Why does this regex/DOM character entity tester return `undefined`?

Question

var str = 'let us pretend that this is a blog about gardening&cooking; here&amp;apos;s an apostrophe &amp; ampersand just for fun.';

This is the string I'm operating on. The desired end result is: "let us pretend that this is a blog about gardening&cooking; here's an apostrophe & ampersand just for fun."

console.log('Before: ' + str);


str = str.replace(/&amp;(?:#x?)?[0-9a-z]+;?/gi, function(m){
  var d = document.createElement('div');
  console.log(m);
  d.innerHTML = m.replace(/&amp;/, '&');
  console.log(d.innerHTML + '|' + d.textContent);
  return !!d.textContent.match(m.replace(/&amp;/, '&')[0]) ? m : d.textContent;
});


console.log('After: ' + str);

You have `!!` at the beginning of your return. I don't believe that is valid syntax, and if it is, I think it cancels itself out. — Shmiddty, Sep 24 '12 at 17:05
@Shmiddty `!!` is used to cast an operand to a Boolean. It's valid syntax and I dont think it is related to the issue. — jbabey, Sep 24 '12 at 17:07
[this question](http://stackoverflow.com/questions/1219860/javascript-jquery-html-encoding) may have some answers for you. HTML encoding is one of those things where you should reuse a proven solution instead of trying to roll your own. — jbabey, Sep 24 '12 at 17:09
Nope... the "inner" (second) one converts it to a boolean expression, (and, like you say, inverts it), the second one re-(un-)inverts it. That's the idea, to cancel it out without having cumbersome nested parentheses. — wwaawaw, Sep 24 '12 at 17:10
I'm not sure what your point of matching the first character of that de-`&`ed string is, could you explain what you intend with that? — Bergi, Sep 24 '12 at 17:49

score 1 · Answer 1 · edited May 23 '17 at 12:12

1

The problem is that HTML doesn't support XML's ' To avoid the issue you should use ' instead of '

For more information look at this post:

Why shouldn't ' be used to escape single quotes?

edited May 23 '17 at 12:12

Community

1
1

answered Sep 24 '12 at 17:41

Warlock

7,321
10
55
75

That definitely isn't the issue. The section containing ' is working just fine, exactly the way I want it to. What's not working is `&`. – wwaawaw Sep 24 '12 at 17:59
Okay, maybe you just need to use more simple function like this: str = str.replace(/&/gi, function(m){ console.log(m); return m.replace(/&/, '&'); }); – Warlock Sep 24 '12 at 18:25

score 0 · Answer 2 · answered Sep 24 '12 at 17:55

0

This should do what you want:

str.replace(/&amp;([#x]\d+;|[a-z]+;)/g, "&$1")

or, with a positive lookahead:

str.replace(/&amp;(?=[#x]\d+;|[a-z]+;)/g, "&")

I don't think you need any HTML2text en-/decoding.

answered Sep 24 '12 at 17:55

Bergi

630,263
148
957
1,375

Why does this regex/DOM character entity tester return `undefined`?

2 Answers2