500202 - Unused variable build warnings.

Status: RESOLVED FIXED

(Core :: Disability Access APIs, defect)

Description

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

Compiler complains about unused variables.

Comment 1

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

Attachment: trivial fix.

Comment 2

[Marco Zehe (:MarcoZ)]

Comment on attachment 384876 [details] [diff] [review]
trivial fix.

trivial indeed, r=me. Thanks!

Comment 3

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

pushed: http://hg.mozilla.org/mozilla-central/rev/113c5916881f

Comment 4

[alexander :surkov (:asurkov)]

I think this requires more investigations because we shouldn't hide errors in general.

Comment 5

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

We aren't hiding errors here. The return value rv was stored in a variable, but the variable was unused in that code path. I think the compiler warning and the fix are valid.

Comment 6

[alexander :surkov (:asurkov)]

(In reply to comment #5)
> We aren't hiding errors here. The return value rv was stored in a variable, but
> the variable was unused in that code path. I think the compiler warning and the
> fix are valid.

I mean it's wrong nsresult variable wan't used - this is what I call hiding errors.

Comment 7

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

Aha I see what you mean and I generally agree. In this case I was thinking the aName.IsEmpty check is sufficient. I suppose we could check the return value and ASSERT but is that what we want here?

Comment 8

[alexander :surkov (:asurkov)]

I think NS_ENSURE_SUCCESS should work for us. We could use ASSERT if we're sure it won't happen. Though it's probably worth to have special case for xblAccessible->GetAccessibleName(aName) to not propagate XBL's mistakes. However if XBL's accessibleName throws Components.results.ERROR then we should use it.