?<=\d\s)cat1 cat white cat
Negative lookahead(?!theatre)the\w*the theatre theme
Negative lookbehind(?<!USD)\d{3,}USD300 = GBP250
Arabic Unicode ∨
Any Arabic character in Unicode
Any Arabic punctuation character in Unicode
Any Arabic diacritic in Unicode
Any Arabic/hindi number in Unicode

This keyboard is an adaptation of Ahmad Ismail's Arabic keyboard to the needs of researchers working with Arabic palaeography and regular expressions. The source code is on GitHub. You can report issues here.