Apps on NTP should be in order of installation

Caused pending extensions to install in the proper order (i.e. the order that users have clicked) rather than alphabetically.

BUG=84686
TEST=Manual


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=141312

[clintstaley, 2012-03-08 23:21:42]

lgtm

[Aaron Boodman, 2012-04-05 22:06:43]

http://codereview.chromium.org/9595001/diff/1/chrome/browser/extensions/pendi...
File chrome/browser/extensions/pending_extension_manager.cc (right):

http://codereview.chromium.org/9595001/diff/1/chrome/browser/extensions/pendi...
chrome/browser/extensions/pending_extension_manager.cc:34:
PendingExtensionList::const_iterator iter = GetExtensionListConstIterById(id);
It would be cleaner to make the pending list store linked_ptr<PendingInfo>
rather than raw PendingInfo structures. That way this can just be PendingInfo*
GetPendingInfoByID() and return NULL in the case where one doesn't exist.

Also you could remove the default constructor from PendingInfo which is only
there to support STL.

http://codereview.chromium.org/9595001/diff/1/chrome/browser/extensions/pendi...
File chrome/browser/extensions/pending_extension_manager.h (right):

http://codereview.chromium.org/9595001/diff/1/chrome/browser/extensions/pendi...
chrome/browser/extensions/pending_extension_manager.h:132: 
extra line

[Aaron Boodman, 2012-04-05 22:07:28]

Also, I've added Dmitry because he is probably a good person to look this over.

[Dmitry Polukhin, 2012-04-06 08:10:05]

http://codereview.chromium.org/9595001/diff/1/chrome/browser/extensions/pendi...
File chrome/browser/extensions/pending_extension_manager.cc (right):

http://codereview.chromium.org/9595001/diff/1/chrome/browser/extensions/pendi...
chrome/browser/extensions/pending_extension_manager.cc:218: for (iter =
pending_extension_list_.begin();
It is basically std::find algorithm and IMHO it would be better to use std::find
instead of these two functions. I don't like function that return only iterator
(not pair(begin,end)) because reading caller code it is hard to make sure that
appropriate end() iterator is used.

I would add extension ID to PendingExtensionInfo struct so PendingExtensionList
will be able to contain PendingExtensionInfo instead of pair and add compare
function to use standard find algorithm.

[Dmitry Polukhin, 2012-04-09 09:34:07]

LGTM

[Aaron Boodman, 2012-04-09 17:02:14]

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/ex...
File chrome/browser/extensions/extension_service.cc (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/ex...
chrome/browser/extensions/extension_service.cc:2191: const PendingExtensionInfo*
pending_extension_info;
= NULL

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
File chrome/browser/extensions/pending_extension_info.h (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
chrome/browser/extensions/pending_extension_info.h:69: class IsId {
This class is such a special-case need, it seems like it should be defined in
the cc module where it is needed and not exposed to the outside-world.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
chrome/browser/extensions/pending_extension_info.h:69: class IsId {
Also, the name is not descriptive enough. How about something like
PendingExtensionIDComparator.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
File chrome/browser/extensions/pending_extension_manager.cc (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
chrome/browser/extensions/pending_extension_manager.cc:35:
std::find_if(pending_extension_list_.begin(),
I know Dmitry suggested std::find in a previous review, but I think he was
imagining that you'd be able to compare PendingInfo structs directly.

Personally, I typically don't find find_if useful. It's just as easy to write
the for loop directly, and the code is easier to understand.

