473569 - Restrict text-position to allowed values

Status: RESOLVED FIXED

(Core :: Disability Access APIs, defect)

Description

[Aaron Leventhal]

IA2 text-position officially can only be baseline, super and sub. Currently Firefox will expose anything that the author specified in vertical-align, including top, text-top, middle, bottom, text-bottom, length, %

Comment 1

[David Bolter [:davidb] (NeedInfo me for attention)]

Will this require a conversion then?  Sounds like new territory.

Comment 2

[alexander :surkov (:asurkov)]

Attachment: patch

Comment 3

[Trevor Saunders (:tbsaunde)]

Comment on attachment 601286 [details] [diff] [review]
patch

add bug to test file

Comment 4

[Trevor Saunders (:tbsaunde)]

ftr we should consider if things like NS_STYLE_VERTICAL_ALIGN_TOP should be consider super or something else.

Comment 5

[alexander :surkov (:asurkov)]

(In reply to Trevor Saunders (:tbsaunde) from comment #4)
> ftr we should consider if things like NS_STYLE_VERTICAL_ALIGN_TOP should be
> consider super or something else.

the best option I could think of is to ask IA2 group for opinions.

Comment 6

[Pete Brunet]

Super/base/sub conveys a certain meaning.  Are there some other meanings that are meant to be conveyed by the various other options?  If so, we can adjust the IA2 spec.

Comment 7

[alexander :surkov (:asurkov)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/f419ad71f1cd

Comment 8

[alexander :surkov (:asurkov)]

(In reply to Pete Brunet from comment #6)
> Super/base/sub conveys a certain meaning.  Are there some other meanings
> that are meant to be conveyed by the various other options?  If so, we can
> adjust the IA2 spec.

CSS vertical-align can take several values (see bug description and http://www.w3schools.com/cssref/pr_pos_vertical-align.asp). This patch maps baseline, sub and super values and numeric values to IA2. We don't map things like middle, top, etc to IA2. If AT would need other values then we should extend the list of allowed IA2 values. But then how we should map numeric values to enum list. Pete, should I send a question to IA2 list?

Comment 9

[Pete Brunet]

Sub and super convey meaning.  Is any meaning conveyed by the other options?  Or are those other options only for visual effect (artistry).

Comment 10

[alexander :surkov (:asurkov)]

(In reply to Pete Brunet from comment #9)
> Sub and super convey meaning.  Is any meaning conveyed by the other options?
> Or are those other options only for visual effect (artistry).

maybe they are rather for visual effect. But who can say certainly what authors on the web do.

Comment 11

[Pete Brunet]

The IA2 group likely won't be able to help answer that question.  What group can be consulted to get a feel for this?

Comment 12

[alexander :surkov (:asurkov)]

(In reply to Pete Brunet from comment #11)
> The IA2 group likely won't be able to help answer that question.  What group
> can be consulted to get a feel for this?

I'm not sure that any group can answer that but we might try to get an opinion from web accessibility people like Steve.

Comment 13

[steve faulkner]

(In reply to alexander :surkov from comment #12)
> (In reply to Pete Brunet from comment #11)
> > The IA2 group likely won't be able to help answer that question.  What group
> > can be consulted to get a feel for this?
> 
> I'm not sure that any group can answer that but we might try to get an
> opinion from web accessibility people like Steve.

I would be wary of assigning meaning based on CSS values, including super and sub, the meaning is meant to be assigned using the appropraite HTML elements (http://dev.w3.org/html5/spec/the-sub-and-sup-elements.html#the-sub-and-sup-elements)

Comment 14

[Matt Brubeck (:mbrubeck)]

https://hg.mozilla.org/mozilla-central/rev/f419ad71f1cd

Comment 15

[alexander :surkov (:asurkov)]

(In reply to steve faulkner from comment #13)

> I would be wary of assigning meaning based on CSS values, including super
> and sub, the meaning is meant to be assigned using the appropraite HTML
> elements
> (http://dev.w3.org/html5/spec/the-sub-and-sup-elements.html#the-sub-and-sup-
> elements)

Ok, let's see if CSS styles causes problems on the web. I filed bug 735645 for sup and sub elements.