Add a partial index to the SQLitePersistentCookieStore.

Add a partial index to the SQLitePersistentCookieStore.

This CL adds a partial index is_transient to the cookies DB. This CL also moves
all the index creation operations into the database version migration, so that
the existence of the other indices don't have to be checked on every startup.
Partial indices are available in SQLite 3.8.0, and Chrome ships a branched
version of 3.8.7.4.

BUG=

Committed: https://crrev.com/d2c42f584723689354b1874bdfaefb098f9ea70d
Cr-Commit-Position: refs/heads/master@{#325592}

[erikchen, 2015-04-16 21:52:29]

erikchen@chromium.org changed reviewers:
+ erikwright@chromium.org

[erikchen, 2015-04-16 21:52:29]

erikwright: Please review.

[erikwright (departed), 2015-04-17 01:20:47]

https://codereview.chromium.org/1083623003/diff/1/content/browser/net/sqlite_...
File content/browser/net/sqlite_persistent_cookie_store.cc (right):

https://codereview.chromium.org/1083623003/diff/1/content/browser/net/sqlite_...
content/browser/net/sqlite_persistent_cookie_store.cc:970: if
(!UpdateIndicesOnTable(db_.get())) {
I don't know that this pattern makes sense.

Imagine a column added in version 10, then an index added on it in version 11.

A user who has version 7 will migrate to version 8 and invoke
UpdateIndicesOnTable before the indexed column is created.

We already have some unsightly duplication with the table definition, it's just
less obvious (because the form of an ALTER TABLE is different from that of a
CREATE TABLE). I don't see any point in trying to get around it with cleverness
unless you want to rethink database definition and migration from scratch.

[erikchen, 2015-04-17 01:34:47]

erikwright: PTAL

https://codereview.chromium.org/1083623003/diff/1/content/browser/net/sqlite_...
File content/browser/net/sqlite_persistent_cookie_store.cc (right):

https://codereview.chromium.org/1083623003/diff/1/content/browser/net/sqlite_...
content/browser/net/sqlite_persistent_cookie_store.cc:970: if
(!UpdateIndicesOnTable(db_.get())) {
On 2015/04/17 01:20:47, erikwright wrote:
> I don't know that this pattern makes sense.
> 
> Imagine a column added in version 10, then an index added on it in version 11.
> 
> A user who has version 7 will migrate to version 8 and invoke
> UpdateIndicesOnTable before the indexed column is created.
> 
> We already have some unsightly duplication with the table definition, it's
just
> less obvious (because the form of an ALTER TABLE is different from that of a
> CREATE TABLE). I don't see any point in trying to get around it with
cleverness
> unless you want to rethink database definition and migration from scratch.

Excellent point. I failed to notice that the pattern I used would make it very
easy to make a mistake in the future.

[erikwright (departed), 2015-04-17 01:43:12]

LGTM

[erikchen, 2015-04-17 01:49:53]

The CQ bit was checked by erikchen@chromium.org

[commit-bot: I haz the power, 2015-04-17 01:50:23]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1083623003/20001

[commit-bot: I haz the power, 2015-04-17 02:01:21]

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

[commit-bot: I haz the power, 2015-04-17 02:01:22]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[erikchen, 2015-04-17 02:15:41]

erikchen@chromium.org changed reviewers:
+ isherman@chromium.org

[erikchen, 2015-04-17 02:15:42]

isherman: Looking for an OWNER review of tools/metrics/histograms/histograms.xml

[Ilya Sherman, 2015-04-17 02:18:02]

LGTM, but: Are these histograms actually used?  If not, can you please remove
them in either a precursor or a follow-up CL?

[erikchen, 2015-04-17 02:18:58]

On 2015/04/17 02:18:02, Ilya Sherman wrote:
> LGTM, but: Are these histograms actually used?  If not, can you please remove
> them in either a precursor or a follow-up CL?

I'll deprecate the old histograms in a follow up CL.

[erikchen, 2015-04-17 02:19:02]

The CQ bit was checked by erikchen@chromium.org

[erikchen, 2015-04-17 02:19:02]

The patchset sent to the CQ was uploaded after l-g-t-m from
erikwright@chromium.org
Link to the patchset: https://codereview.chromium.org/1083623003/#ps40001
(title: "Fix typo.")

[commit-bot: I haz the power, 2015-04-17 02:19:23]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1083623003/40001

[commit-bot: I haz the power, 2015-04-17 03:33:46]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-04-17 03:34:33]

Patchset 3 (id:??) landed as
https://crrev.com/d2c42f584723689354b1874bdfaefb098f9ea70d
Cr-Commit-Position: refs/heads/master@{#325592}