But you can leave it this way if you like, so long as you move the comparator
inside this module as suggested in my other comments.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
File chrome/browser/extensions/pending_extension_manager.h (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
chrome/browser/extensions/pending_extension_manager.h:111: // A list of pairs of
<extension id, pending extension info>
Comment is out of date.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
File chrome/browser/extensions/updater/extension_updater.cc (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
chrome/browser/extensions/updater/extension_updater.cc:267: for
(ExtensionSet::const_iterator iter = extensions->begin();
Name this iterator extension_iter and the one below pending_id_iter to increase
clarity.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
chrome/browser/extensions/updater/extension_updater.cc:311:
DCHECK(pending_extension_manager);
I think you meant to DCHECK(info) here. Same with next line. But actually, I
think you can just remove both these lines. We just got these items from
pending_extension_manager, so they should be valid.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
File chrome/browser/extensions/updater/extension_updater.h (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
chrome/browser/extensions/updater/extension_updater.h:10: #include <set>
Is the set include still necessary?

[Dmitry Polukhin, 2012-04-10 08:57:26]

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
File chrome/browser/extensions/pending_extension_manager.cc (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
chrome/browser/extensions/pending_extension_manager.cc:35:
std::find_if(pending_extension_list_.begin(),
On 2012/04/09 17:02:14, Aaron Boodman wrote:
> I know Dmitry suggested std::find in a previous review, but I think he was
> imagining that you'd be able to compare PendingInfo structs directly.
> 
> Personally, I typically don't find find_if useful. It's just as easy to write
> the for loop directly, and the code is easier to understand.
> 
> But you can leave it this way if you like, so long as you move the comparator
> inside this module as suggested in my other comments.

Actually you can use std::find directly you just need to make
PendingExtensionInfo comparable std::string. std::find prototype is:

template<class InputIterator, class T>
InputIterator find ( InputIterator first, InputIterator last, const T& value )

So T will be std::string (i.e. extensions ID) so deference iterator will return
PendingExtensionInfo that will be compared with 'value'. But IMHO, find_if is
also fine and might be easier to understand than direct compare between
PendingExtensionInfo and string.

[mitchellwrosen, 2012-04-13 22:20:18]

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/ex...
File chrome/browser/extensions/extension_service.cc (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/ex...
chrome/browser/extensions/extension_service.cc:2191: const PendingExtensionInfo*
pending_extension_info;
On 2012/04/09 17:02:14, Aaron Boodman wrote:
> = NULL

Done.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
File chrome/browser/extensions/pending_extension_manager.cc (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
chrome/browser/extensions/pending_extension_manager.cc:35:
std::find_if(pending_extension_list_.begin(),
On 2012/04/09 17:02:14, Aaron Boodman wrote:
> I know Dmitry suggested std::find in a previous review, but I think he was
> imagining that you'd be able to compare PendingInfo structs directly.
> 
> Personally, I typically don't find find_if useful. It's just as easy to write
> the for loop directly, and the code is easier to understand.
> 
> But you can leave it this way if you like, so long as you move the comparator
> inside this module as suggested in my other comments.

I went back and changed them all to explicit iteration, and THEN I saw Dmitry's
comment posted. I'll leave it be for this patch set, and if you guys want me to
change it back to using find_if, or find, that's fine too.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
File chrome/browser/extensions/pending_extension_manager.h (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/pe...
chrome/browser/extensions/pending_extension_manager.h:111: // A list of pairs of
<extension id, pending extension info>
On 2012/04/09 17:02:14, Aaron Boodman wrote:
> Comment is out of date.

Done.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
File chrome/browser/extensions/updater/extension_updater.cc (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
chrome/browser/extensions/updater/extension_updater.cc:267: for
(ExtensionSet::const_iterator iter = extensions->begin();
On 2012/04/09 17:02:14, Aaron Boodman wrote:
> Name this iterator extension_iter and the one below pending_id_iter to
increase
> clarity.

Done.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
chrome/browser/extensions/updater/extension_updater.cc:311:
DCHECK(pending_extension_manager);
On 2012/04/09 17:02:14, Aaron Boodman wrote:
> I think you meant to DCHECK(info) here. Same with next line. But actually, I
> think you can just remove both these lines. We just got these items from
> pending_extension_manager, so they should be valid.

Done.

