743573 - Build warnings in dom/

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Patch v1

Comment 1

[Mounir Lamouri (:mounir)]

Comment on attachment 613186 [details] [diff] [review]
Patch v1

Review of attachment 613186 [details] [diff] [review]:
-----------------------------------------------------------------

::: dom/base/nsDOMClassInfo.cpp
@@ +8795,5 @@
>    }
>  
> +  nsHTMLDocument* doc = GetDocument(obj);
> +  nsISupports* result = NULL;
> +  nsWrapperCache* cache = NULL;

Could you use nsnull instead?

::: dom/plugins/ipc/PluginModuleParent.cpp
@@ +585,5 @@
>  
>  PluginIdentifierParent*
>  PluginModuleParent::GetIdentifierForNPIdentifier(NPP npp, NPIdentifier aIdentifier)
>  {
> +    PluginIdentifierParent* ident = NULL;

nsnull?

Comment 2

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Patch v2

I don't remember what went wrong with the previous patch, but this passes try: <https://tbpl.mozilla.org/?tree=Try&rev=cc52db937c0f>.

Comment 3

[Mounir Lamouri (:mounir)]

Comment on attachment 640633 [details] [diff] [review]
Patch v2

Review of attachment 640633 [details] [diff] [review]:
-----------------------------------------------------------------

Unfortunately, I don't have cycles for that. Moving to khuey hoping he has some.

Comment 4

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Patch v3

Comment 5

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Patch v4

Comment 6

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Patch v5

Comment 7

[Mounir Lamouri (:mounir)]

Comment on attachment 648484 [details] [diff] [review]
Patch v5

Review of attachment 648484 [details] [diff] [review]:
-----------------------------------------------------------------

Does that make dom/ fully warning-free? If so, maybe we could add something in the Makfile rules that make dom/ subdirectories warning-free.

Comment 8

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: More fixes

The static assertions are because of clang errors like:

../../../dom/indexedDB/OpenDatabaseHelper.cpp:27:1: error: unused function 'pr_static_assert' [-Werror,-Wunused-function]
PR_STATIC_ASSERT(JS_STRUCTURED_CLONE_VERSION == 1);
^
../../dist/include/prtypes.h:535:17: note: expanded from macro 'PR_STATIC_ASSERT'
    extern void pr_static_assert(int arg[(condition) ? 1 : -1])
                ^
1 error generated.

Comment 9

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

https://hg.mozilla.org/mozilla-central/rev/1bbc0b65dffb

Comment 10

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

And I pushed

https://hg.mozilla.org/integration/mozilla-inbound/rev/08b1e1a4b851

with fixes from Aryeh.

Comment 11

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/08b1e1a4b851

Comment 12

[Jason Duell]

This is now causing 

dom/base/nsDOMClassInfo.cpp:9069:29: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]

Casting 32 bit int to void *: tsk, tsk...

Comment 13

[Jason Duell]

This is on a x86-64 bit linux system.  

I'm confused as to why the buildbots didn't catch this--do we not turn on warnings-as-errors there?

Comment 14

[Jason Duell]

These too:

dom/plugins/base/nsPluginNativeWindowGtk2.cpp:202:147: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]

dom/plugins/ipc/PluginInstanceParent.cpp:409:48: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]

Comment 15

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

I'll fix those in bug 781058.