20663 - [MLK] Possible reference count leak

Status: VERIFIED FIXED

(Core :: Layout, defect, P3)

Description

[Patrick C. Beard]

In the URL I included above you see an NS_RELEASE(regSel);, but there isn't a
release done when you break out of the loop. Is this a refcount leak?

Comment 1

[Patrick C. Beard]

Fixing this keeps a nsJSEvenetListener object from leaking. Here's my patch:

Index: mozilla/layout/events/src/nsEventListenerManager.cpp
===================================================================
RCS file: /cvsroot/mozilla/layout/events/src/nsEventListenerManager.cpp,v
retrieving revision 1.47
diff -r1.47 nsEventListenerManager.cpp
211d210
<   nsIScriptEventListener* regSel;
224c223,225
<       if (NS_OK == ls->mListener->QueryInterface(kIScriptEventListenerIID,
(void**)&regSel)) {
---
>       nsresult rv;
>       nsCOMPtr<nsIScriptEventListener> regSel = do_QueryInterface(ls->mListener, &rv);
>       if (NS_SUCCEEDED(rv) && nsnull != regSel) {
231d231
<         NS_RELEASE(regSel);

Comment 2

[joki (gone)]

Okay, fix in.  Thanks Patrick.

Comment 3

[Chris Petersen]

Marking verified per last comments.

Comment 4

[Pulsebot]

Pushed by ffxbld-merge:
https://hg.mozilla.org/releases/mozilla-beta/rev/2606e1dc15e7
[fenix] Issue https://github.com/mozilla-mobile/fenix/issues/20663: Make inactive card collapsible