Encountering compile errors w/ blink_platform_impl.cc with Debug builds

113 views
Skip to first unread message

dhn...@chromium.org

unread,
Jul 25, 2014, 1:46:00 PM7/25/14
to chromi...@chromium.org
Is anyone else experiencing these compile errors? I've tried clobbering, but it was unsuccessful.

I am not encountering issue when building a Release build.
 
../../content/child/blink_platform_impl.cc:759:34: error: use of undeclared identifier 'IDR_INSPECTOR_OVERLAY_PAGE_HTML'
  { "InspectorOverlayPage.html", IDR_INSPECTOR_OVERLAY_PAGE_HTML,
                                 ^
../../content/child/blink_platform_impl.cc:762:5: error: use of undeclared identifier 'IDR_INSPECTOR_INJECTED_SCRIPT_CANVAS_MODULE_SOURCE_JS'
    IDR_INSPECTOR_INJECTED_SCRIPT_CANVAS_MODULE_SOURCE_JS,
    ^
../../content/child/blink_platform_impl.cc:764:32: error: use of undeclared identifier 'IDR_INSPECTOR_INJECTED_SCRIPT_SOURCE_JS'
  { "InjectedScriptSource.js", IDR_INSPECTOR_INJECTED_SCRIPT_SOURCE_JS,
                               ^
../../content/child/blink_platform_impl.cc:766:32: error: use of undeclared identifier 'IDR_INSPECTOR_DEBUGGER_SCRIPT_SOURCE_JS'
  { "DebuggerScriptSource.js", IDR_INSPECTOR_DEBUGGER_SCRIPT_SOURCE_JS,
                               ^
../../content/child/blink_platform_impl.cc:799:26: error: no matching function for call to 'ArraySizeHelper'
  for (size_t i = 0; i < arraysize(kDataResources); ++i) {
                         ^~~~~~~~~~~~~~~~~~~~~~~~~
../../base/macros.h:74:34: note: expanded from macro 'arraysize'
#define arraysize(array) (sizeof(ArraySizeHelper(array)))
                                 ^~~~~~~~~~~~~~~
../../base/macros.h:64:8: note: candidate template ignored: could not match 'T [N]' against 'const content::(anonymous namespace)::DataResource []'
char (&ArraySizeHelper(T (&array)[N]))[N];
       ^
../../base/macros.h:71:8: note: candidate template ignored: could not match 'const T [N]' against 'const content::(anonymous namespace)::DataResource []'
char (&ArraySizeHelper(const T (&array)[N]))[N];
       ^


Thiago Farina

unread,
Jul 25, 2014, 2:26:11 PM7/25/14
to dhn...@chromium.org, Chromium-dev
This is probably my change.

But should be easy to fix.

Sorry for the breakage. 

--
Thiago Farina

Zhenyao Mo

unread,
Jul 25, 2014, 3:32:05 PM7/25/14
to Thiago Farina, dhn...@chromium.org, Chromium-dev
My local linux build also failed to compile because of this.  It the fix landing soon?

Why it's not caught by any bots?


--
--
Chromium Developers mailing list: chromi...@chromium.org
View archives, change email options, or unsubscribe:
http://groups.google.com/a/chromium.org/group/chromium-dev

To unsubscribe from this group and stop receiving emails from it, send an email to chromium-dev...@chromium.org.

Robert Shield

unread,
Jul 25, 2014, 3:44:46 PM7/25/14
to Zhenyao Mo, Thiago Farina, dhn...@chromium.org, Chromium-dev
I hit this earlier and fixed it by searching my out/ dir and finding two generated blink_resources.h files, one which had the new symbols and one which didn't. 

I deleted the one that didn't and the build succeeded. Afaict, how these files are generated changed, but if you don't do a full clobber build, the old blink_resources.h file will stick around and I guess it is found first causing this problem. 

mich...@chromium.org

unread,
Jul 25, 2014, 5:56:03 PM7/25/14
to chromi...@chromium.org
This worked for me too. Specifically:

rm out/Debug/gen/webkit/grit/blink_resources.h

To unsubscribe from this group and stop receiving emails from it, send an email to chromium-dev+unsubscribe@chromium.org.

Thiago Farina

unread,
Jul 25, 2014, 6:51:53 PM7/25/14
to Michael Giuffrida, Chromium-dev
On Fri, Jul 25, 2014 at 6:56 PM, <mich...@chromium.org> wrote:
This worked for me too. Specifically:

rm out/Debug/gen/webkit/grit/blink_resources.h


On Friday, July 25, 2014 12:44:46 PM UTC-7, Robert Shield wrote:
I hit this earlier and fixed it by searching my out/ dir and finding two generated blink_resources.h files, one which had the new symbols and one which didn't. 

