getElementsByClass fails on mozilla richedit content

[werpu@…]

getElementsByClass function seems to work fine on the html editor richtext part in ie, but fails miserably to deliver anything in content within the mozilla rich text editor.

a href within the rich text content with a class="myFaces_SpellingError". The href is shown and triggerf

now if I trigger following code: var errorHrefs = dojo.html.getElementsByClass("myfaces_SpellingError", document,"*",0); errorHrefs is an empty array while on the same content the whole thing is returned back as an array with all the hrefs set to class="myfaces_SpellingError"

I am not sure if this is fixable, a workaround is to set the code directly within the href which has to be triggered instead of pushing it in afterwards. via dom element alteration, so that getElementsByClass calling is not needed. As I said I am not sure if this is fixable, but definitely worth an investigation.

[dylan]

Bryan, is this still an issue after your html refactor?

[dylan]

(In [6203]) references #456, references #959, temporary workaround until we figure out why the xpath version of getElementsByClass does not work 100% reliably in firefox

[bill]

Functionality replace by dojo.query()