[sql] Remove misleading AutoRecoverTable() parameter.

[sql] Remove misleading AutoRecoverTable() parameter.

|extend_columns| was intended to be used in the case where the target
schema differed slightly from the schema of the table being recovered.
Unfortunately, it actually implemented part of the solution for when the
target has fewer columns and described it as the solution for when the
target has more columns.  Remove the unnecessary code and parameter.

[In general, table schema only add new columns, removing is more
infrequent.]

BUG=none

Committed: https://crrev.com/6f68bd3907164c404842a54b217fe048ebf80f93
Cr-Commit-Position: refs/heads/master@{#373586}

[Scott Hess - ex-Googler, 2016-02-03 20:30:11]

Give me a night to think about it, and I'll make a pointless change.

[Scott Hess - ex-Googler, 2016-02-03 20:36:18]

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

[Scott Hess - ex-Googler, 2016-02-03 20:36:18]

Another tidbit.  I meant to dump like three of them on you, but got distracted
by a bunny.

https://codereview.chromium.org/1666473003/diff/20001/sql/recovery.cc
File sql/recovery.cc (left):

https://codereview.chromium.org/1666473003/diff/20001/sql/recovery.cc#oldcode468
sql/recovery.cc:468: }
Summary of "why": Recovery works by defining a virtual table over the data of a
real table.  That virtual table allows the regular SQLite case of "row has fewer
columns than expected, return NULL for extras", but if there are more columns
than requested it assumes the row is corrupt.  This code was meant to handle the
fewer-columns-than-requested case, but actually handles the
more-columns-than-requested case.  So it is unnecessary.

[Ryan Hamilton, 2016-02-04 01:36:34]

lgtm

https://codereview.chromium.org/1666473003/diff/20001/sql/recovery.cc
File sql/recovery.cc (left):

https://codereview.chromium.org/1666473003/diff/20001/sql/recovery.cc#oldcode468
sql/recovery.cc:468: }
On 2016/02/03 20:36:18, Scott Hess wrote:
> Summary of "why": Recovery works by defining a virtual table over the data of
a
> real table.  That virtual table allows the regular SQLite case of "row has
fewer
> columns than expected, return NULL for extras", but if there are more columns
> than requested it assumes the row is corrupt.  This code was meant to handle
the
> fewer-columns-than-requested case, but actually handles the
> more-columns-than-requested case.  So it is unnecessary.

Cute!

[Scott Hess - ex-Googler, 2016-02-04 18:39:59]

thanks!

https://codereview.chromium.org/1666473003/diff/20001/sql/recovery.cc
File sql/recovery.cc (left):

https://codereview.chromium.org/1666473003/diff/20001/sql/recovery.cc#oldcode468
sql/recovery.cc:468: }
On 2016/02/04 01:36:34, Ryan Hamilton wrote:
> On 2016/02/03 20:36:18, Scott Hess wrote:
> > Summary of "why": Recovery works by defining a virtual table over the data
of
> a
> > real table.  That virtual table allows the regular SQLite case of "row has
> fewer
> > columns than expected, return NULL for extras", but if there are more
columns
> > than requested it assumes the row is corrupt.  This code was meant to handle
> the
> > fewer-columns-than-requested case, but actually handles the
> > more-columns-than-requested case.  So it is unnecessary.
> 
> Cute!

I was prototyping an alternative recovery method which didn't use the virtual
table, and found that in that case I _did_ have to actually implement this, and
that my test which tested that implementation broke this implementation.  It was
embarrassing.

[Scott Hess - ex-Googler, 2016-02-04 18:40:04]

The CQ bit was checked by shess@chromium.org

[commit-bot: I haz the power, 2016-02-04 18:42:17]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1666473003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1666473003/20001

[commit-bot: I haz the power, 2016-02-04 18:52:39]

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

[commit-bot: I haz the power, 2016-02-04 18:52:39]

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...)

[Scott Hess - ex-Googler, 2016-02-04 19:01:21]

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

[Scott Hess - ex-Googler, 2016-02-04 19:01:22]

sdefresne@ for OWNERS stamp.  Change is trivial, removes a parameter which
didn't actually accomplish anything.

[sdefresne, 2016-02-04 19:15:29]

components lgtm

[Scott Hess - ex-Googler, 2016-02-04 19:17:22]

The CQ bit was checked by shess@chromium.org

[commit-bot: I haz the power, 2016-02-04 19:19:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1666473003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1666473003/20001

[commit-bot: I haz the power, 2016-02-04 19:29:53]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-02-04 19:30:52]

Description was changed from

==========
[sql] Remove misleading AutoRecoverTable() parameter.

|extend_columns| was intended to be used in the case where the target
schema differed slightly from the schema of the table being recovered.
Unfortunately, it actually implemented part of the solution for when the
target has fewer columns and described it as the solution for when the
target has more columns.  Remove the unnecessary code and parameter.

[In general, table schema only add new columns, removing is more
infrequent.]

BUG=none
==========

to

==========
[sql] Remove misleading AutoRecoverTable() parameter.

|extend_columns| was intended to be used in the case where the target
schema differed slightly from the schema of the table being recovered.
Unfortunately, it actually implemented part of the solution for when the
target has fewer columns and described it as the solution for when the
target has more columns.  Remove the unnecessary code and parameter.

[In general, table schema only add new columns, removing is more
infrequent.]

BUG=none

Committed: https://crrev.com/6f68bd3907164c404842a54b217fe048ebf80f93
Cr-Commit-Position: refs/heads/master@{#373586}
==========

[commit-bot: I haz the power, 2016-02-04 19:30:53]

Patchset 2 (id:??) landed as
https://crrev.com/6f68bd3907164c404842a54b217fe048ebf80f93
Cr-Commit-Position: refs/heads/master@{#373586}