I have the following line of code to remove illegal characters from a file name:
str= str.replace(/([^a-z0-9]+)/gi, '-');
That works fine but it also removes the spaces, how can I only remove the illegal characters but leave the spaces?
Asked
Active
Viewed 3.5k times
33
-
How do you determine "illegal" character? Your regex says that `.`,`_` and many other characters are "illegal", Why? – Toto Feb 13 '17 at 17:55
-
3Just add the space to the class `[^a-z0-9 ]` that way it won't get matched (removed). – Feb 13 '17 at 19:20
4 Answers
72
Illegal characters are listed here. To replace them use this regex /[/\\?%*:|"<>]/g like this:
var filename = "f?:i/le> n%a|m\\e.ext";
filename = filename.replace(/[/\\?%*:|"<>]/g, '-');
console.log(filename);
ibrahim mahrir
- 31,174
- 5
- 48
- 73
-
These characters depend on OS. Under Linux you can have any character, even space in a filename. – Toto Feb 13 '17 at 17:56
-
@Toto I excluded the case of `.` and ` ` because they aren't really illegal! – ibrahim mahrir Feb 13 '17 at 18:52
-
You missed unprintable characters in ascii-range 0-31 (inclusive). – Stefan Steiger Sep 23 '21 at 08:19
-
2Down-vote, because it is more secure to allow-list allowed characters than trying to block-list invalid characters. – halllo Jun 10 '22 at 21:12
-
-
@EugeneP. there is [no need to escape those](https://stackoverflow.com/a/19976308/9867451) because inside a character class, they are considered literals. – ibrahim mahrir Mar 23 '23 at 16:59
9
You are searching for all non numeric and roman letters. If you want to exclude the space from being replaced:
Just add a space to the selection to exclude :)
str= str.replace(/([^a-z0-9 ]+)/gi, '-');
// add a space here -------^
Pineda
- 7,435
- 3
- 30
- 45
-
Sorry but I don't think I understand what you mean, could you explain? – user3378165 Feb 13 '17 at 17:49
-
Your regexp looks to match characters that are not: lowercase characters from a-z, uppercase from A-Z and numbers from 0-9, and replaces the match. To avoid replacing spaces, adding a space as I've recommended will achieve what you want because you're trying to match anything _**not**_ included between the `[ .. ]` square brackets – Pineda Feb 13 '17 at 17:50
9
You can use \W+
\W (non-word-character) matches any single character that doesn't match by \w (same as [^a-zA-Z0-9_])
Source: https://www.ntu.edu.sg/home/ehchua/programming/howto/Regexe.html
str = str.replace(/(\W+)/gi, '-');
user2226755
- 12,494
- 5
- 50
- 73
2
Add the '\s' whitespace to your exclusion.
str = str.replace(/([^a-z0-9\s]+)/gi, '-');
Kit
- 724
- 4
- 11