RegExpJsToJavaConverter \p{L} Unicode property escapes

[Jaaap]

We have an HTML pattern that looks like this:

<input type="text" pattern="[\p{L}0-9 _'@&,?/]*"/>

Observe the \p{L}. This is a Unicode property escape for the Letter group.
The RegExpJsToJavaConverter.convert() method replaces this with [p\{L\}0-9 _'@&,?/]*.
Notice the missing backslash before the p and the added backslashed before the { and }.

Is it possible to fix this?
Also see https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_expressions/Unicode_property_escapes

[rbri]

Yes will try to fix that - hope you are already at version 3.1

[rbri]

Have done a fix for this - please check the latest 3.2.0-SNAPSHOT.

[Jaaap]

We're waiting for a spring update to allow our project to use the 3.x release of HtmlUnit.
I did take a look at your code and it looks good.
Can we perhaps add a test like:

assertEquals("[^\\p{gc=Co}\\p{gc=Cn}]+", regExpJsToJavaConverter.convert("[^\\p{gc=Co}\\p{gc=Cn}]+"));

[rbri]

yes, passes also
can you provide some input that passes this and some that fails...
To add a test case similar like this

@Test
@Alerts("true")
public void validationPatternUnicodePropertyEscapeLowercaseLetter() throws Exception {
    validation("\\p{Ll}*", "&#x043C&#x0439&#x0440");
}

[rbri]

Ok found something....

[rbri]

Please have a look at my last commit.

[Jaaap]

The Cn group is currently empty (so untestable), according to https://www.fileformat.info/info/unicode/category/Cn/index.htm
The Co group has six characters: https://www.fileformat.info/info/unicode/category/Co/list.htm

@Test
@Alerts("true")
public void validationPatternUnicodePropertyEscapeCategoryPrivateUse() throws Exception {
    validation("[\\p{gc=Co}]+", ""); //Not sure this will copy/paste right
    validation("[\\p{gc=Co}]+", "󰀀");
}

[rbri]

Done, will close this. Thanks.