Register Chrome in Open With for File Associations at install on Windows.

Register Chrome in Open With... for File Associations at install on Windows.

Refactor shell_util.cc to use ScopedVector<RegistryEntry> instead of std::list<RegistryEntry*>

BUG=139597
TEST=Chrome shows up in the Open With context menu immediately post-install (not after first-run/make default).
Make sure App Paths still work (i.e. launch "chrome.exe" from Run dialog).

For the OpenWith test (say you're testing OpenWith for .htm files), make sure that you delete Chrome ProgIds from:
HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer\FileExts\.htm\OpenWithProgids
(they are automatically copied there by Windows so if you have previously installed Chrome on this machine and made it default at least once Windows will have picked this up already).

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=150234

[gab, 2012-08-05 07:00:41]

PTAL.

I also took this as an opportunity to slightly refactor registration calls so
that the things (RegisteredApplications) that need to go in HKLM in Win7- are
separate from the rest which can go in HKCU for user-level installs on all
Windows versions.

For OpenWith, I do it XP style
(http://msdn.microsoft.com/en-us/library/bb166549(v=vs.80).aspx); these values
are copied to the appropriate location in later versions of Windows anyways so
it's safe to only do it here and it works on all versions of Windows :).

The only other side-effect of this refactoring is that App Paths are now
registered in HKCU for user-level installs as they should always have been
(http://msdn.microsoft.com/en-us/library/windows/desktop/ee872121(v=vs.85).asp...).

Cheers,
Gab

[gab, 2012-08-05 07:01:49]

[+chrome-win8-eng]

On 2012/08/05 07:00:41, gab wrote:
> PTAL.
> 
> I also took this as an opportunity to slightly refactor registration calls so
> that the things (RegisteredApplications) that need to go in HKLM in Win7- are
> separate from the rest which can go in HKCU for user-level installs on all
> Windows versions.
> 
> For OpenWith, I do it XP style
> (http://msdn.microsoft.com/en-us/library/bb166549%28v=vs.80%29.aspx%29; these
values
> are copied to the appropriate location in later versions of Windows anyways so
> it's safe to only do it here and it works on all versions of Windows :).
> 
> The only other side-effect of this refactoring is that App Paths are now
> registered in HKCU for user-level installs as they should always have been
>
(http://msdn.microsoft.com/en-us/library/windows/desktop/ee872121%28v=vs.85%29...).
> 
> Cheers,
> Gab

[robertshield, 2012-08-05 15:01:12]

looking good, comments and such

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/setup/unin...
File chrome/installer/setup/uninstall.cc (right):

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/setup/unin...
chrome/installer/setup/uninstall.cc:652: // Cleanup OpenWithList
Please add links to
http://msdn.microsoft.com/en-us/library/windows/desktop/ee872121 and
http://msdn.microsoft.com/en-us/library/bb166549 in the comment.

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
File chrome/installer/util/shell_util.cc (right):

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
chrome/installer/util/shell_util.cc:394: // In particular: App Paths and File
Associations.
Please add links to
http://msdn.microsoft.com/en-us/library/windows/desktop/ee872121 and
http://msdn.microsoft.com/en-us/library/bb166549 in the comment.

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
chrome/installer/util/shell_util.cc:399: std::list<RegistryEntry*>* entries) {
any reason not to use base::ScopedVector here? it would reduce the burden on
callers a bit

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
chrome/installer/util/shell_util.cc:461: std::list<RegistryEntry*>* entries) {
same comment about base::ScopedVector

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
chrome/installer/util/shell_util.cc:898: // Change the default browser for
current user.
nit: imo should add a blank line before the comment.

[gab, 2012-08-06 19:39:27]

Done.

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/setup/unin...
File chrome/installer/setup/uninstall.cc (right):

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/setup/unin...
chrome/installer/setup/uninstall.cc:652: // Cleanup OpenWithList
On 2012/08/05 15:01:12, robertshield wrote:
> Please add links to
> http://msdn.microsoft.com/en-us/library/windows/desktop/ee872121 and
> http://msdn.microsoft.com/en-us/library/bb166549 in the comment.

Done.

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
File chrome/installer/util/shell_util.cc (right):

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
chrome/installer/util/shell_util.cc:394: // In particular: App Paths and File
Associations.
On 2012/08/05 15:01:12, robertshield wrote:
> Please add links to
> http://msdn.microsoft.com/en-us/library/windows/desktop/ee872121 and
> http://msdn.microsoft.com/en-us/library/bb166549 in the comment.

Done.

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
chrome/installer/util/shell_util.cc:399: std::list<RegistryEntry*>* entries) {
On 2012/08/05 15:01:12, robertshield wrote:
> any reason not to use base::ScopedVector here? it would reduce the burden on
> callers a bit

shell_util has been using std::list forever, but I'm also in favor of this
change so, done :)!

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
chrome/installer/util/shell_util.cc:461: std::list<RegistryEntry*>* entries) {
On 2012/08/05 15:01:12, robertshield wrote:
> same comment about base::ScopedVector

Done.

http://codereview.chromium.org/10823178/diff/9001/chrome/installer/util/shell...
chrome/installer/util/shell_util.cc:898: // Change the default browser for
current user.
On 2012/08/05 15:01:12, robertshield wrote:
> nit: imo should add a blank line before the comment.

Done.

[robertshield, 2012-08-06 21:12:48]

lgtm

[commit-bot: I haz the power, 2012-08-06 21:20:09]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gab@chromium.org/10823178/6

[commit-bot: I haz the power, 2012-08-06 22:39:03]

Try job failure for 10823178-6 (previous was lost) (retry) on linux_rel for step
"compile" (clobber build).
It's a second try, previously, step "compile" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_rel&...

[commit-bot: I haz the power, 2012-08-06 22:45:01]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gab@chromium.org/10823178/6

[commit-bot: I haz the power, 2012-08-06 23:19:03]

Try job failure for 10823178-6 (retry) on linux_chromeos for step "compile"
(clobber build).
It's a second try, previously, step "compile" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_chro...

[commit-bot: I haz the power, 2012-08-07 00:24:12]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gab@chromium.org/10823178/6

[commit-bot: I haz the power, 2012-08-07 02:16:09]

Change committed as 150234