Check that processed i18n nodes are defined before marking them.

Check that processed i18n nodes are defined before marking them.

On iOS, |root.children| was undefined for IDR_NET_ERROR_HTML webui,
which prevented the further conversion of i18n nodes.

BUG=512085
Committed: https://crrev.com/b46a8bb14f620471292c2ce4c12e7694c9fd7595
Cr-Commit-Position: refs/heads/master@{#339792}

[kkhorimoto, 2015-07-21 21:21:45]

kkhorimoto@chromium.org changed reviewers:
+ dbeam@chromium.org, jyquinn@chromium.org

[kkhorimoto, 2015-07-21 21:21:45]

This fixes a bug introduced in https://codereview.chromium.org/1233883005/, and
described in crbug.com/512085.  For the error page webui, at some point |root|
is a DocumentFragment with undefined children, so accessing |processed.length|
caused an error that prevented conversion of the heading element.

[Dan Beam, 2015-07-21 21:26:21]

https://codereview.chromium.org/1246633007/diff/1/ui/webui/resources/js/i18n_...
File ui/webui/resources/js/i18n_template_no_process.js (right):

https://codereview.chromium.org/1246633007/diff/1/ui/webui/resources/js/i18n_...
ui/webui/resources/js/i18n_template_no_process.js:187: if (processed) {
I'm confused. if ([]) always triggers.  root.children should always be there. 
when is this actually happening?

[Dan Beam, 2015-07-21 21:28:18]

On 2015/07/21 21:21:45, kkhorimoto_ wrote:
> This fixes a bug introduced in https://codereview.chromium.org/1233883005/,
and
> described in crbug.com/512085.  For the error page webui, at some point |root|
> is a DocumentFragment with undefined children, so accessing |processed.length|
> caused an error that prevented conversion of the heading element.

ugh, yet another cross-platform difference

https://developer.mozilla.org/en-US/docs/Web/API/ParentNode/children#Browser_...

stop using our code

lgtm

[kkhorimoto, 2015-07-21 21:33:21]

The CQ bit was checked by kkhorimoto@chromium.org

[commit-bot: I haz the power, 2015-07-21 21:35:14]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1246633007/1

[commit-bot: I haz the power, 2015-07-22 00:12:29]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-07-22 00:13:21]

Patchset 1 (id:??) landed as
https://crrev.com/b46a8bb14f620471292c2ce4c12e7694c9fd7595
Cr-Commit-Position: refs/heads/master@{#339792}

[Dan Beam, 2015-07-22 00:20:40]

https://codereview.chromium.org/1246633007/diff/1/ui/webui/resources/js/i18n_...
File ui/webui/resources/js/i18n_template_no_process.js (right):

https://codereview.chromium.org/1246633007/diff/1/ui/webui/resources/js/i18n_...
ui/webui/resources/js/i18n_template_no_process.js:186: var processed = isElement
? [root] : root.children;
fwiw: this would probably work on iOS as well

  var processed = isElement ? [root] : root.childNodes;
  for (var i = 0; i < process.length; ++i) {
    if (processed.nodeType == Node.ELEMENT_NODE)
      processed[i].setAttribute('i18n-processed', '');
  }

or just

    if (processed instanceof Element)

[Dan Beam, 2015-07-22 00:21:16]

On 2015/07/22 00:20:40, Dan Beam wrote:
>
https://codereview.chromium.org/1246633007/diff/1/ui/webui/resources/js/i18n_...
> File ui/webui/resources/js/i18n_template_no_process.js (right):
> 
>
https://codereview.chromium.org/1246633007/diff/1/ui/webui/resources/js/i18n_...
> ui/webui/resources/js/i18n_template_no_process.js:186: var processed =
isElement
> ? [root] : root.children;
> fwiw: this would probably work on iOS as well
> 
>   var processed = isElement ? [root] : root.childNodes;
>   for (var i = 0; i < process.length; ++i) {
>     if (processed.nodeType == Node.ELEMENT_NODE)

if (processed[i].nodeType == Node.ELEMENT_NODE) **

>       processed[i].setAttribute('i18n-processed', '');
>   }
> 
> or just
> 
>     if (processed instanceof Element)

if (processed[i] instanceof Element) **