Remove Gears from Chrome

Remove Gears from Chrome.

There are probably a few more bits and pieces that can be removed, but I think this is the majority of it.

BUG=51934
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=77888

[evanm, 2011-02-25 21:47:13]

LGTM

http://codereview.chromium.org/6576020/diff/5001/chrome/common/render_message...
File chrome/common/render_messages_params.h (right):

http://codereview.chromium.org/6576020/diff/5001/chrome/common/render_message...
chrome/common/render_messages_params.h:487: // TODO(aa): Can these be removed
now? Unclear if they are being used.
Why not?  I guess we can find it later by grepping again for Gears, but we
shouldn't close the Gears bug until we get all of these...

http://codereview.chromium.org/6576020/diff/5001/tools/emacs/trybot-mac.txt
File tools/emacs/trybot-mac.txt (left):

http://codereview.chromium.org/6576020/diff/5001/tools/emacs/trybot-mac.txt#o...
tools/emacs/trybot-mac.txt:1059: /Developer/usr/bin/distcc
/Developer/usr/bin/gcc-4.2 -x c++ -arch i386 -fmessage-length=0 -pipe
-Wno-trigraphs -fno-exceptions -fno-rtti -O0 -Werror -Wnewline-eof
-DCHROMIUM_BUILD -DENABLE_REMOTING=1 -DENABLE_GPU=1 -DNACL_WINDOWS=0
-DNACL_LINUX=0 -DNACL_OSX=1 -DNACL_TARGET_SUBARCH=32 -DNACL_BUILD_SUBARCH=32
-DCHROME_V8 -DGOOGLE_PROTOBUF_NO_RTTI -DXML_STATIC -DFEATURE_ENABLE_SSL
-DFEATURE_ENABLE_VOICEMAIL -DEXPAT_RELATIVE_PATH -DOSX -DPOSIX -DLIBXML_STATIC
-DHUNSPELL_STATIC -DHUNSPELL_CHROME_CLIENT -DUSE_HUNSPELL -D__STDC_FORMAT_MACROS
-DDYNAMIC_ANNOTATIONS_ENABLED=1 -D_DEBUG -isysroot
/Developer/SDKs/MacOSX10.5.sdk -fvisibility=hidden -fvisibility-inlines-hidden
-fno-threadsafe-statics -mmacosx-version-min=10.5 -gdwarf-2 -Wall -Wendif-labels
-Wextra -Wno-unused-parameter -Wno-missing-field-initializers -fpch-preprocess
-F/b/build/slave/mac/build/src/chrome/../xcodebuild/Debug
-F/Developer/SDKs/MacOSX10.5.sdk/System/Library/Frameworks/ApplicationServices.framework/Frameworks
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/Debug/include
-I../third_party/icu/public/common -I../third_party/icu/public/i18n -I..
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/chrome.build/DerivedSources/Debug
-I../third_party/apple -I../third_party/GTM -I../third_party/GTM/AppKit
-I../third_party/GTM/Foundation -I../third_party/GTM/DebugUtils -I../gpu
-I../third_party -I../third_party/mesa/MesaLib/include
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/DerivedSources/Debug/app
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/DerivedSources/Debug/chrome
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/DerivedSources/Debug/protoc_out
-I../third_party/protobuf -I../third_party/protobuf/src
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/DerivedSources/Debug/app/app_resources
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/DerivedSources/Debug/app/app_locale_settings
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/DerivedSources/Debug/app/app_strings
-I../skia/config -I../third_party/skia/include/config
-I../third_party/skia/include/core -I../third_party/skia/include/effects
-I../skia/ext -I../third_party/bzip2 -I../third_party/expat/files/lib
-I../third_party/libjingle/overrides -I../third_party/libjingle/source
-I../third_party/expat/files -I../third_party/libxml/mac/include
-I../third_party/libxml/src/include -I../third_party/npapi
-I../third_party/npapi/bindings -I../third_party/speex/include
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/DerivedSources/Debug/webkit
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/chrome.build/Debug/browser.build/DerivedSources/i386
-I/b/build/slave/mac/build/src/chrome/../xcodebuild/chrome.build/Debug/browser.build/DerivedSources
-g1 -c /b/build/slave/mac/build/src/chrome/browser/chrome_plugin_host.cc -o
/b/build/slave/mac/build/src/chrome/../xcodebuild/chrome.build/Debug/browser.build/Objects-normal/i386/chrome_plugin_host.o
heh

