776481 - accessible/src/html/HTMLTableAccessible.cpp(1208) : warning C4305: 'return' : truncation from 'nsresult' to 'bool'

Status: RESOLVED FIXED

(Core :: Disability Access APIs, defect)

Description

[Makoto Kato [:m_kato]]

don't return NS_ERROR_xxxx and NS_OK for bool.

Comment 1

[Makoto Kato [:m_kato]]

Attachment: fix

Comment 2

[Trevor Saunders (:tbsaunde)]

Comment on attachment 644889 [details] [diff] [review]
fix


>   nsresult rv = GetCellAt(0, 0, *getter_AddRefs(cellElement));
>-  NS_ENSURE_SUCCESS(rv, NS_ERROR_FAILURE);
>+  NS_ENSURE_SUCCESS(rv, false);
> 
>   nsCOMPtr<nsIContent> cellContent(do_QueryInterface(cellElement));
>-  NS_ENSURE_TRUE(cellContent, NS_ERROR_FAILURE);
>+  NS_ENSURE_TRUE(cellContent, false);

so, in an ideal world we'd use the RETURN_LAYOUT_ANSWEr() macro, or really just make this stuff infalable, but because of the warnings that's tricky so these seem fine.

>   nsIFrame *cellFrame = cellContent->GetPrimaryFrame();
>   if (!cellFrame) {
>-    return NS_OK;
>+    return true;

it seems you should return false here too, why don't oyu? also using the macro here is probably fine.

Comment 3

[alexander :surkov (:asurkov)]

Falable is fine. I agree, those should be false since we had false as default value (see http://hg.mozilla.org/mozilla-central/annotate/de43aa36ceef/accessible/src/html/nsHTMLTableAccessible.cpp#l1361).

Comment 5

[Aryeh Gregor (:ayg) (no longer with Mozilla)]

Attachment: Patch

Note that this changes the return value from true to false in the first two cases.  The first of those can't be easily made infallible: it might fail in a lot of ways.  The second should really be infallible in principle, but currently we have no way to make it so in practice.  (Really what we'd want is a method nsIDOMElement::AsNative or something that returns dom::Element, but we don't have one yet.)

Comment 6

[Trevor Saunders (:tbsaunde)]

Comment on attachment 652451 [details] [diff] [review]
Patch

r=me fwiw I'll be redoing all that in bug 781409 rsn anyway :/

Comment 7

[Aryeh Gregor (:ayg) (no longer with Mozilla)]

Green try except for Windows: https://tbpl.mozilla.org/?tree=Try&rev=fc3eb12e62f4  Windows isn't included because I'm trying to fix Windows build errors in other patches in the series, but it shouldn't behave any differently on Windows.  (The patch isn't listed at the left, you have to go up to the parent several times to find it.)

https://hg.mozilla.org/integration/mozilla-inbound/rev/e9e13da6a652

Comment 8

[Phil Ringnalda (:philor)]

https://hg.mozilla.org/mozilla-central/rev/e9e13da6a652