[dev-platform] Intent to prototype SVG new path data API

2024-12-11 Thread Robert Longson
Summary: SVG 1.1 had a path API that was deprecated in SVG 2 and removed 
from all browsers. A new API has been created.

A polyfill already exists for part of this API 
https://github.com/jarek-foksa/path-data-polyfill We can compare our 
implementation's functionality against that.

Bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1934525

Specification: https://svgwg.org/specs/paths/#DOMInterfaces

Standards body: w3c

-- 
You received this message because you are subscribed to the Google Groups 
"dev-platform@mozilla.org" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dev-platform+unsubscr...@mozilla.org.
To view this discussion visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/26b0fe3a-57f3-42aa-b497-d058ad4be71bn%40mozilla.org.


Re: [dev-platform] Intent to prototype SVG new path data API

2024-12-11 Thread 'Mike Conca' via dev-platform@mozilla.org
Hi Robert,

Can you please share the status of this API in the other browser engines?

Thanks,
Mike

Mike Conca
Director of Product, Firefox Web Platform


On Wed, Dec 11, 2024 at 7:41 AM Robert Longson  wrote:

> Summary: SVG 1.1 had a path API that was deprecated in SVG 2 and removed
> from all browsers. A new API has been created.
>
> A polyfill already exists for part of this API
> https://github.com/jarek-foksa/path-data-polyfill We can compare our
> implementation's functionality against that.
>
> Bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1934525
>
> Specification: https://svgwg.org/specs/paths/#DOMInterfaces
>
> Standards body: w3c
>
> --
> You received this message because you are subscribed to the Google Groups "
> dev-platform@mozilla.org" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to dev-platform+unsubscr...@mozilla.org.
> To view this discussion visit
> https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/26b0fe3a-57f3-42aa-b497-d058ad4be71bn%40mozilla.org
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"dev-platform@mozilla.org" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dev-platform+unsubscr...@mozilla.org.
To view this discussion visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/CAMcy6JCOv_tcUjdCdZHnZqEW2bRayN9%3Dr8AbWx78XZr4oS4fzg%40mail.gmail.com.


Re: [dev-platform] Intent to prototype SVG new path data API

2024-12-11 Thread Robert Longson
Hi Mike,

Various requests in https://issues.chromium.org/issues/40441025 for it to 
be implemented but seems like little progress.

Looks like Safari hasn't implemented either the path d property 
(https://bugs.webkit.org/show_bug.cgi?id=234227) or removed the old path 
API (https://bugs.webkit.org/show_bug.cgi?id=168519). Implementing the new 
API without removing the old one would make things much more complicated so 
I imagine they would want to fix both those issues first.

Robert

On Wednesday, 11 December 2024 at 14:54:52 UTC mco...@mozilla.com wrote:

> Hi Robert, 
>
> Can you please share the status of this API in the other browser engines?
>
> Thanks,
> Mike
>
> Mike Conca
> Director of Product, Firefox Web Platform
>
>
> On Wed, Dec 11, 2024 at 7:41 AM Robert Longson  wrote:
>
>> Summary: SVG 1.1 had a path API that was deprecated in SVG 2 and removed 
>> from all browsers. A new API has been created.
>>
>> A polyfill already exists for part of this API 
>> https://github.com/jarek-foksa/path-data-polyfill We can compare our 
>> implementation's functionality against that.
>>
>> Bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1934525
>>
>> Specification: https://svgwg.org/specs/paths/#DOMInterfaces
>>
>> Standards body: w3c
>>
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "dev-pl...@mozilla.org" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to dev-platform...@mozilla.org.
>> To view this discussion visit 
>> https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/26b0fe3a-57f3-42aa-b497-d058ad4be71bn%40mozilla.org
>>  
>> 
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"dev-platform@mozilla.org" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dev-platform+unsubscr...@mozilla.org.
To view this discussion visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/b9ace74d-92b2-48cf-b113-10a5e623d6f2n%40mozilla.org.


[dev-platform] New crash ping dashboard now available

2024-12-11 Thread 'Alex Franchuk' via dev-platform@mozilla.org

I’m announcing the availability of a new crash ping dashboard:
https://crash-pings.mozilla.org.

