Add breakpad support for me2me and it2me native messaging host on Mac.

Add breakpad support for me2me and it2me native messaging host on Mac.

Changes in this CL include:
1. Initialize breakpad in it2me native messaging host
2. Create bundles for it2me and me2me native messaging hosts and move them out of the me2me host bundle (and update the manifests accordingly).
3. Dump symbols of the it2me and me2me native messaging hosts
4. Sign the me2me and it2me native messaging hosts
4. Rename the me2me host bundle and the prefPane bundle to get rid of the internal chromium/chromoting names and to conform to Mac convention
5. Refactor GYP, build-installer-archive.py, and do_signing.sh, etc to use name constants from branding_chrome and branding_chromium instead of hard coded names. This ensures that all the name constants are defined in a central location and reduces duplicate name definitions. 

Here are the names of the host bundles:

me2me host:
old name: org.chromium.chromoting.me2me_host
new name: ChromeRemoteDesktop.Host

Pref pane:
old name: org.chromium.chromoting.prefPane
new name: ChromeRemoteDesktop.prefPane

it2me nm host:
name: NativeMessagingHost 
Full Path: /Library/Google/ChromeRemoteDesktop/NativeMessagingHost.app

me2me nm host:
name: RemoteAssistanceHost 
Full path: /Library/Google/ChromeRemoteDesktop/RemoteAssistanceHost.app

BUG=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=259287

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=260075

[weitao, 2014-03-06 18:54:34]

Ideally I should separate the renaming and refactoring changes from the breakpad
change. But that is not trivial to do as these changes are all tightly tangled
together. And I want to minimize the number we update do_signing.sh to reduce
the change of bot failures.

I made up the new host bundle names based on my observation of the convention
used on my macbook. Please review them and let me know if you prefer different
names. We certainly don't want to rename them too many times.

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome
File remoting/branding_Chrome (right):

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome...
remoting/branding_Chrome:9: MAC_HOST_BUNDLE_NAME=ChromeRemoteDesktop.Host
Any comment on the bundle names?

https://codereview.chromium.org/186763004/diff/40001/remoting/host/installer/...
File remoting/host/installer/mac/ChromotingHostService.pkgproj (right):

https://codereview.chromium.org/186763004/diff/40001/remoting/host/installer/...
remoting/host/installer/mac/ChromotingHostService.pkgproj:324:
<string>PreferencePanes/@@PREFPANE_BUNDLE_NAME@@</string>
Why does the pkgproj files use tabs instead of spaces? Should I clean them up
(in a separate CL of course)?

https://codereview.chromium.org/186763004/diff/40001/remoting/tools/get_mac_c...
File remoting/tools/get_mac_crd_version.sh (right):

https://codereview.chromium.org/186763004/diff/40001/remoting/tools/get_mac_c...
remoting/tools/get_mac_crd_version.sh:7:
ME2ME_HOST="/Library/PrivilegedHelperTools/ChromeRemoteDesktop.Host.app"
This tool doesn't seem to be used anywhere so I didn't spend time enhance it to
include support for the chromium build.

The path to the official installation is hardcoded here. So it won't be able to
find the version of the local build (Chromoting instead of ChromeRemoteDesktop)
if one is installed.

[weitao, 2014-03-06 18:56:24]

Please disregard the previous incorrectly titled CR request email. :)
 
Ideally I should separate the renaming and refactoring changes from the breakpad
change. But that is not trivial to do as these changes are all tightly tangled
together. And I want to minimize the number we update do_signing.sh to reduce
the change of bot failures.

I made up the new host bundle names based on my observation of the convention
used on my macbook. Please review them and let me know if you prefer different
names. We certainly don't want to rename them too many times.

[weitao, 2014-03-06 19:17:07]

Forgot to mention this:

I don't believe signing script on Google3 needs to be updated. It only
references the zip archive name (remoting-me2me-host-mac) which I didn't change.
Let me know if you think otherwise.

