ASCII Char remove not functioning

Question

If a string contains any non-ASCII value, i need to to delete it.

I have tried replaceAll mrthod which is working fine with jdk 1.8 and above. but i want to deploy the same on jdk 1.6 and is it not functioning.

String Remarks2 ="hii:╘’i";

String Remarks = Remarks2.replaceAll("[^\\p{ASCII}]", "");
System.out.println("ans: "+Remarks);

Output in jdk 1.8: hii:i Output in jdk 1.6: hii:╘’i

Actual result must be- hii:i

Have you tried to use "[^\\x00-\\x7F]" instead of "[^\\p{ASCII}]". Maybe "\p[ASCII}" is not defined in 1.6 — SirFartALot, Mar 28 '19 at 09:59
`\p{ASCII}` is defined in Java 6, see https://docs.oracle.com/javase/6/docs/api/java/util/regex/Pattern.html — Mincong Huang, Mar 28 '19 at 11:29
@T.J.Crowder Thanks. but what i have understand is that when you press and hold down ALT while typing the character code we can generate the ascii key. For example, to insert the degree (º) symbol, press and hold down ALT while typing 0176 on the numeric keypad. Also people sometime copy paste the content from excel sheet resulting the change in schema when we try to store the same through input field from the html text box. — Shreyas, Mar 29 '19 at 06:44
@kometen yes!! need to replace it with blank character i.e "" — Shreyas, Mar 29 '19 at 06:45
I tried it with jdk1.6.0_24 and the code you provided DID output "hii:i". Have you really tested this? On which version of JDK1.6 did you test it? — SirFartALot, Mar 29 '19 at 08:28
@SirFartALot once again i checked, it is not working. we have 1.6.0_45 installed. Please suggest — Shreyas, Mar 30 '19 at 09:11
Okay there's one update, ascii is working absolutely fine. but when a string contains single quote as ’ rather than ' it is not working. ’ vs ' (copy paste in notepad to see the difference). what kind of characters are they?? how to identify such characters. @SirFartALot — Shreyas, Mar 30 '19 at 09:44

score 0 · Accepted Answer · answered Mar 29 '19 at 08:55

As commented \p{ASCII} came later. As ASCII is the first 7-bits range:

String remarks = remarks2.replaceAll("[^\u0000-\u007F]", "");

You might try to normalize text decomposing é into e and a zero-width combining mark ´ first:

remarks2 = Normalizer.normalize(remarks2, Form.NFD);

1 Answers1