[jam, 2011-02-25 22:19:33]

http://codereview.chromium.org/6576020/diff/6138/content/browser/plugin_proce...
File content/browser/plugin_process_host.cc (left):

http://codereview.chromium.org/6576020/diff/6138/content/browser/plugin_proce...
content/browser/plugin_process_host.cc:73: const ViewHostMsg_Resource_Request&
resource_request) {
one minor change that could be done now: GetRequestContext can just take in the
process type, since that's the only information that's needed by
BrowserRenderProcessHost.

[Aaron Boodman, 2011-02-25 22:29:28]

ptal

http://codereview.chromium.org/6576020/diff/5001/chrome/common/render_message...
File chrome/common/render_messages_params.h (right):

http://codereview.chromium.org/6576020/diff/5001/chrome/common/render_message...
chrome/common/render_messages_params.h:487: // TODO(aa): Can these be removed
now? Unclear if they are being used.
On 2011/02/25 21:47:13, evanm wrote:
> Why not?  I guess we can find it later by grepping again for Gears, but we
> shouldn't close the Gears bug until we get all of these...

It seemed like maybe external automation clients could have been relying on it,
but I think I misread the code. Removed.

[Aaron Boodman, 2011-02-25 22:48:28]

http://codereview.chromium.org/6576020/diff/6138/content/browser/plugin_proce...
File content/browser/plugin_process_host.cc (left):

http://codereview.chromium.org/6576020/diff/6138/content/browser/plugin_proce...
content/browser/plugin_process_host.cc:73: const ViewHostMsg_Resource_Request&
resource_request) {
On 2011/02/25 22:19:33, John Abd-El-Malek wrote:
> one minor change that could be done now: GetRequestContext can just take in
the
> process type, since that's the only information that's needed by
> BrowserRenderProcessHost.

You mean "resource type"?  If so, done.

[jam, 2011-02-25 22:51:17]

On Fri, Feb 25, 2011 at 2:48 PM, <aa@chromium.org> wrote:

>
>
>
http://codereview.chromium.org/6576020/diff/6138/content/browser/plugin_proce...
> File content/browser/plugin_process_host.cc (left):
>
>
>
http://codereview.chromium.org/6576020/diff/6138/content/browser/plugin_proce...
> content/browser/plugin_process_host.cc:73: const
> ViewHostMsg_Resource_Request& resource_request) {
> On 2011/02/25 22:19:33, John Abd-El-Malek wrote:
>
>> one minor change that could be done now: GetRequestContext can just
>>
> take in the
>
>> process type, since that's the only information that's needed by
>> BrowserRenderProcessHost.
>>
>
> You mean "resource type"?  If so, done.


yep, sorry that's what I meant. lgtm


>
>
> http://codereview.chromium.org/6576020/
>

[Aaron Boodman, 2011-03-11 06:25:59]

+kuchhal

Rahul, can you please comment on the change to create_installer_archive.py? It
seems like it would work, but I'm not sure how to test it.

[kuchhal, 2011-03-11 19:10:35]

This is the diff I see:

--- a/chrome/tools/build/win/create_installer_archive.py +++
b/chrome/tools/build/win/create_installer_archive.py @@ -141,11 +141,11 @@
def GetPrevVersion(output_dir, temp_dir, last_chrome_installer):  lzma_exec
= GetLZMAExec(options.output_dir) prev_archive_file =
os.path.join(options.last_chrome_installer, - options.output_name +
ARCHIVE_SUFFIX) - cmd = '%s x -o"%s" "%s" Chrome-bin/*/gears.dll' %
(lzma_exec, temp_dir, - prev_archive_file) + *options.ogenereated
viutput_name* + ARCHIVE_SUFFIX) + cmd = '%s x -o"%s" "%s"
Chrome-bin/*/chrome.dll' % (lzma_exec, temp_dir, + prev_archive_file)

Looks like a typo to me.

