On Mon, 02 Mar 2009 15:31:16 +0100, Steve Block <[email protected]> wrote:

Thanks for the new patch Stig, but you haven't addressed any of my
original comments, copied below.

Sorry about that. Will do.

2009/1/27 Steve Block <[email protected]>:
Almost all of these changes have either already been made to the Gears
repository or are specific to Win32. I've marked the latter with 'Win32'. Please
sync to tip-of-tree before submitting the next set of patches.

========================================================================
http://mondrian.corp.google.com/file/9881868///depot/googleclient/gears/opensource/gears/Makefile?a=1
File //depot/googleclient/gears/opensource/gears/Makefile (snapshot 1)
------------------------------------
Line 31: # are not clobbered by the Chrome trickery at the end of the file
We've agreed not to make Win32-specific changes at this point.

These changes are needed by Opera for WinCE as well. When you do dynamic assignment (= operator) of the Chrome specific sources to the NPAPI sources the NPAPI values become a mix of NPAPI and Chrome code. We do not want the Chrome code in a Opera build so we need to statically assign (:= operator) the NPAPI values before Chrome clobbers them.

------------------------------------
Line 100: $(NULL)
Win32

Removed.

------------------------------------
Line 935: message_queue_cr.cc \
Win32

Moved from NPAPI to the Chrome section since it is chrome specific and not NPAPI specific.

------------------------------------
Line 969: detect_version_collision_win32.cc \
This is already added in base/opera.

Removed.

------------------------------------
Line 971: time_utils_win32.cc \
These two are already added in base/common.

Removed.

------------------------------------
Line 992: $(OPERA_OUTDIR)/setup.res \
This is already added in installer/opera.

Removed.

------------------------------------
Line 1002: paths_op.cc \
These three are already added in base/opera.

Removed.

------------------------------------
Line 1901: opera_local_server_impl.cc \
This has yet to be added to the repository, so this will break the build.

It was included in the localserver patch with the http_request implementation.

------------------------------------
Line 1902: safe_http_request.cc \
This is already added in localserver/common.

Removed.

========================================================================
http://mondrian.corp.google.com/file/9881868///depot/googleclient/gears/opensource/gears/Makefile?a=0
File //depot/googleclient/gears/opensource/gears/Makefile
------------------------------------
Line 2220: ifeq ($(OS),wince)
Win32, though we could remove the BROWSER==OPERA guard.

I'm not sure what part of the code this refers to. Take a look in the attached patch and please give a new reference to it.


========================================================================
http://mondrian.corp.google.com/file/9881868///depot/googleclient/gears/opensource/gears/tools/config.mk?a=1
File //depot/googleclient/gears/opensource/gears/tools/config.mk (snapshot 1)
------------------------------------
Line 927: OPERA_LIBS = delayimp.lib /DELAYLOAD:"comdlg32.dll" comdlg32.lib
Win32

Removed.

------------------------------------
Line 988: M4FLAGS  += -DPRODUCT_BROWSER=$(BROWSER)
Replicated below.

Removed.

========================================================================
http://mondrian.corp.google.com/file/9881868///depot/googleclient/gears/opensource/gears/tools/rules.mk?a=1
File //depot/googleclient/gears/opensource/gears/tools/rules.mk (snapshot 1)
------------------------------------
Line 260: OPERA_INSTALLER_MSI    = $(INSTALLERS_OUTDIR)/$(INSTALLER_BASE_NAME)-opera.msi
Win32

Removed.

------------------------------------
Line 326: $(MAKE) modules    BROWSER=OPERA
Win32

Removed.

========================================================================
http://mondrian.corp.google.com/file/9881868///depot/googleclient/gears/opensource/gears/tools/rules.mk?a=0
File //depot/googleclient/gears/opensource/gears/tools/rules.mk
------------------------------------
Line 430: # desktop.
Win32

Removed

========================================================================
http://mondrian.corp.google.com/file/9881868///depot/googleclient/gears/opensource/gears/tools/rules.mk?a=1
File //depot/googleclient/gears/opensource/gears/tools/rules.mk (snapshot 1)
------------------------------------
Line 482: installers:: $(FFMERGED_INSTALLER_XPI) $(WIN32_INSTALLER_MSI) $(NPAPI_INSTALLER_MSI) $(OPERA_INSTALLER_MSI)
Win32

Removed

------------------------------------
Line 782: $(RC) $(RCFLAGS) /DBROWSER_OPERA=1 /DBROWSER_NPAPI=1 $<
I don't think that this is needed?

Removed

------------------------------------
Line 858: $(shell $(GGUIDGEN) $(NAMESPACE_GUID) OUR_COMPONENT_GUID_OPERA_REGISTRY-$(VERSION))
Win32

Removed

------------------------------------
Line 894: -dOurComponentGUID_OperaRegistry=$(OUR_COMPONENT_GUID_OPERA_REGISTRY) \
Win32

Removed

------------------------------------
Line 1296: $(NULL)
These files are already listed in OPERA_RESOURCES.

Removed

------------------------------------
Line 1299: cp $< $(OPERA_OUTDIR)/genfiles
We don't want to copy all files from ui/common. The rule is already present at
line 785.

Removed

------------------------------------
Line 1301: $(OPERA_WINCE_INSTALLER_CAB): $(OPERA_INFSRC) $(OPERA_MODULE_DLL) $(OPERA_WINCESETUP_DLL) $(patsubst %,$(OPERA_OUTDIR)/genfiles/%,$(SOURCE_HTML_IMG_FILES))
What about the dependency on the HTML files?

Done

------------------------------------
Line 1315: light.exe -out $@ $(OPERA_INSTALLER_WIXOBJ) -sice:ICE39 -sice:ICE64 -sice:ICE91
Win32

Removed

========================================================================
http://mondrian.corp.google.com/file/9881868///depot/googleclient/gears/opensource/gears/tools/sanity.mk?a=1
File //depot/googleclient/gears/opensource/gears/tools/sanity.mk (snapshot 1)
------------------------------------
Line 44: ifneq ($(BROWSER),OPERA)
Win32.

Removed


--
Stig Halvorsen
Senior Core Developer
Opera Software ASA

Attachment: makefiles3.patch
Description: Binary data

Reply via email to