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
File chrome/browser/extensions/updater/extension_updater.h (right):

http://codereview.chromium.org/9595001/diff/8001/chrome/browser/extensions/up...
chrome/browser/extensions/updater/extension_updater.h:10: #include <set>
On 2012/04/09 17:02:14, Aaron Boodman wrote:
> Is the set include still necessary?
Nope!

[Aaron Boodman, 2012-05-01 15:50:22]

Just a few more comments.

Also, are there any tests that rely on this path that we can extend to ensure
the ordering?

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
File chrome/browser/extensions/pending_extension_info.h (right):

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_info.h:51: const std::string& id_;
No need for this to be const since it is private.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
File chrome/browser/extensions/pending_extension_manager.cc (right):

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:5: #include
"chrome/browser/extensions/pending_extension_manager.h"
Thanks for fixing this

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:36: iter !=
pending_extension_list_.end();
Nit: You should bring the typedef back that was in the .h file so you don't have
to repeat this definition so much.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:38: if
(!id.compare(iter->id()))
Nit: We don't usually use compare() with strings. Just use the == operator.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:45: bool
PendingExtensionManager::Remove(const std::string& id) {
Same comments as GetById

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:61: for (iter =
pending_extension_list_.begin();
Same comments as GetById

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:185: for (iter =
pending_extension_list_.begin();
if (GetById(id)) ?

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
File chrome/browser/extensions/pending_extension_manager.h (right):

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.h:46: // Remove |id| from
the set of pending extensions.
Document what the return value means?

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
File chrome/browser/extensions/updater/extension_updater.cc (right):

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/updater/extension_updater.cc:278: pending_ids.begin(),
Nit: no need to wrap so vigorously. These can go on the same line when they fit.

[mitchellwrosen, 2012-05-11 05:45:03]

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
File chrome/browser/extensions/pending_extension_info.h (right):

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_info.h:51: const std::string& id_;
On 2012/05/01 15:50:22, Aaron Boodman wrote:
> No need for this to be const since it is private.

Compiler barks about initializing a non-const std::string& from a const
std::string& in the constructor

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
File chrome/browser/extensions/pending_extension_manager.cc (right):

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:36: iter !=
pending_extension_list_.end();
On 2012/05/01 15:50:22, Aaron Boodman wrote:
> Nit: You should bring the typedef back that was in the .h file so you don't
have
> to repeat this definition so much.

Done.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:38: if
(!id.compare(iter->id()))
On 2012/05/01 15:50:22, Aaron Boodman wrote:
> Nit: We don't usually use compare() with strings. Just use the == operator.

Done.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:45: bool
PendingExtensionManager::Remove(const std::string& id) {
On 2012/05/01 15:50:22, Aaron Boodman wrote:
> Same comments as GetById

Done.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:45: bool
PendingExtensionManager::Remove(const std::string& id) {
On 2012/05/01 15:50:22, Aaron Boodman wrote:
> Same comments as GetById

Done.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:61: for (iter =
pending_extension_list_.begin();
On 2012/05/01 15:50:22, Aaron Boodman wrote:
> Same comments as GetById

Done.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.cc:185: for (iter =
pending_extension_list_.begin();
On 2012/05/01 15:50:22, Aaron Boodman wrote:
> if (GetById(id)) ?

Done.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
File chrome/browser/extensions/pending_extension_manager.h (right):

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/pending_extension_manager.h:46: // Remove |id| from
the set of pending extensions.
On 2012/05/01 15:50:22, Aaron Boodman wrote:
> Document what the return value means?

Done.

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
File chrome/browser/extensions/updater/extension_updater.cc (right):

https://chromiumcodereview.appspot.com/9595001/diff/15001/chrome/browser/exte...
chrome/browser/extensions/updater/extension_updater.cc:278: pending_ids.begin(),
On 2012/05/01 15:50:22, Aaron Boodman wrote:
> Nit: no need to wrap so vigorously. These can go on the same line when they
fit.

