Ignore soft hyphens, word breaks & zero-width stuff

Please add support for highlighting content that includes a soft hyphen (­), a word break opportunity HTML tag (<wbr>) or a zero-width space (​).

In this demo, I added a ­, <wbr> and ​ at various locations inside of the word “Lorem”. And when using mark.js to highlight “Lorem”, only two out of the five results are marked. Performing a browser search reveals all five matches.

I believe this applies to all browsers, but I did test and got the same results in the latest versions of Chrome & Firefox (Windows 10) with mark.js version (8.1.1).

References:

• http://www.quirksmode.org/oddsandends/wbr.html
• https://developer.mozilla.org/en-US/docs/Web/HTML/Element/wbr
• http://www.fileformat.info/info/unicode/char/200b/index.htm

I also found that zero-width non-joiner (‌) and zero-width joiner (‍) also produces the same result. Odd that they aren’t discussed as much… I guess because these last two are joiners don’t appear work with css hyphens - demo modified from this page.

[Mottie]

Ok, I think I’m going to go with ignoreJoiners because my other idea of using killWhitey didn’t seem too appropriate either.

I’ve started working on tests, and I have an odd question - why does the across elements filter file not have any elements to span across?

[Mottie]

They don’t seem to be really known.

I’m all about weird stuff. Like me! 😀

If this doesn’t require a lot of code we can support this.

I’ve already got something working. The regular expression starts to look really ugly, so I think disabling this option by default would be preferred.

So what would you want to name this option? I started out with ignoreWhitespace but that isn’t accurate enough. It only applies to soft hyphens and some zero-width characters.