On 2014/03/06 18:56:24, weitaosu wrote:
> Please disregard the previous incorrectly titled CR request email. :)
>  
> Ideally I should separate the renaming and refactoring changes from the
breakpad
> change. But that is not trivial to do as these changes are all tightly tangled
> together. And I want to minimize the number we update do_signing.sh to reduce
> the change of bot failures.
> 
> I made up the new host bundle names based on my observation of the convention
> used on my macbook. Please review them and let me know if you prefer different
> names. We certainly don't want to rename them too many times.

[weitao, 2014-03-08 00:20:24]

ping... ...

[Sergey Ulanov, 2014-03-08 02:37:55]

That's awesome stuff!!! Thank you for working on this!

I think you also need to update
remoting/host/installer/mac/PrivilegedHelperTools/org.chromium.chromoting.me2me.sh
to make sure it runs the host from the right location.

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome
File remoting/branding_Chrome (right):

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome...
remoting/branding_Chrome:9: MAC_HOST_BUNDLE_NAME=ChromeRemoteDesktop.Host
On 2014/03/06 18:54:35, weitaosu wrote:
> Any comment on the bundle names?

Why does it have .Host at the end. I think it can be ChromeRemoteDesktopHost.
Also it looks it's common to use '.bundle' extensions for bundles that are not
"Apps" (and host bundle is not an app, it's not normally started by a user), see
http://en.wikipedia.org/wiki/Application_bundle. So I think this should be
"ChromeRemoteDesktopHost.bundle"

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome...
remoting/branding_Chrome:11:
MAC_NATIVE_MESSAGING_HOST_BUNDLE_NAME=NativeMessagingHost
NativeMessagingHost.bundle
But I'm not sure about this one. it can be something like
"RemoteDesktopNMHost.bundle"

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome...
remoting/branding_Chrome:15:
MAC_REMOTE_ASSISTANCE_HOST_BUNDLE_NAME=RemoteAssistanceHost
RemoteAssistanceHost.bundle

https://codereview.chromium.org/186763004/diff/40001/remoting/remoting_host_m...
File remoting/remoting_host_mac.gypi (right):

https://codereview.chromium.org/186763004/diff/40001/remoting/remoting_host_m...
remoting/remoting_host_mac.gypi:92: 'native_messaging_host_bundle_name':
'<!(python <(version_py_path) -f <(branding_path) -t
"@MAC_NATIVE_MESSAGING_HOST_BUNDLE_NAME@")',
Can we put the NM host bundle inside the me2me bundle (e.g. in
Content/Resources/NativeMessagingHost.bundle)? seems like it would be the right
place for it.

[weitao, 2014-03-19 07:22:36]

PTAL.

1. Feedback on bundle names and locations addressed
2. Fixed the uninstaller
3. Updated the references I missed in the previous revision.
3. renamed more mac files to replace "org.chromium.chromoting" with
"com.google.chromeremotedesktop": the .sh file, the log file, etc.

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome
File remoting/branding_Chrome (right):

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome...
remoting/branding_Chrome:9: MAC_HOST_BUNDLE_NAME=ChromeRemoteDesktop.Host
On 2014/03/08 02:37:56, Sergey Ulanov wrote:
> On 2014/03/06 18:54:35, weitaosu wrote:
> > Any comment on the bundle names?
> 
> Why does it have .Host at the end. I think it can be ChromeRemoteDesktopHost.
> Also it looks it's common to use '.bundle' extensions for bundles that are not
> "Apps" (and host bundle is not an app, it's not normally started by a user),
see
> http://en.wikipedia.org/wiki/Application_bundle. So I think this should be
> "ChromeRemoteDesktopHost.bundle"

Done.

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome...
remoting/branding_Chrome:11:
MAC_NATIVE_MESSAGING_HOST_BUNDLE_NAME=NativeMessagingHost
Changed the extension from ".app" to ".bundle".
I prefer to keep original base name though. RemoteDesktopNMHost sounds a bit
redundant as this bundle is already nested in ChromeRemoteDesktopHost.bundle. It
is also more consistent with remote_assistance_host where the bundle name and
the binary name are identical.

