Bug 116699 – [WK2][EFL][QT] Build break after r150610

• Component: WebKit2
• Status: RESOLVED FIXED

[Ryuan Choi]

EFL and QT using -Werror=maybe-uninitialized got below error.

/home/ryuan/workspace/project/webkit/efl-webkit/Source/WebKit2/UIProcess/WebPageProxy.cpp: In member function ‘void WebKit::WebPageProxy::findPlugin(const WTF::String&, uint32_t, const WTF::String&, const WTF::String&, const WTF::String&, bool, uint64_t&, WTF::String&, uint32_t&)’:
/home/ryuan/workspace/project/webkit/efl-webkit/Source/WebKit2/UIProcess/WebPageProxy.cpp:1440:155: error: ‘pluginProcessSandboxPolicy’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
cc1plus: all warnings being treated as errors
make[2]: *** [Source/WebKit2/CMakeFiles/WebKit2.dir/UIProcess/WebPageProxy.cpp.o] Error 1

[Ryuan Choi]

Created attachment 202758 [details]
Patch

[Anders Carlsson]

Comment on attachment 202758 [details]
Patch

This fix (while OK) has the downside of making the compiler not warn if we ever add more cases. I think a better fix is to just initialize the variable.

[Ryuan Choi]

Created attachment 202762 [details]
Patch

[Ryuan Choi]

(In reply to comment #2)
> (From update of attachment 202758 [details])
> This fix (while OK) has the downside of making the compiler not warn if we ever add more cases. I think a better fix is to just initialize the variable.

OK. I initialized the variable.

[Csaba Osztrogonác]

Comment on attachment 202762 [details]
Patch

LGTM, I think it's safe if we use stricter policy if the variable is uninitialized.

I'm not an owner, but I don't think if we should wait for 
an owner to get up for this trivial buildfix. So r=me. 

And feel free to find a better fix after you get up.

[Csaba Osztrogonác]

Comment on attachment 202762 [details]
Patch

Clearing flags on attachment: 202762

Committed r150624: <http://trac.webkit.org/changeset/150624>

[Csaba Osztrogonác]

All reviewed patches have been landed.  Closing bug.