Some of you may be familiar with Socorro 
, our crash report dashboard. The 
difference is that crash pings are default-on with telemetry, and don’t 
require user input to be sent (because they do not contain PII: 
Personally identifiable information), whereas reports require user input 
to be sent. We have far less volume of reports than pings, and they are 
far more biased. This means that reports are not the best source of data 
to, for example, determine top crashers: to get a comprehensive view of 
stability, use ping data!


This dashboard was adapted and reworked from the existing 
Telemetry-based dashboard 
. It uses Glean crash 
ping data, which is now available for all platforms. This includes 
Android crashes (native and Java), which we didn’t have in Telemetry!


Our dashboard currently shows a sampling of the received pings in the 
past week, where we limit results within groups of release channel and 
process type. This ensures each channel and process type has some pings, 
but allows us to reasonably process the total volume of data. We are 
working on improving symbolication to be able to increase the sample 
size, however the current sample size results in less than about 1.5% 
error in the 99% confidence interval.


The dashboard is still being developed and improved, but we are 
announcing it now as it is moderately functional and may prove useful to 
you. If you encounter any issues or have suggestions, please submit bugs 
on BMO under the Crash Pings
:: Website 
 
component.


We plan to move the existing automated topcrasher bug flagging to use 
this crash ping data rather than the crash report data it currently 
uses. Note, however, that this will never be a replacement for Socorro 
and crash reports: crash ping data is a good place to start when 
triaging issues and assessing stability, but the depth of information 
available from reports is essential for debugging issues.


Thanks all,

Alex Franchuk

​

--
You received this message because you are subscribed to the Google Groups 
"dev-platform@mozilla.org" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dev-platform+unsubscr...@mozilla.org.
To view this discussion visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/24ac77b1-ef3a-4b30-9024-77e2ebcc6772%40mozilla.com.


[dev-platform] BackstagePass is renamed to SystemGlobal

2024-12-11 Thread Tooru Fujisawa
In bug 1767888, we've renamed the "BackstagePass" object to "SystemGlobal".

https://bugzilla.mozilla.org/show_bug.cgi?id=1767888

"BackstagePass" was a name of global, which is mostly used for the shared 
system global for system modules (previously .jsm and now .sys.mjs), and also 
used for xpcshell global etc, and historically used for some other things.

The name sometimes appears in the log, test output, profile, debugger, etc, and 
had been a source of confusion.

Given that the main consumer is now the shared system global, we've reflected 
it to the name.

This may affect some tooling or automation, such as intermittent bug 
classification, crash reports, etc.
Ping me when you hit some issues, or if you have any questions.

-- 
arai

-- 
You received this message because you are subscribed to the Google Groups 
"dev-platform@mozilla.org" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dev-platform+unsubscr...@mozilla.org.
To view this discussion visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/4323963A-2E4D-47A8-A17A-5E729982ACED%40gmail.com.


Re: [dev-platform] New crash ping dashboard now available

2024-12-11 Thread 'Jeff Muizelaar' via dev-platform@mozilla.org
I'm excited to see this finally done!

I noticed that there are no GPU process crashes on Android. Any idea why?

-Jeff

On Wed, Dec 11, 2024 at 10:14 AM 'Alex Franchuk' via
dev-platform@mozilla.org  wrote:

