385731 - ensure nsAccessibilityService::GetAccessible doesn't fail if there is no accessible for the given node

Status: RESOLVED WORKSFORME

(Core :: Disability Access APIs, defect)

Description

[alexander :surkov (:asurkov)]

Comment 1

[alexander :surkov (:asurkov)]

<aaronlev> surkov, i mean, if someone asks for an accessible object for a regular <span> object or something
<aaronlev> we will return null for the accessible and NS_ERROR_FAILURE
<aaronlev> but, an error should usually be for something not expected
<aaronlev> however, it's very expected that someone will ask for a given node if ther eis an accessible object for it
<aaronlev> technically we should only return an error for things that are unexpected

Comment 2

[alexander :surkov (:asurkov)]

initial work was done in bug 347019.

Comment 3

[alexander :surkov (:asurkov)]

(In reply to comment #7 bug 347019)
> Surkov, didn't we need to return NS_OK here?
> http://lxr.mozilla.org/seamonkey/source/accessible/src/base/nsAccessibilityService.cpp#1131
> 
> 1130   if (!aAccessibleIn) {
> 1131     return NS_ERROR_FAILURE; // No accessible to init
> 1132   }
> 

Either we should use NS_ENSURE_ARG() or return NS_OK. Now we do not check whether aAccessibleIn is null when we call InitAccessible(), probably we should because it doesn't make sense to call it with nsnull.

Comment 4

[Aaron Leventhal]

> it doesn't make sense to call it with nsnull.
Okay.

I think we used to have more places where we called it, and it made the code a little easier to read at that time. But I agree with you.

Comment 5

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

Mass un-assigning bugs assigned to Aaron.

Comment 6

[alexander :surkov (:asurkov)]

It doesn't fail anymore, was fixed at some point. Closing as worksforme.