Make FTS2 inclusion in SQLite optional

Make FTS2 inclusion in SQLite optional

Only platforms/products that need to support old history files need
FTS2 support in the included SQLite library.

BUG=418082
R=shess@chromium.org
R=michaeln@chromium.org
R=erikwright@chromium.org

Committed: https://crrev.com/ffdda02dd3b523b35f63c0580fcde3f17cdf1e9a
Cr-Commit-Position: refs/heads/master@{#298022}

[the_jk, 2014-09-26 13:38:44]

the_jk@opera.com changed reviewers:
+ erikwright@chromium.org, michaeln@chromium.org

[erikwright (departed), 2014-09-30 14:08:44]

https://codereview.chromium.org/606033002/diff/1/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/606033002/diff/1/build/common.gypi#newcode2342
build/common.gypi:2342: 'sqlite_enable_fts2%': 1,
I think there should be a variable named "support_pre_MXX_history_files". If
that's set, turn on 'sqlite_enable_fts2'.

A bit of archaeology will be required to find out what XX should be.

[michaeln, 2014-10-01 20:04:15]

conditionally building fts2 sgtm, but my gyp kungfu is sorely lacking, can you
run tryjobs?

[the_jk, 2014-10-02 09:54:19]

On 2014/10/01 20:04:15, michaeln wrote:
> conditionally building fts2 sgtm, but my gyp kungfu is sorely lacking, can you
> run tryjobs?

I don't have tryjobs access, sorry

[erikwright (departed), 2014-10-02 14:13:54]

LGTM.

[michaeln, 2014-10-02 19:40:48]

lgtm 2

[Scott Hess - ex-Googler, 2014-10-03 00:22:12]

I'm on leave, so don't take this as authoritative or blocking (if you ask
questions, I might be very delayed in answering, or never answer), but since we
use a single SQLite for Chromium as a whole, it may be possible that someone out
there is using fts2 in websql.  I don't know how to evaluate that chance.  It's
pretty unlikely, though, fts3 existed before Chrome shipped, and I was a little
irate that fts2 had even been included in the first place, so I would fully
support just pulling the plug and watching the bug tracker for reports and
dealing with that fallout.

Personally, I would also support simply dropping fts2 support entirely rather
than having optional stuff in the codebase, and letting people who insist on
having fts2 fork their gyp file using whatever mechanism they already use (note
that those history files are already being deleted by default unless the code is
modified).

[the_jk, 2014-10-03 09:01:42]

On 2014/10/03 00:22:12, Scott Hess - On Leave wrote:
> I'm on leave, so don't take this as authoritative or blocking (if you ask
> questions, I might be very delayed in answering, or never answer), but since
we
> use a single SQLite for Chromium as a whole, it may be possible that someone
out
> there is using fts2 in websql.  I don't know how to evaluate that chance. 
It's
> pretty unlikely, though, fts3 existed before Chrome shipped, and I was a
little
> irate that fts2 had even been included in the first place, so I would fully
> support just pulling the plug and watching the bug tracker for reports and
> dealing with that fallout.
> 
> Personally, I would also support simply dropping fts2 support entirely rather
> than having optional stuff in the codebase, and letting people who insist on
> having fts2 fork their gyp file using whatever mechanism they already use
(note
> that those history files are already being deleted by default unless the code
is
> modified).

I think I'll go the cowards route for this my first chromium commit ;) with a
less
disruptive change. If more platforms than iOS wish to test dropping fts2 support
then can now do that more easily.

[the_jk, 2014-10-03 09:02:42]

The CQ bit was checked by the_jk@opera.com

[commit-bot: I haz the power, 2014-10-03 09:03:41]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/606033002/20001

[commit-bot: I haz the power, 2014-10-03 09:10:06]

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

[commit-bot: I haz the power, 2014-10-03 09:10:07]

Try jobs failed on following builders:
  linux_gpu on tryserver.chromium.gpu ()
  win_gpu on tryserver.chromium.gpu ()
  android_aosp on tryserver.chromium.linux ()
  chromium_presubmit on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)
  ios_rel_device on tryserver.chromium.mac ()
  ios_rel_device_ninja on tryserver.chromium.mac ()
  win8_chromium_rel on tryserver.chromium.win ()

[the_jk, 2014-10-03 10:34:45]

The CQ bit was checked by the_jk@opera.com

[commit-bot: I haz the power, 2014-10-03 10:35:49]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/606033002/40001

[commit-bot: I haz the power, 2014-10-03 11:36:27]

Committed patchset #3 (id:40001) as c9e6c64615b17073444fb33cb4e4d8a0bff634cc

[commit-bot: I haz the power, 2014-10-03 11:37:06]

Patchset 3 (id:??) landed as
https://crrev.com/ffdda02dd3b523b35f63c0580fcde3f17cdf1e9a
Cr-Commit-Position: refs/heads/master@{#298022}