1145326 - String.prototype.normalize error when normalization form parameter is not an atom

Status: RESOLVED FIXED

(Core :: JavaScript: Internationalization API, defect)

Description

[Patrick Hall]

Attachment: unicodeNormalization.html

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:37.0) Gecko/20100101 Firefox/37.0
Build ID: 20150305191659

Steps to reproduce:

Assume a button with an id of "NFKC":

<button id="NFKC">NFKC</button>

Stick that string in a variable:

var s1 = document.body.querySelector('#NFKC').id

By way of comparison, assign the very same string to a variable directly:

var s2 = 'NFKC'

So of course:

s1 === s2
// true

And:

s1 == s2
// true

To normalize a string, you pass one of NFC, NFD, NFKC, or NFKD to .normalize(), like this:

'á'.normalize('NFKC')
// "á"

Depending on the normalization form you choose, you get different codepoints:

'á'.normalize('NFC').length == 1
// true
'á'.normalize('NFD').length == 2
// true




Actual results:


Since we know that s1 (the string we retrieved from the DOM) and s2 are THE SAME STRING (s1 === s2 is true), then obviously we can use either to normalize a string:

'á'.normalize(s2)
"á"
// well yeah, because s2 IS 'NFKC'. 

Naturally, s1 will behave exactly the same way, right?

'á'.normalize(s1)
 // RangeError: form must be one of 'NFC', 'NFD', 'NFKC', or 'NFKD'

Why the error here, but not with s2?


Expected results:

If I pass one of four strings corresponding to normalization forms to .normalize(), it should carry out normalization. Instead, I get a RangeError.

Comment 1

[Patrick Hall]

As pointed out on StackOverflow by Rick Hitchcock, the string doesn't have to be extracted from the DOM to cause the error:


    var s2 = 'NFKC'.split('').join('');
    var s2 = 'NFKCabc'.replace('abc','');

Rick points out that this version this doesn't: 

    var s2 = 'N'+'F'+'K'+'C';

http://stackoverflow.com/questions/29153533/when-a-string-is-not-a-string-unicode-normalization-weirdness-in-javascript#comment46529785_29153533

Comment 2

[Tom S [:evilpie]]

So the problem here is that the form parameter can be a "linear string", but not an atom. And of course if it's not an atom the pointer comparison to any atom is always going to fail.

Comment 3

[Tom S [:evilpie]]

Attachment: normalize

Comment 4

[Jeff Walden [:Waldo]]

Comment on attachment 8580637 [details] [diff] [review]
normalize

Review of attachment 8580637 [details] [diff] [review]:
-----------------------------------------------------------------

Oh fail, how did I review that.  :-(

Comment 5

[Tom S [:evilpie]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/8c968097ff98

Comment 6

[Phil Ringnalda (:philor)]

https://hg.mozilla.org/mozilla-central/rev/8c968097ff98