On 2014/03/08 02:37:56, Sergey Ulanov wrote:
> NativeMessagingHost.bundle
> But I'm not sure about this one. it can be something like
> "RemoteDesktopNMHost.bundle"

https://codereview.chromium.org/186763004/diff/40001/remoting/branding_Chrome...
remoting/branding_Chrome:15:
MAC_REMOTE_ASSISTANCE_HOST_BUNDLE_NAME=RemoteAssistanceHost
On 2014/03/08 02:37:56, Sergey Ulanov wrote:
> RemoteAssistanceHost.bundle

Done.

https://codereview.chromium.org/186763004/diff/40001/remoting/remoting_host_m...
File remoting/remoting_host_mac.gypi (right):

https://codereview.chromium.org/186763004/diff/40001/remoting/remoting_host_m...
remoting/remoting_host_mac.gypi:92: 'native_messaging_host_bundle_name':
'<!(python <(version_py_path) -f <(branding_path) -t
"@MAC_NATIVE_MESSAGING_HOST_BUNDLE_NAME@")',
Agreed that the NM hosts should be nested in the me2me bundle.

But I don't think Resources/ is the right location. I placed them under
Content/MacOS. This is more consistent with the other nested bundles I examined.
 
On 2014/03/08 02:37:56, Sergey Ulanov wrote:
> Can we put the NM host bundle inside the me2me bundle (e.g. in
> Content/Resources/NativeMessagingHost.bundle)? seems like it would be the
right
> place for it.

[weitao, 2014-03-24 17:09:30]

I reverted the renaming of the config file, the .conf, .log, .plist, and .sh
files. Also included are a couple of bug fixes:
1. Fixed the installation path of the nm hosts.
2. Add the missing VERSION definition in the build.

[Sergey Ulanov, 2014-03-24 19:11:04]

https://codereview.chromium.org/186763004/diff/80001/remoting/branding_Chrome
File remoting/branding_Chrome (right):

https://codereview.chromium.org/186763004/diff/80001/remoting/branding_Chrome...
remoting/branding_Chrome:10:
MAC_NATIVE_MESSAGING_HOST_BUNDLE_ID=com.google.chromeremotedesktop.native-messaging-host
can this be in com.google.chrome.remote_desktop.native_messaging_host, to make
it more consistent with the naming of the NM hosts

https://codereview.chromium.org/186763004/diff/80001/remoting/branding_Chrome...
remoting/branding_Chrome:14:
MAC_REMOTE_ASSISTANCE_HOST_BUNDLE_ID=com.google.chromeremotedesktop.remote-assistance-host
same here - com.google.chrome.remote_desktop.remote_assistance

https://codereview.chromium.org/186763004/diff/80001/remoting/host/constants_...
File remoting/host/constants_mac.cc (right):

https://codereview.chromium.org/186763004/diff/80001/remoting/host/constants_...
remoting/host/constants_mac.cc:10: #define PREFPANE_BUNDLE_NAME
"ChromeRemoteDesktop.prefPane"
Pass these from gyp, so that they are correct for chromium builds?

https://codereview.chromium.org/186763004/diff/80001/remoting/host/installer/...
File
remoting/host/installer/mac/PrivilegedHelperTools/org.chromium.chromoting.me2me.sh
(right):

https://codereview.chromium.org/186763004/diff/80001/remoting/host/installer/...
remoting/host/installer/mac/PrivilegedHelperTools/org.chromium.chromoting.me2me.sh:1:
#!/bin/sh
rietveld shows me this file as Deleted. Is it really deleted?

[weitao, 2014-03-24 21:21:33]

https://codereview.chromium.org/186763004/diff/80001/remoting/branding_Chrome
File remoting/branding_Chrome (right):

https://codereview.chromium.org/186763004/diff/80001/remoting/branding_Chrome...
remoting/branding_Chrome:10:
MAC_NATIVE_MESSAGING_HOST_BUNDLE_ID=com.google.chromeremotedesktop.native-messaging-host
On 2014/03/24 19:11:04, Sergey Ulanov wrote:
> can this be in com.google.chrome.remote_desktop.native_messaging_host, to make
> it more consistent with the naming of the NM hosts

Done.

https://codereview.chromium.org/186763004/diff/80001/remoting/branding_Chrome...
remoting/branding_Chrome:14:
MAC_REMOTE_ASSISTANCE_HOST_BUNDLE_ID=com.google.chromeremotedesktop.remote-assistance-host
On 2014/03/24 19:11:04, Sergey Ulanov wrote:
> same here - com.google.chrome.remote_desktop.remote_assistance

