Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Proposal to remove `console.exception` from the console spec

8 views
Skip to first unread message

Brian Grinstead

unread,
Feb 17, 2018, 6:49:13 PM2/17/18
to dev-developer-tools
There’s a discussion about removing `console.exception` from the specification at https://github.com/whatwg/console/issues/27#issuecomment-366326423. The method was added in Bug 922214, and came over from Firebug’s console implementation. Some points in favor of removing it:

1) It's basically just an alias for console.error (I can’t find any difference in the implementation between the two methods).
2) There are only a few hits within m-c for it (https://searchfox.org/mozilla-central/search?q=console.exception&path=) and would be easy to migrate to `console.error`.
3) Chrome, Safari, and node haven’t ever implemented it, and it’s not likely to get added to the spec unless if we advocate for it.

I don't have a strong argument for keeping it in the spec, but the main argument against removing our implementation would be web compat breakage for sites currently using it. Given that only Firefox and Edge have implemented it, I would expect it to generally be used alongside a polyfill.

Feedback is welcome. Also, do we have existing telemetry data on web api usage that would capture console methods? If we do and someone could point me towards that it would be helpful.

Thanks,
Brian

Harald Kirschner

unread,
Feb 18, 2018, 10:03:42 AM2/18/18
to Brian Grinstead, dev-developer-tools, Michael Taylor
[+miket for web compat input]

Given the potential web compat issues and prior issues we recently found
with synthetic errors; I’d like to get some more data to inform this. It
might be ok as only Edge and FF have it, but lets get more data points.

API calls are not instrumented by telemetry, other than few manual
exceptions.

As we have work lined up for improving console error formatting (usage of
error objects and error logging methods); we could actually invest in
adding a probe that can inform this using at least some Nightly data.

/Harald

On Sat, Feb 17, 2018 at 15:49 Brian Grinstead <bgrin...@mozilla.com>
wrote:
> _______________________________________________
> dev-developer-tools mailing list
> dev-devel...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-developer-tools
>

Brian Grinstead

unread,
Feb 20, 2018, 4:18:52 PM2/20/18
to Harald Kirschner, dev-developer-tools, Michael Taylor
What do we need to do to start gathering this data? I’d like to provide a response in the spec issue with a timeline for when we'll be able to make a final decision one way or another.

Brian

> On Feb 18, 2018, at 7:03 AM, Harald Kirschner <har...@mozilla.com> wrote:
>
> [+miket for web compat input]
>
> Given the potential web compat issues and prior issues we recently found with synthetic errors; I’d like to get some more data to inform this. It might be ok as only Edge and FF have it, but lets get more data points.
>
> API calls are not instrumented by telemetry, other than few manual exceptions.
>
> As we have work lined up for improving console error formatting (usage of error objects and error logging methods); we could actually invest in adding a probe that can inform this using at least some Nightly data.
>
> /Harald
>
> On Sat, Feb 17, 2018 at 15:49 Brian Grinstead <bgrin...@mozilla.com> wrote:
> There’s a discussion about removing `console.exception` from the specification at https://github.com/whatwg/console/issues/27#issuecomment-366326423. The method was added in Bug 922214, and came over from Firebug’s console implementation. Some points in favor of removing it:
>
> 1) It's basically just an alias for console.error (I can’t find any difference in the implementation between the two methods).
> 2) There are only a few hits within m-c for it (https://searchfox.org/mozilla-central/search?q=console.exception&path=) and would be easy to migrate to `console.error`.

bgrin...@mozilla.com

unread,
Feb 23, 2018, 12:56:13 PM2/23/18
to mozilla-dev-d...@lists.mozilla.org
On Tuesday, February 20, 2018 at 1:18:52 PM UTC-8, Brian Grinstead wrote:
> What do we need to do to start gathering this data? I’d like to provide a response in the spec issue with a timeline for when we'll be able to make a final decision one way or another.

I filed https://bugzilla.mozilla.org/show_bug.cgi?id=1440729 to collect the data.

Brian

Mike Taylor

unread,
Feb 23, 2018, 5:13:58 PM2/23/18
to bgrin...@mozilla.com, mozilla-dev-d...@lists.mozilla.org

> On Feb 23, 2018, at 12:56 PM, bgrin...@mozilla.com wrote:
>
> On Tuesday, February 20, 2018 at 1:18:52 PM UTC-8, Brian Grinstead wrote:
>> What do we need to do to start gathering this data? I’d like to provide a response in the spec issue with a timeline for when we'll be able to make a final decision one way or another.
>
Thanks Brian. I’ll add some thoughts to the original GitHub issue.

--
Mike Taylor
Web Compat, Mozilla


0 new messages