Convert the sqlite cookie database and web database to use the new sqlite...

Convert the sqlite cookie database and web database to use the new sqlite
wrapper. This also moves and renamed the old cookie_monster_sqlite file to
match the class name.

BUG=none
TEST=none

[cpu_(ooo_6.6-7.5), 2009-09-14 23:06:21]

LGTM

Some nits below

http://codereview.chromium.org/201099/diff/3002/5002
File app/sql/connection.h (right):

http://codereview.chromium.org/201099/diff/3002/5002#newcode199
Line 199: bool DoesTableExist( const char* table_name);
-extra space ( const 
-can this be made const?

http://codereview.chromium.org/201099/diff/3002/5002#newcode210
Line 210: // is closed.
soo scary that it does not take the statement. Is our reliance  on this safe?

http://codereview.chromium.org/201099/diff/3002/5008
File chrome/browser/net/sqlite_persistent_cookie_store.cc (left):

http://codereview.chromium.org/201099/diff/3002/5008#oldcode338
Line 338: sqlite3_close(db);
that was loco to close the db just like that!

http://codereview.chromium.org/201099/diff/3002/5008
File chrome/browser/net/sqlite_persistent_cookie_store.cc (right):

http://codereview.chromium.org/201099/diff/3002/5008#newcode346
Line 346: if (!cc.get())
345-347 remove?

http://codereview.chromium.org/201099/diff/3002/5004
File chrome/browser/webdata/web_database.cc (right):

http://codereview.chromium.org/201099/diff/3002/5004#newcode297
Line 297: // First time.
remove 297 and 327?

http://codereview.chromium.org/201099/diff/3002/5005
File chrome/browser/webdata/web_database.h (right):

http://codereview.chromium.org/201099/diff/3002/5005#newcode113
Line 113: bool RemoveLoginsCreatedBetween(base::Time delete_begin,
const reference?

The time structure seems pretty fat, maybe there is a reason for this?

http://codereview.chromium.org/201099/diff/3002/5005#newcode157
Line 157: base::Time delete_end);
const reference?

[brettw, 2009-09-15 00:45:27]

http://codereview.chromium.org/201099/diff/3002/5002
File app/sql/connection.h (right):

http://codereview.chromium.org/201099/diff/3002/5002#newcode210
Line 210: // is closed.
On 2009/09/14 23:06:22, cpu wrote:
> soo scary that it does not take the statement. Is our reliance  on this safe?

This is sqlite's model (like the last inserted row ID). I think this is only
used in one place, so yes.

http://codereview.chromium.org/201099/diff/3002/5005
File chrome/browser/webdata/web_database.h (right):

http://codereview.chromium.org/201099/diff/3002/5005#newcode113
Line 113: bool RemoveLoginsCreatedBetween(base::Time delete_begin,
On 2009/09/14 23:06:22, cpu wrote:
> const reference?
> 
> The time structure seems pretty fat, maybe there is a reason for this?

Time is just an int64. I intuitively think of it like a nice wrapper around
time_t, so pass by value (I bet this is cheaper than dereferencing the reference
too).

[cpu_(ooo_6.6-7.5), 2009-09-21 21:30:24]

Ok. Go for it.


On 2009/09/15 00:45:27, brettw wrote:
> http://codereview.chromium.org/201099/diff/3002/5002
> File app/sql/connection.h (right):
> 
> http://codereview.chromium.org/201099/diff/3002/5002#newcode210
> Line 210: // is closed.
> On 2009/09/14 23:06:22, cpu wrote:
> > soo scary that it does not take the statement. Is our reliance  on this
safe?
> 
> This is sqlite's model (like the last inserted row ID). I think this is only
> used in one place, so yes.
> 
> http://codereview.chromium.org/201099/diff/3002/5005
> File chrome/browser/webdata/web_database.h (right):
> 
> http://codereview.chromium.org/201099/diff/3002/5005#newcode113
> Line 113: bool RemoveLoginsCreatedBetween(base::Time delete_begin,
> On 2009/09/14 23:06:22, cpu wrote:
> > const reference?
> > 
> > The time structure seems pretty fat, maybe there is a reason for this?
> 
> Time is just an int64. I intuitively think of it like a nice wrapper around
> time_t, so pass by value (I bet this is cheaper than dereferencing the
reference
> too).