I deleted the one that didn't and the build succeeded. Afaict, how these files are generated changed, but if you don't do a full clobber build, the old blink_resources.h file will stick around and I guess it is found first causing this problem. 

Yeah, thanks for sharing the fix.

This is because the new file lives in out/Debug/gen/blink/public/resources.

I apologize everyone affected by my change. Hopefully the fix was something simple and quick.

--
Thiago Farina

Kenneth Russell

unread,
Jul 25, 2014, 7:13:07 PM7/25/14
to Thiago Farina, Michael Giuffrida, Chromium-dev
A separate PSA to chromium-dev would be helpful. Also in the future
please try to avoid requiring manual intervention. Even a temporary
build rule to handle renaming of a generated file would be better than
breakage.

-Ken

Thiago Farina

unread,
Jul 25, 2014, 7:35:07 PM7/25/14
to Kenneth Russell, Michael Giuffrida, Chromium-dev
On Fri, Jul 25, 2014 at 8:12 PM, Kenneth Russell <k...@chromium.org> wrote:
On Fri, Jul 25, 2014 at 3:51 PM, Thiago Farina <tfa...@chromium.org> wrote:
>
>
>
> On Fri, Jul 25, 2014 at 6:56 PM, <mich...@chromium.org> wrote:
>>
>> This worked for me too. Specifically:
>>
>> rm out/Debug/gen/webkit/grit/blink_resources.h
>>
>>
>>
>> On Friday, July 25, 2014 12:44:46 PM UTC-7, Robert Shield wrote:
>>>
>>> I hit this earlier and fixed it by searching my out/ dir and finding two
>>> generated blink_resources.h files, one which had the new symbols and one
>>> which didn't.
>>>
>>> I deleted the one that didn't and the build succeeded. Afaict, how these
>>> files are generated changed, but if you don't do a full clobber build, the
>>> old blink_resources.h file will stick around and I guess it is found first
>>> causing this problem.
>>>
> Yeah, thanks for sharing the fix.
>
> This is because the new file lives in out/Debug/gen/blink/public/resources.
>
> I apologize everyone affected by my change. Hopefully the fix was something
> simple and quick.

A separate PSA to chromium-dev would be helpful. Also in the future
please try to avoid requiring manual intervention. Even a temporary
build rule to handle renaming of a generated file would be better than
breakage.

Yeah, I fully agree. I will certainly do a PSA in future for patches like that. I know it is not an excuse, but I was not aware that it could happen and thus would require that manual fix, certainly my bad, and I will try to remember that in future.

--
Thiago Farina

Kenneth Russell

unread,
Jul 25, 2014, 8:14:51 PM7/25/14
to Thiago Farina, Michael Giuffrida, Chromium-dev
Completely understood. Your patches went through the CQ, and it's
impossible to predict all failure modes. Still, if you wouldn't mind
sending a separate PSA to chromium-dev now, indicating people will
need to do a clobber build if they encounter this, I think it would
still be useful. (Something like "PSA: clobber build needed for
IDR_INSPECTOR_* compile errors in blink_platform_impl.cc")

Thanks,

-Ken

Thiago Farina

unread,
Jul 25, 2014, 8:21:15 PM7/25/14
to Kenneth Russell, Michael Giuffrida, Chromium-dev
On Fri, Jul 25, 2014 at 9:14 PM, Kenneth Russell <k...@chromium.org> wrote:
Completely understood. Your patches went through the CQ, and it's
impossible to predict all failure modes. Still, if you wouldn't mind
sending a separate PSA to chromium-dev now, indicating people will
need to do a clobber build if they encounter this, I think it would
still be useful. (Something like "PSA: clobber build needed for
IDR_INSPECTOR_* compile errors in blink_platform_impl.cc")

Sure thing. Sent.

--
Thiago Farina

Paweł Hajdan, Jr.

unread,
Jul 28, 2014, 5:13:40 AM7/28/14
to Thiago Farina, Robert Iannucci, Kenneth Russell, Michael Giuffrida, Chromium-dev
Shouldn't a landmine be used for this? Please see http://www.chromium.org/Home/chromium-clobber-landmines

Paweł


--

Jochen Eisinger

unread,
Jul 28, 2014, 8:41:31 AM7/28/14
to Paweł Hajdan, Jr., Thiago Farina, Robert Iannucci, Kenneth Russell, Michael Giuffrida, Chromium-dev
landmines don't clobber developer checkouts (it's compile.py that does it)

best
-jochen
Reply all
Reply to author
Forward
0 new messages