663132 - about:support opens up empty in my main profile

Status: RESOLVED DUPLICATE of bug 662901

(Toolkit :: General, defect)

Description

[(no longer active)]

Here's what I get in the error console:

Error: not well-formed
Source File: 
Line: 1, Column: 82
Source Code:
<td xmlns="http://www.w3.org/1999/xhtml">http://www.google.com/search?ie=UTF-8&oe=UTF-8&sourceid=navclient&gfns=1&q=</

Error: uncaught exception: [Exception... "An invalid or illegal string was specified"  code: "12" nsresult: "0x8053000c (NS_ERROR_DOM_SYNTAX_ERR)"  location: "chrome://global/content/aboutSupport.js Line: 350"]

Line 350 of aboutSupport.js maps to the second statement in this function:

function createElement(tagName, content, opt_class) {
  let elem = document.createElement(tagName);
  elem.innerHTML = content;
  elem.className = opt_class || "";
  return elem;
}

What's happening here is that createElement is trying to add a td element for my keyword.URL pref which has this value: "http://www.google.com/search?ie=UTF-8&oe=UTF-8&sourceid=navclient&gfns=1&q=", and this fails because we're in an XHTML document, which is not forgiving with those ampersand characters.

Comment 1

[(no longer active)]

Attachment: Patch (v1)

Comment 3

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

Both this bug and bug 662901 have patches to deal with this.

Probably only one should go forward.

Comment 4

[pal-moz]

dupe of bug 662901

Comment 6

[Dave Townsend [:mossop]]

Do we have no tests for about:support?

Comment 7

[:Gavin Sharp [email: [e-mail]]]

Comment on attachment 538266 [details] [diff] [review]
Patch (v1)

This breaks bug 660732 (which is what caused this bug in the first place), per bug 662901.

Comment 8

[:Gavin Sharp [email: [e-mail]]]

(In reply to comment #6)
> Do we have no tests for about:support?

Probably wouldn't have caught this since it only manifests itself if you have an extension or pref that's an invalid value for .innerHTML, but yeah, looks like we don't have any tests for it. Simple test that checks that it loads without throwing an uncaught exception would probably be good!

Comment 9

[:Gavin Sharp [email: [e-mail]]]

Blair's fix was better, but I think we probably should just be even more conservative here and handle that one specific case separately. We can add three strings (graphicsLinkPre, graphicsLink, graphicsLinkPost) and then construct the link HTML manually in that one case to avoid the possibility of a localization mistake causing trouble.