Done.

[Aaron Boodman, 2012-05-11 19:23:55]

http://codereview.chromium.org/9595001/diff/15001/chrome/browser/extensions/p...
File chrome/browser/extensions/pending_extension_info.h (right):

http://codereview.chromium.org/9595001/diff/15001/chrome/browser/extensions/p...
chrome/browser/extensions/pending_extension_info.h:51: const std::string& id_;
On 2012/05/11 05:45:03, mitchellwrosen wrote:
> On 2012/05/01 15:50:22, Aaron Boodman wrote:
> > No need for this to be const since it is private.
> 
> Compiler barks about initializing a non-const std::string& from a const
> std::string& in the constructor

It should work if its non-reference. Just plain old std::string. In fact I'm not
really sure what it means to have a reference member.

http://codereview.chromium.org/9595001/diff/24003/chrome/browser/extensions/u...
File chrome/browser/extensions/updater/extension_updater_unittest.cc (right):

http://codereview.chromium.org/9595001/diff/24003/chrome/browser/extensions/u...
chrome/browser/extensions/updater/extension_updater_unittest.cc:238: // Require
that there are no pending extensions in the pending extension manager
Put this comment above the assertion. Or just kill it entirely, the assertion
says the same thing.

http://codereview.chromium.org/9595001/diff/24003/chrome/browser/extensions/u...
chrome/browser/extensions/updater/extension_updater_unittest.cc:267: for
(std::list<std::string>::const_iterator it = pending_ids.begin();
This is fine, but not sure if it's really worth testing.

[mitchellwrosen, 2012-05-24 03:09:00]

Made a few more changes, could one of you look it over one more time before I
send this CL to the trybots? Thanks.

[commit-bot: I haz the power, 2012-05-25 20:22:11]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/9595001/32003

[commit-bot: I haz the power, 2012-05-25 20:22:25]

Presubmit check for 9595001-32003 failed and returned exit status 1.

Running presubmit commit checks ...

** Presubmit ERRORS **
Missing LGTM from an OWNER for files in these directories:
    chrome/browser/sync

Presubmit checks took 2.0s to calculate.

[mitchellwrosen, 2012-05-25 20:27:25]

+akalin for browser/sync

[mitchellwrosen, 2012-06-02 16:56:53]

On 2012/05/25 20:27:25, mitchellwrosen wrote:
> +akalin for browser/sync

[mitchellwrosen, 2012-06-08 02:49:59]

+atwilson for browser/sync

[Andrew T Wilson (Slow), 2012-06-08 15:17:52]

On 2012/06/08 02:49:59, mitchellwrosen wrote:
> +atwilson for browser/sync

browser/sync LGTM

[commit-bot: I haz the power, 2012-06-08 18:30:25]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/9595001/43003

[commit-bot: I haz the power, 2012-06-08 18:51:47]

Try job failure for 9595001-43003 (retry) on mac_rel for step "compile" (clobber
build).
It's a second try, previously, step "compile" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac_rel&nu...

[commit-bot: I haz the power, 2012-06-08 21:03:23]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/9595001/51003

[commit-bot: I haz the power, 2012-06-08 21:03:25]

Failed to apply patch for chrome/browser/extensions/extension_service.cc:
While running patch -p1 --forward --force;
patching file chrome/browser/extensions/extension_service.cc
Hunk #2 FAILED at 575.
1 out of 5 hunks FAILED -- saving rejects to file
chrome/browser/extensions/extension_service.cc.rej

[commit-bot: I haz the power, 2012-06-08 21:12:48]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/9595001/53003

[commit-bot: I haz the power, 2012-06-08 21:31:51]

Try job failure for 9595001-53003 (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, 2012-06-08 21:38:45]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/9595001/49006

[commit-bot: I haz the power, 2012-06-08 22:53:22]

Change committed as 141312