[sqlite] Indirect recover.c dependencies on Pager.

[sqlite] Indirect recover.c dependencies on Pager.

Replace dependencies on SQLite's Pager and DbPage structures with local
wrappers, in preparation for rewriting those wrappers in terms of
sqlite3_file.  The primary substantive change is that RecoverPager is an
owned reference, while Pager was a reference shared with SQLite at
large.  DbPage already needed ref-counting so RecoverPage matches that.

BUG=584407
Committed: https://crrev.com/8089a8ae75369aac3784910276feff69b2f1f6cf
Cr-Commit-Position: refs/heads/master@{#374768}

[Scott Hess - ex-Googler, 2016-02-05 20:38:21]

Fix a potential leak.

[Scott Hess - ex-Googler, 2016-02-05 20:39:25]

shess@chromium.org changed reviewers:
+ rch@chromium.org

[Scott Hess - ex-Googler, 2016-02-05 20:39:26]

OK, here's a more-substantial one!

As for the earlier CL, the changes to sqlite3.c is just copy-pasta of the
changes to recover.c generated by a script.

https://codereview.chromium.org/1671043003/diff/1/third_party/sqlite/src/src/...
File third_party/sqlite/src/src/recover.c (right):

https://codereview.chromium.org/1671043003/diff/1/third_party/sqlite/src/src/...
third_party/sqlite/src/src/recover.c:384: }
The stylized destroy/create groupings are basically my way of dealing with lack
of C++.  I assume they are mostly inlined, but realistically it doesn't matter
for this code.

https://codereview.chromium.org/1671043003/diff/1/third_party/sqlite/src/src/...
third_party/sqlite/src/src/recover.c:514: case 10 : case 11 : assert( "RESERVED
TYPE"==NULL ); return 0;
Oops, didn't intend these to get in, but they seem safe to leave.  At some point
I was compiling outside of third_party/sqlite and they got warned.  sqlite.gyp
has -Wstring-conversion for this reason.

https://codereview.chromium.org/1671043003/diff/1/third_party/sqlite/src/src/...
third_party/sqlite/src/src/recover.c:1184: RecoverPager *pPager;            /*
Page provider. */
Previously found via reference in pPage.  Added explicitly here for purposes of
managing ownership.  There can be pages from the same pager in pParent
(essentially a linked list to the top node of the tree), but pParent will be
unwound before pPager.

[Scott Hess - ex-Googler, 2016-02-05 20:39:47]

The CQ bit was checked by shess@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-05 20:41:22]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1671043003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1671043003/20001

[commit-bot: I haz the power, 2016-02-05 22:06:52]

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

[commit-bot: I haz the power, 2016-02-05 22:06:53]

Dry run: This issue passed the CQ dry run.

[Ryan Hamilton, 2016-02-10 19:44:08]

lgtm

[Scott Hess - ex-Googler, 2016-02-10 20:25:10]

The CQ bit was checked by shess@chromium.org

[commit-bot: I haz the power, 2016-02-10 20:27:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1671043003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1671043003/20001

[commit-bot: I haz the power, 2016-02-10 21:19:09]

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

[commit-bot: I haz the power, 2016-02-10 21:19:09]

Try jobs failed on following builders:
  linux_android_rel_ng on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/linux_androi...)

[Scott Hess - ex-Googler, 2016-02-10 21:23:03]

On 2016/02/10 21:19:09, commit-bot: I haz the power wrote:
> Try jobs failed on following builders:
>   linux_android_rel_ng on tryserver.chromium.android (JOB_FAILED,
>
https://build.chromium.org/p/tryserver.chromium.android/builders/linux_androi...)

Bot exception, AFAICT, pushing it again.

[Scott Hess - ex-Googler, 2016-02-10 21:23:10]

The CQ bit was checked by shess@chromium.org

[commit-bot: I haz the power, 2016-02-10 21:23:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1671043003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1671043003/20001

[commit-bot: I haz the power, 2016-02-10 23:12:33]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-02-16 22:32:11]

Description was changed from

==========
[sqlite] Indirect recover.c dependencies on Pager.

Replace dependencies on SQLite's Pager and DbPage structures with local
wrappers, in preparation for rewriting those wrappers in terms of
sqlite3_file.  The primary substantive change is that RecoverPager is an
owned reference, while Pager was a reference shared with SQLite at
large.  DbPage already needed ref-counting so RecoverPage matches that.

BUG=584407
==========

to

==========
[sqlite] Indirect recover.c dependencies on Pager.

Replace dependencies on SQLite's Pager and DbPage structures with local
wrappers, in preparation for rewriting those wrappers in terms of
sqlite3_file.  The primary substantive change is that RecoverPager is an
owned reference, while Pager was a reference shared with SQLite at
large.  DbPage already needed ref-counting so RecoverPage matches that.

BUG=584407

Committed: https://crrev.com/8089a8ae75369aac3784910276feff69b2f1f6cf
Cr-Commit-Position: refs/heads/master@{#374768}
==========

[commit-bot: I haz the power, 2016-02-16 22:32:12]

Patchset 2 (id:??) landed as
https://crrev.com/8089a8ae75369aac3784910276feff69b2f1f6cf
Cr-Commit-Position: refs/heads/master@{#374768}