The change looks OK in principle. We are trying to find out the last version
of chrome against which a diff installer will be generated. To test it - in
visual studio there used to be a property of mini_installer project where
you would give the directory that has chrome.7z of last chrome version
released. After that building mini_installer should exercise this code path.

On Thu, Mar 10, 2011 at 10:25 PM, <aa@chromium.org> wrote:

> +kuchhal
>
> Rahul, can you please comment on the change to create_installer_archive.py?
> It
> seems like it would work, but I'm not sure how to test it.
>
> http://codereview.chromium.org/6576020/
>

[Aaron Boodman, 2011-03-11 21:04:27]

On 2011/03/11 19:10:35, kuchhal wrote:
> This is the diff I see:
> 
> --- a/chrome/tools/build/win/create_installer_archive.py +++
> b/chrome/tools/build/win/create_installer_archive.py @@ -141,11 +141,11 @@
> def GetPrevVersion(output_dir, temp_dir, last_chrome_installer):  lzma_exec
> = GetLZMAExec(options.output_dir) prev_archive_file =
> os.path.join(options.last_chrome_installer, - options.output_name +
> ARCHIVE_SUFFIX) - cmd = '%s x -o"%s" "%s" Chrome-bin/*/gears.dll' %
> (lzma_exec, temp_dir, - prev_archive_file) + *options.ogenereated
> viutput_name* + ARCHIVE_SUFFIX) + cmd = '%s x -o"%s" "%s"
> Chrome-bin/*/chrome.dll' % (lzma_exec, temp_dir, + prev_archive_file)
> 
> Looks like a typo to me.
> 
> The change looks OK in principle. We are trying to find out the last version
> of chrome against which a diff installer will be generated. To test it - in
> visual studio there used to be a property of mini_installer project where
> you would give the directory that has chrome.7z of last chrome version
> released. After that building mini_installer should exercise this code path.

Oh, yeah. I fixed that typo. The question I had was about gears.dll->chrome.dll.

[kuchhal, 2011-03-11 21:10:40]

I think changing gears.dll to chrome.dll should be fine since the version is
read from the parent directory. I don't remember why we chose gears.dll but
it might have been due to smaller size and hence less time taken to extract
out of .7z archive.

On Fri, Mar 11, 2011 at 1:04 PM, <aa@chromium.org> wrote:

> On 2011/03/11 19:10:35, kuchhal wrote:
>
>> This is the diff I see:
>>
>
>  --- a/chrome/tools/build/win/create_installer_archive.py +++
>> b/chrome/tools/build/win/create_installer_archive.py @@ -141,11 +141,11 @@
>> def GetPrevVersion(output_dir, temp_dir, last_chrome_installer):
>>  lzma_exec
>> = GetLZMAExec(options.output_dir) prev_archive_file =
>> os.path.join(options.last_chrome_installer, - options.output_name +
>> ARCHIVE_SUFFIX) - cmd = '%s x -o"%s" "%s" Chrome-bin/*/gears.dll' %
>> (lzma_exec, temp_dir, - prev_archive_file) + *options.ogenereated
>> viutput_name* + ARCHIVE_SUFFIX) + cmd = '%s x -o"%s" "%s"
>> Chrome-bin/*/chrome.dll' % (lzma_exec, temp_dir, + prev_archive_file)
>>
>
>  Looks like a typo to me.
>>
>
>  The change looks OK in principle. We are trying to find out the last
>> version
>> of chrome against which a diff installer will be generated. To test it -
>> in
>> visual studio there used to be a property of mini_installer project where
>> you would give the directory that has chrome.7z of last chrome version
>> released. After that building mini_installer should exercise this code
>> path.
>>
>
> Oh, yeah. I fixed that typo. The question I had was about
> gears.dll->chrome.dll.
>
>
> http://codereview.chromium.org/6576020/
>