Regex any ASCII character

Question

What is the regex to match xxx[any ASCII character here, spaces included]+xxx?

I am trying xxx[(\w)(\W)(\s)]+xxx, but it doesn't seem to work.

score 140 · Answer 1 · edited Aug 07 '18 at 22:05

140

[ -~]

It was seen here. It matches all ASCII characters from the space to the tilde.

So your implementation would be:

xxx[ -~]+xxx

edited Aug 07 '18 at 22:05

Peter Mortensen

30,738
21
105
131

answered Jan 30 '13 at 16:26

luk3thomas

2,512
1
18
20

This is perfect, since the accepted answer did not work with [RegularExpression] attribute in ASP.NET MVC - it gets rendered as Unicode chars and this breaks the validation. – Art Dec 08 '15 at 03:36
4

Really elegant solution, highly readable and semantically exactly what I was looking for. – machine yearning Apr 28 '16 at 13:49
@Art This is not working for me with a `[RegularExpression]` attribute... Did you have to do anything else special for it to work? My attribute is just `` but won't even accept "asdf". – Zack Aug 26 '16 at 18:44
1

I guess never mind. I just realized that will only match one character, so I had to add the + to make it `[ -~]+` "any ascii character, one or more times" for my usage. – Zack Aug 26 '16 at 18:49
1

@Zack check out this article I wrote with the fully working example code: http://nimblegecko.com/how-to-validate-a-textfield-for-only-printable-characters-in-aspnet-mvc/ Hope it helps and let me know if you're stuck! – Art Aug 27 '16 at 00:11
@Art Yeah it is working, I just didn't think about how I would need the + on the end to make it match a character "one or more" times, and I got confused when it wouldn't accept anything I tried to test with like "qwert", "1234" etc, because they were all longer than one character. – Zack Aug 29 '16 at 13:55
Very helpful. I used `[^ -~]` and replace every match by empty string to remove non ASCII characters. – Sebastian Oct 11 '19 at 07:09
great answer, see live demo (space excluded) https://regex101.com/r/Xuj0Pf/1 – humble_wolf Sep 26 '21 at 08:39

score 114 · Accepted Answer · edited Apr 11 '23 at 03:52

114

If you really mean any ASCII (e.g. not all Unicode characters):

xxx[\x00-\x7F]+xxx

JavaScript example:

var re = /xxx[\x00-\x7F]+xxx/;

re.test('xxxabcxxx')
// true

re.test('xxx☃☃☃xxx')
// false

edited Apr 11 '23 at 03:52

Nate Anderson

18,334
18
100
135

answered Jul 08 '10 at 11:52

Matthew Flaschen

278,309
50
514
539

1

this doesn't seem to work the way I thought it would. This matches the BYTE representations of ascii characters, not the visual printed / typed characters themselves? – fIwJlxSzApHEZIl Feb 27 '13 at 20:49
1

@advocate, yes it does. What language did you test in? That could be it. – Matthew Flaschen Feb 28 '13 at 06:09
1

oh I was hoping to find something to match the printed versions of ASCII characters :) – fIwJlxSzApHEZIl Feb 28 '13 at 09:36
16

@advocate, I think you're looking for `^[\x20-\x7F]+$`. – wrygiel Dec 19 '13 at 19:46
1

Such a nice solution, I have seen so many bad ones. I was not aware about regex being able to filter by the charcode, elegant! – John Mar 16 '14 at 13:48
1

wrygiel - !!! :) Just a tip for removing CR LF `e.replace(/\x0A/g,"").replace(/\x0D/g,"")` (hex of 10 and 13 are 0A and 0D respectively) – gordon Feb 18 '16 at 15:07
3

@wrygiel Small correction as `x7F` is not printable (it's the `DEL` char). It should be `^[\x20-\x7E]+$`. – AnorZaken Jun 28 '21 at 20:30

score 36 · Answer 3 · edited Aug 07 '18 at 22:04

36

You can use the [[:ascii:]] class.

edited Aug 07 '18 at 22:04

Peter Mortensen

30,738
21
105
131

answered Jul 08 '10 at 11:56

catwalk

6,340
25
16

@catwalk You are my hero for today! It wasn't accepting \x00-\x7F for some reason, but it took [:ascii:]. Thanks! – n0nag0n Jul 20 '12 at 22:31
12

Note that [[:ascii:]] would match any ASCII character, even non-printable characters, whereas [ -~] would match just the printable subset of ASCII. – elolos Sep 02 '14 at 11:58
2

*"character class"*? – Peter Mortensen Aug 07 '18 at 22:04
1

Note: A better name would be "POSIX expression", instead of "class". Be aware that it is not supported in every language, for example JavaScript, does not support it. – Daniel B. Jun 03 '20 at 14:21
I wish that worked in powershell. – js2010 Nov 01 '22 at 19:47

score 11 · Answer 4 · answered Jul 08 '10 at 11:55

11

Since US-ASCII characters are in the byte range of 0x00–0x7F (0–127):

xxx[\x00-\x7F]+xxx

answered Jul 08 '10 at 11:55

Gumbo

643,351
109
780
844

score 8 · Answer 5 · answered May 09 '19 at 09:50

8

Accepts / Matches only ASCII characters

/^[\x00-\x7F]*$/

answered May 09 '19 at 09:50

Vaibhav Gaikwad

811
2
12
21

score 2 · Answer 6 · answered Jul 08 '10 at 11:53

2

Try using .+ instead of [(\w)(\W)(\s)]+.

Note that this actually includes more than you need - ASCII only defines the first 128 characters.

answered Jul 08 '10 at 11:53

Mark Byers

811,555
193
1,581
1,452

score 0 · Answer 7 · edited Aug 07 '18 at 22:03

0

Depending on what you mean with "ASCII character" you could simply try:

xxx.+xxx

edited Aug 07 '18 at 22:03

Peter Mortensen

30,738
21
105
131

answered Jul 08 '10 at 11:54

RoToRa

37,635
12
69
105

score -1 · Answer 8 · answered Jul 08 '10 at 11:54

-1

. stands for any char, so you write your regex like this:

xxx.+xxx

answered Jul 08 '10 at 11:54

m_vitaly

11,856
5
47
63

Regex any ASCII character

8 Answers8

Linked

Related