\d only matchs 0-9 digits?

Question

As far as I know, \d should matchs non-english digits, e.g. ۱۲۳۴۵۶۷۸۹۰ but it doesn't work properly in JavaScript.

See this jsFiddle: http://jsfiddle.net/xZpam/

Is this a normal behavior?

It is normal. The question is whether you can enable the unicode behavior in javascript regexes. Chrome doesn't like the `u` flag. — John Dvorak, May 21 '13 at 05:49
http://stackoverflow.com/questions/150033/regular-expression-to-match-non-english-characters — Ravi Gadag, May 21 '13 at 05:52
Since it only matches [0-9] why not try something like `^[۱۲۳۴۵۶۷۸۹۰]+$`? — Menno, May 21 '13 at 05:53

score 11 · Answer 1 · edited May 23 '17 at 12:34

11

It seems that JavaScript does not support this (along with other weaknesses of the language in RegExp). However there's a library called XRegExp that has a unicode addon, which enables unicode support through \p{} category definition. For example if you use \p{Nd} instead of \d it will match digits:

<script src="xregexp-all.js" type="text/javascript"></script>
<script type="text/javascript">
    var englishDigits = '123123';
    var nonEnglishDigits = '۱۲۳۱۲۳';

    var digitsPattern = XRegExp('\\p{Nd}+');
    if (digitsPattern.test(nonEnglishDigits)) {
        alert('Non-english using xregexp');
    }

    if (digitsPattern.test(englishDigits)) {
        alert('English using xregexp');
    }
</script>

EDIT:

Used \p{Nd} instead of \p{N} as it seems that \d is equivalent to \p{Nd} in non ECMA Script Regex engines. Thanks go to Shervin for pointing it out. See also this fiddle by Shervin.

edited May 23 '17 at 12:34

Community

1
1

answered May 21 '13 at 06:27

Sina Iravanian

16,011
4
34
45

And if you want to limit your input to [Arabic-Indic](http://unicode.org/cldr/utility/list-unicodeset.jsp?a=%5Cp%7Bsubhead%3DArabic-Indic+digits%7D&g=) or [Eastern Arabic-Indic Digits](http://unicode.org/cldr/utility/list-unicodeset.jsp?a=\p{subhead=Eastern%20Arabic-Indic%20digits}) you can use Unicode block criteria: `alert(XRegExp("^\\p{InArabic}\\p{N}").test('۱۲۳۴٤۵٥۶۷۸۹۰')); // True` `alert(XRegExp("^\\p{InArabic}\\p{N}").test('1234567890')); // False` – Shervin May 21 '13 at 06:43
2

@Sina, I think that `\\p{N}` (Number) should be replaced with `\\p{Nd}` (Decimal Number) since we don't want to match [non-decimal numeral characters](http://unicode.org/cldr/utility/list-unicodeset.jsp?a=%5Cp%7BN%7D-%5Cp%7BNd%7D&g=) like ➋, ⅑, Ⅷ, etc.: http://jsfiddle.net/wZXZ3/2/ – Shervin May 21 '13 at 07:26
1

@Shervin Thanks, I updated the answer, and linked to your fiddle. – Sina Iravanian May 21 '13 at 11:40

score 10 · Accepted Answer · answered May 21 '13 at 05:50

10

JavaScript does not support Unicode regex matching (and it is far from the only language where such is true).

http://www.regular-expressions.info/unicode.html

answered May 21 '13 at 05:50

Amber

507,862
82
626
550

2

Adding to what Amber said, `JavaScript uses ASCII for \d and \w, but Unicode for \s`. [Source](http://www.regular-expressions.info/charclass.html) – Vimal Stan May 21 '13 at 05:51
yes, and now it's a trouble having good client validation with JS regexps – RusAlex May 21 '13 at 06:00
Languages that properly support Unicode regex matching tend to be the exception, actually. – R. Martinho Fernandes May 21 '13 at 10:47

score 3 · Answer 3 · answered May 21 '13 at 05:51

3

In the documention of Mozilla Firefox (https://developer.mozilla.org/en-US/docs/JavaScript/Reference/Global_Objects/RegExp) you will find that:

\d  

Matches a digit character in the basic Latin alphabet. Equivalent to [0-9].

answered May 21 '13 at 05:51

kaljak

1,263
1
16
34

score 2 · Answer 4 · answered May 21 '13 at 05:51

2

\d is equivalent to [0-9], according to MDN.

answered May 21 '13 at 05:51

Arjan

9,784
1
31
41

score 1 · Answer 5 · answered May 21 '13 at 05:51

1

From MDN . RegEx Test

Matches a digit character in the basic Latin alphabet. Equivalent to [0-9].

answered May 21 '13 at 05:51

Ravi Gadag

15,735
5
57
83

score 1 · Answer 6 · answered May 21 '13 at 06:05

1

Matches a digit character. Equivalent to [0-9].

For example, /\d/ or /[0-9]/ matches '2' in "B2 is the suite number."

From MDN

answered May 21 '13 at 06:05

Dinever

690
5
13

score 1 · Answer 7 · answered May 21 '13 at 06:49

Yes, it is normal and correct that \d matches the Ascii digits 0 to 9 only. The authoritative reference is the ECMAScript standard. It is not particularly easy reading, but clause 15.10.2.12 (CharacterClassEscape) specifies that \d denotes “the ten-element set of characters containing the characters 0 through 9 inclusive”.

score 0 · Answer 8 · answered Mar 06 '18 at 10:53

Yes, using \d not matching properly to none-English numbers in JavaScript, But like other weird parts of JavaScript, you can still check none-English numbers (like Persian numbers for example) in JavaScript, using something like the code below:

/[۰, ۹]/.test("۱۲۳۴۵۶۷۸۹۰"); //true

\d only matchs 0-9 digits?

8 Answers8

EDIT:

Linked