Done.

https://codereview.chromium.org/186763004/diff/80001/remoting/host/constants_...
File remoting/host/constants_mac.cc (right):

https://codereview.chromium.org/186763004/diff/80001/remoting/host/constants_...
remoting/host/constants_mac.cc:10: #define PREFPANE_BUNDLE_NAME
"ChromeRemoteDesktop.prefPane"
We need to do this for HOST_BUNDLE_NAME as well as PREFPANE_BUNDEL_NAME, and
maybe other stuff too. I plan to refactor this in the subsequent CL. Does that
sound OK?

On 2014/03/24 19:11:04, Sergey Ulanov wrote:
> Pass these from gyp, so that they are correct for chromium builds?

https://codereview.chromium.org/186763004/diff/80001/remoting/host/installer/...
File
remoting/host/installer/mac/PrivilegedHelperTools/org.chromium.chromoting.me2me.sh
(right):

https://codereview.chromium.org/186763004/diff/80001/remoting/host/installer/...
remoting/host/installer/mac/PrivilegedHelperTools/org.chromium.chromoting.me2me.sh:1:
#!/bin/sh
I don't intend to delete it. But how do I correct it? Git add this file is a
no-op as the file already exists in Git in my local checkout.

On 2014/03/24 19:11:04, Sergey Ulanov wrote:
> rietveld shows me this file as Deleted. Is it really deleted?

[Sergey Ulanov, 2014-03-24 23:32:35]

lgtm

[weitao, 2014-03-25 00:01:29]

The CQ bit was checked by weitaosu@chromium.org

[commit-bot: I haz the power, 2014-03-25 00:02:32]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/weitaosu@chromium.org/186763004/100001

[commit-bot: I haz the power, 2014-03-25 00:33:08]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-03-25 00:33:08]

Retried try job too often on chromium_presubmit for step(s) presubmit
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=chromium_p...

[weitao, 2014-03-25 05:51:27]

Craig, could you please give the owner's nod to
chrome/test/pyautolib/chromoting_helper.py?

[Sergey Ulanov, 2014-03-25 07:48:45]

https://codereview.chromium.org/186763004/diff/100001/chrome/test/pyautolib/c...
File chrome/test/pyautolib/chromoting_helper.py (right):

https://codereview.chromium.org/186763004/diff/100001/chrome/test/pyautolib/c...
chrome/test/pyautolib/chromoting_helper.py:127: 'ChromeRemoteDesktop.prefPane')
pyauto tests are on death row - is it worth updating them?

[craigdh, 2014-03-25 14:29:50]

The one-word change in the pyautolib lgtm.

[weitao, 2014-03-25 16:55:53]

The CQ bit was checked by weitaosu@chromium.org

[commit-bot: I haz the power, 2014-03-25 16:56:06]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/weitaosu@chromium.org/186763004/100001

[commit-bot: I haz the power, 2014-03-25 19:29:18]

Change committed as 259287

[weitao, 2014-03-27 19:01:57]

The CQ bit was checked by weitaosu@chromium.org

[commit-bot: I haz the power, 2014-03-27 19:03:49]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/weitaosu@chromium.org/186763004/270001

[commit-bot: I haz the power, 2014-03-27 20:25:23]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-03-27 20:25:23]

Retried try job too often on win_rel for step(s) content_browsertests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[weitao, 2014-03-27 21:29:26]

The CQ bit was checked by weitaosu@chromium.org

[commit-bot: I haz the power, 2014-03-27 21:35:14]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/weitaosu@chromium.org/186763004/270001

[commit-bot: I haz the power, 2014-03-28 02:00:01]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-03-28 02:00:02]

Commit queue rejected this change because the description was changed
between the time the change entered the commit queue and the time it
was ready to commit. You can safely check the commit box again.

[weitao, 2014-03-28 05:19:10]

The CQ bit was checked by weitaosu@chromium.org

[commit-bot: I haz the power, 2014-03-28 05:19:37]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/weitaosu@chromium.org/186763004/270001

[commit-bot: I haz the power, 2014-03-28 05:22:02]

Change committed as 260075