Problem Reading ShopRite's Website with Internet Explorer--It's Blank #5569

Closed
n2fgc opened this Issue Dec 3, 2015 · 7 comments

3 participants

@n2fgc

Good evening,

I tried to access the http://Shoprite.com website with Internet Explorer 11 and NVDA and came up blank. It started out in a search box, but when I tabbed out, all I heard all the way down the page was that lovely word, "Blank." In desperation, I quit NVDA and brought up another screen reader which read the site links and shoed me the page wasn't blank. I quit the other screen reader and started up NVDA; the links were no longer readable.

I've been given to understand that Firefox with NVDA works well on the site which is an okay work around, but for those who wouldn't know to try that solution, could you take a look at the site and help figure out how to make it and others like it work?

Thanks very much!

Lee

@jcsteh jcsteh added the bug label Dec 3, 2015
@jcsteh jcsteh added this to the 2016.1 milestone Dec 3, 2015
@jcsteh

Thanks for the report.

Technical: Wow. This site is broken. It looks like the author accidentally used the wrong quote symbol in the code, so the lang attribute on the html tag is the following:

en”>
<head>
    <meta http-equiv=

instead of just the intended en. Still, we need to be able to cope with such brokenness.

What I don't quite understand yet is why our XML escaping for attribute values isn't handling this.

@jcsteh jcsteh added a commit that referenced this issue Dec 3, 2015
@jcsteh jcsteh In Internet Explorer, the shoprite.com website can now be read in bro…
…wse mode instead of reporting as blank. (Specifically, malformed lang attributes are now handled gracefully.)

Normally, the lang attribute should only contain specific characters, but it's entirely possible for an author to break this as is the case on this site. We must XML escape the language to avoid complete breakage where this happens.
Fixes #5569.
771c32f
@jcsteh jcsteh added a commit that referenced this issue Dec 3, 2015
@jcsteh jcsteh In Internet Explorer, the shoprite.com website can now be read in bro…
…wse mode instead of reporting as blank. (Specifically, malformed lang attributes are now handled gracefully.)

Normally, the lang attribute should only contain specific characters, but it's entirely possible for an author to break this as is the case on this site. We must XML escape the language to avoid complete breakage where this happens.
Fixes #5569.
49f6833
@nvaccessAuto

Incubated in 046306f.

@jcsteh jcsteh was assigned by nvaccessAuto Dec 3, 2015
@jcsteh

What I don't quite understand yet is why our XML escaping for attribute values isn't handling this.

That'd be because the language attribute is handled separately for MSHTML (and wasn't previously being escaped).

@n2fgc
@n2fgc
@jcsteh
@n2fgc
@jcsteh jcsteh added a commit that closed this issue Dec 17, 2015
@jcsteh jcsteh In Internet Explorer, the shoprite.com website can now be read in bro…
…wse mode instead of reporting as blank. (Specifically, malformed lang attributes are now handled gracefully.)

Normally, the lang attribute should only contain specific characters, but it's entirely possible for an author to break this as is the case on this site. We must XML escape the language to avoid complete breakage where this happens.
Fixes #5569.
cb8b486
@jcsteh jcsteh closed this in cb8b486 Dec 17, 2015
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment