824167 - (SQLite3.7.15.1) Upgrade to SQLite 3.7.15.1

Status: RESOLVED FIXED

(Core :: SQLite and Embedded Database Bindings, defect)

Description

[Ryan VanderMeulen [:RyanVM]]

SQLite Release 3.7.15 On 2012-12-12 (3.7.15)

    Added the sqlite3_errstr() interface.
    Avoid invoking the sqlite3_trace() callback multiple times when a statement is automatically reprepared due to SQLITE_SCHEMA errors.
    Added support for Windows Phone 8 platforms
    Enhance IN operator processing to make use of indices with numeric affinities.
    Do full-table scans using covering indices when possible, under the theory that an index will be smaller and hence can be scanned with less I/O.
    Enhance the query optimizer so that ORDER BY clauses are more aggressively optimized, especially in joins where various terms of the ORDER BY clause come from separate tables of the join.
    Add the ability to implement FROM clause subqueries as coroutines rather that manifesting the subquery into a temporary table.
    Enhancements the command-line shell:
        Added the ".print" command
        Negative numbers in the ".width" command cause right-alignment
        Add the ".wheretrace" command when compiled with SQLITE_DEBUG 
    Added the busy_timeout pragma.
    Added the instr() SQL function.
    Added the SQLITE_FCNTL_BUSYHANDLER file control, used to allow VFS implementations to get access to the busy handler callback.
    The xDelete method in the built-in VFSes now return SQLITE_IOERR_DELETE_NOENT if the file to be deleted does not exist.
    Enhanced support for QNX.
    Work around an optimizer bug in the MSVC compiler when targeting ARM.
    Bug fix: Avoid various concurrency problems in shared cache mode.
    Bug fix: Avoid a deadlock or crash if the backup API, shared cache, and the SQLite Encryption Extension are all used at once.
    Bug fix: SQL functions created using the TCL interface honor the "nullvalue" setting.
    Bug fix: Fix a 32-bit overflow problem on CREATE INDEX for databases larger than 16GB.
    Bug fix: Avoid segfault when using the COLLATE operator inside of a CHECK constraint or view in shared cache mode. 


SQLite Release 3.7.15.1 On 2012-12-19 (3.7.15.1)

    Fix a bug, introduced in version 3.7.15, that causes a segfault if the AS name of a result column of a SELECT statement is used as a logical term in the WHERE clause. Ticket a7b7803e8d1e869.


Green on Try.
https://tbpl.mozilla.org/?tree=Try&rev=969877c4de60

Comment 1

[Ryan VanderMeulen [:RyanVM]]

Attachment: Upgrade to SQLite 3.7.15.1 - Mozilla changes

Not going to bother attaching the SQLite changes patch.

Comment 2

[Marco Bonardo [:mak]]

(In reply to Ryan VanderMeulen from comment #0)
>     Enhance IN operator processing to make use of indices with numeric
> affinities.
>     Do full-table scans using covering indices when possible, under the
> theory that an index will be smaller and hence can be scanned with less I/O.
>     Enhance the query optimizer so that ORDER BY clauses are more
> aggressively optimized, especially in joins where various terms of the ORDER
> BY clause come from separate tables of the join.

These look like may change our perf numbers, after landing we'll have to check talos strictly.  A pre-comparison done through Try (push a head, and the same head with the new version, then compare them with compare-talos) would be nice.

>     Added the instr() SQL function.

sounds interesting to use with substr()

>     Added the SQLITE_FCNTL_BUSYHANDLER file control, used to allow VFS
> implementations to get access to the busy handler callback.
>     The xDelete method in the built-in VFSes now return
> SQLITE_IOERR_DELETE_NOENT if the file to be deleted does not exist.

I suppose they didn't bump the vfs version, since otherwise our build would have failed http://mxr.mozilla.org/mozilla-central/source/storage/src/TelemetryVFS.cpp#503 so we should be fine

Are there changes to sqlite_quota?

Comment 3

[Marco Bonardo [:mak]]

Ah, looks like we don't use it anymore, bug 787804.

Comment 4

[Marco Bonardo [:mak]]

I'd say to remove the "(We no longer use test_quota.h and test_quota.c)" message, we usually don't track what we don't use and it complicates the instructions without any benefix.

Comment 5

[Ryan VanderMeulen [:RyanVM]]

Attachment: Upgrade to SQLite 3.7.15.1 - Mozilla changes

(In reply to Marco Bonardo [:mak] (intermittently avail. until 3 Jan) from comment #2)
> These look like may change our perf numbers, after landing we'll have to
> check talos strictly.  A pre-comparison done through Try (push a head, and
> the same head with the new version, then compare them with compare-talos)
> would be nice.

Argh, I meant to run Talos on this push.
https://tbpl.mozilla.org/?tree=Try&rev=0a5da2ef06d5

(In reply to Marco Bonardo [:mak] (intermittently avail. until 3 Jan) from comment #4)
> I'd say to remove the "(We no longer use test_quota.h and test_quota.c)"
> message, we usually don't track what we don't use and it complicates the
> instructions without any benefix.

Fine by me.

Comment 6

[Marco Bonardo [:mak]]

(In reply to Ryan VanderMeulen from comment #5)
> Argh, I meant to run Talos on this push.
> https://tbpl.mozilla.org/?tree=Try&rev=0a5da2ef06d5

what's the reference push to compare to?

Comment 7

[Ryan VanderMeulen [:RyanVM]]

Parent cset was https://hg.mozilla.org/mozilla-central/rev/84320dffec6e

Comment 8

[Marco Bonardo [:mak]]

Comment on attachment 695270 [details] [diff] [review]
Upgrade to SQLite 3.7.15.1 - Mozilla changes

Review of attachment 695270 [details] [diff] [review]:
-----------------------------------------------------------------

Based on various checks on code changes and talos (even if the current compare looks outdate) I'd say we can proceed.
Please keep an eye on tree-management for eventual notifications.

Comment 9

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/7ef62857f69f
https://hg.mozilla.org/integration/mozilla-inbound/rev/c9a232a7c318

Comment 10

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/7ef62857f69f
https://hg.mozilla.org/mozilla-central/rev/c9a232a7c318

Comment 11

[Ryan VanderMeulen [:RyanVM]]

(In reply to Marco Bonardo [:mak] (intermittently avail. until 3 Jan) from comment #8)
> Please keep an eye on tree-management for eventual notifications.

No complaints thus far.