Intent to remove WebSQL from Workers

346 views
Skip to first unread message

Elliott Sprehn

unread,
Jul 30, 2014, 4:32:24 PM7/30/14
to blink-dev
Primary eng (and PM) emails
esprehn@, abarth@

Summary
Remove WebSQL support from workers.

Motivation
This feature is unused and adds complexity.

Compatibility Risk
Since the count is < 0% (rounded down) I suspect none.

Alternative implementation suggestion for web developers
Use indexeddb.

Usage information from UseCounter

Entry on chromestatus.com, crbug.com, or MDN

but that's not specific to workers.

Requesting approval to remove too?
Yes.

- E

Adam Barth

unread,
Jul 30, 2014, 4:41:40 PM7/30/14
to esp...@chromium.org, blin...@chromium.org
LGTM

For comparison, WebSQL on the main thread is still pretty widely used at ~0.33%:

Adam Klein

unread,
Jul 30, 2014, 5:00:52 PM7/30/14
to Elliott Sprehn, blink-dev, Kent Tamura
How does this relate to tkent's thread from just last week? Link: https://groups.google.com/a/chromium.org/d/msg/blink-dev/2SWnbXAL7Wg/OYQYBjL_bhsJ

Adam Barth

unread,
Jul 30, 2014, 5:15:27 PM7/30/14
to ad...@chromium.org, esp...@chromium.org, blin...@chromium.org, tk...@chromium.org
Hum...  With such low usage, it seems likely we'll remove this feature.  Maybe we should put it behind a runtime flag now to start getting feedback from the field?  We should check the data from the stable channel before actually removing the code, however, in case something surprising happens and we need to reverse course.

Adam

TAMURA, Kent

unread,
Jul 30, 2014, 9:34:31 PM7/30/14
to Elliott Sprehn, blink-dev
I thought we should not do the sudden removal because the counters are not in stable channel.
Do you have any reason to hurry it?

--
TAMURA Kent
Software Engineer, Google


Adam Barth

unread,
Jul 30, 2014, 10:06:09 PM7/30/14
to tk...@chromium.org, esp...@chromium.org, blin...@chromium.org
On Wed Jul 30 2014 at 6:34:29 PM TAMURA, Kent <tk...@chromium.org> wrote:
I thought we should not do the sudden removal because the counters are not in stable channel.
Do you have any reason to hurry it?

No particular reason.

Adam

Adam Barth

unread,
Aug 5, 2014, 1:35:00 PM8/5/14
to tk...@chromium.org, esp...@chromium.org, blin...@chromium.org
The API OWNERS met today and decided that we should wait before removing this feature until the deprecation reaches the stable channel.

Adam

TAMURA, Kent

unread,
Sep 1, 2014, 1:56:00 AM9/1/14
to Adam Barth, Elliott Sprehn, blink-dev
Now stable channel of Google Chrome has their use counters, and we see no significant increase.
LGTM to remove.


Kentaro Hara

unread,
Sep 1, 2014, 2:12:18 AM9/1/14
to TAMURA, Kent, Adam Barth, Elliott Sprehn, blink-dev
LGTM

By removing WebSQL from workers, we can:

- move webdatabase/ to Oilpan.
- make the main thread join with worker threads when shutting down. We've spent a lot of time in fixing threading races between the main thread and terminating worker threads. These threading races will be gone by joining the worker threads gracefully.




--
Kentaro Hara, Tokyo, Japan

PhistucK

unread,
Sep 1, 2014, 2:16:00 AM9/1/14
to TAMURA, Kent, Adam Barth, Elliott Sprehn, blink-dev
Please, give it two weeks for the release to be widely adopted.


☆PhistucK


To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org.

Jochen Eisinger

unread,
Sep 1, 2014, 8:37:56 AM9/1/14
to PhistucK, TAMURA, Kent, Adam Barth, Elliott Sprehn, blink-dev
LGTM assuming the numbers don't suddenly spike in the next two weeks

best
-jochen

PhistucK

unread,
Sep 1, 2014, 8:51:36 AM9/1/14
to TAMURA, Kent, Adam Barth, Elliott Sprehn, blink-dev
I think the second link is unrelated... something about animation.


☆PhistucK


On Mon, Sep 1, 2014 at 8:55 AM, TAMURA, Kent <tk...@chromium.org> wrote:

TAMURA, Kent

unread,
Sep 10, 2014, 2:13:42 AM9/10/14
to Adam Barth, Elliott Sprehn, blink-dev
FYI.

Last day:
OpenDatabaseInWorker: 0.0000046%
OpenDatabaseSyncInWorker: 0.0000039%

Last 7 days:
OpenDatabaseInWorker: 0.0000040%
OpenDatabaseSyncInWorker: 0.0000041%




On Mon, Sep 1, 2014 at 2:55 PM, TAMURA, Kent <tk...@chromium.org> wrote:

Jochen Eisinger

unread,
Sep 10, 2014, 3:12:51 AM9/10/14
to TAMURA, Kent, Adam Barth, Elliott Sprehn, blink-dev
that's close enough to zero

-jochen

Eric Seidel

unread,
Sep 10, 2014, 12:42:39 PM9/10/14
to Jochen Eisinger, TAMURA, Kent, Adam Barth, Elliott Sprehn, blink-dev
lgtm

Adam Barth

unread,
Sep 10, 2014, 4:35:19 PM9/10/14
to Eric Seidel, Jochen Eisinger, TAMURA, Kent, Elliott Sprehn, blink-dev
LGTM

TAMURA, Kent

unread,
Sep 16, 2014, 8:11:28 PM9/16/14
to Adam Barth, Eric Seidel, Jochen Eisinger, Elliott Sprehn, blink-dev
The code was removed.  This will be effective since Google Chrome 39.

Reply all
Reply to author
Forward
0 new messages