> I’m announcing the availability of a new crash ping dashboard:
> https://crash-pings.mozilla.org.
>
> Some of you may be familiar with Socorro ,
> our crash report dashboard. The difference is that crash pings are
> default-on with telemetry, and don’t require user input to be sent (because
> they do not contain PII: Personally identifiable information), whereas
> reports require user input to be sent. We have far less volume of reports
> than pings, and they are far more biased. This means that reports are not
> the best source of data to, for example, determine top crashers: to get a
> comprehensive view of stability, use ping data!
>
> This dashboard was adapted and reworked from the existing Telemetry-based
> dashboard . It uses Glean
> crash ping data, which is now available for all platforms. This includes
> Android crashes (native and Java), which we didn’t have in Telemetry!
>
> Our dashboard currently shows a sampling of the received pings in the past
> week, where we limit results within groups of release channel and process
> type. This ensures each channel and process type has some pings, but allows
> us to reasonably process the total volume of data. We are working on
> improving symbolication to be able to increase the sample size, however the
> current sample size results in less than about 1.5% error in the 99%
> confidence interval.
>
> The dashboard is still being developed and improved, but we are announcing
> it now as it is moderately functional and may prove useful to you. If you
> encounter any issues or have suggestions, please submit bugs on BMO under
> the Crash Pings
> :: Website
> 
> component.
>
> We plan to move the existing automated topcrasher bug flagging to use this
> crash ping data rather than the crash report data it currently uses. Note,
> however, that this will never be a replacement for Socorro and crash
> reports: crash ping data is a good place to start when triaging issues and
> assessing stability, but the depth of information available from reports is
> essential for debugging issues.
>
> Thanks all,
>
> Alex Franchuk
> ​
>
> --
> You received this message because you are subscribed to the Google Groups "
> dev-platform@mozilla.org" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to dev-platform+unsubscr...@mozilla.org.
> To view this discussion visit
> https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/24ac77b1-ef3a-4b30-9024-77e2ebcc6772%40mozilla.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"dev-platform@mozilla.org" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dev-platform+unsubscr...@mozilla.org.
To view this discussion visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/CAKb2zs_H24gzCnYKPGKcGa%3Dig%3DsrNL4kjmRsJhFWjKgNCwSKkA%40mail.gmail.com.


Re: [dev-platform] New crash ping dashboard now available

2024-12-11 Thread 'Bobby Holley' via dev-platform@mozilla.org
Great work everyone! This is a big milestone in a multi-year journey to
level-up our stability infrastructure. Thanks to everyone involved along
the way for getting us here.

On Wed, Dec 11, 2024 at 7:14 AM 'Alex Franchuk' via dev-platform@mozilla.org
 wrote:

> I’m announcing the availability of a new crash ping dashboard:
> https://crash-pings.mozilla.org.
>
> Some of you may be familiar with Socorro ,
> our crash report dashboard. The difference is that crash pings are
> default-on with telemetry, and don’t require user input to be sent (because
> they do not contain PII: Personally identifiable information), whereas
> reports require user input to be sent. We have far less volume of reports
> than pings, and they are far more biased. This means that reports are not
> the best source of data to, for example, determine top crashers: to get a
> comprehensive view of stability, use ping data!
>
> This dashboard was adapted and reworked from the existing Telemetry-based
> dashboard . It uses Glean
> crash ping data, which is now available for all platforms. This includes
> Android crashes (native and Java), which we didn’t have in Telemetry!
>
> Our dashboard currently shows a sampling of the received pings in the past
> week, where we limit results within groups of release channel and process
> type. This ensures each channel and process type has some pings, but allows
> us to reasonably process the total volume of data. We are working on
> improving symbolication to be able to increase the sample size, however the
> current sample size results in less than about 1.5% error in the 99%
> confidence interval.
>
> The dashboard is still being developed and improved, but we are announcing
> it now as it is moderately functional and may prove useful to you. If you
> encounter any issues or have suggestions, please submit bugs on BMO under
> the Crash Pings
> :: Website
> 
> component.
>
> We plan to move the existing automated topcrasher bug flagging to use this
> crash ping data rather than the crash report data it currently uses. Note,
> however, that this will never be a replacement for Socorro and crash
> reports: crash ping data is a good place to start when triaging issues and
> assessing stability, but the depth of information available from reports is
> essential for debugging issues.
>
> Thanks all,
>
> Alex Franchuk
> ​
>
> --
> You received this message because you are subscribed to the Google Groups "
> dev-platform@mozilla.org" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to dev-platform+unsubscr...@mozilla.org.
> To view this discussion visit
> https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/24ac77b1-ef3a-4b30-9024-77e2ebcc6772%40mozilla.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"dev-platform@mozilla.org" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dev-platform+unsubscr...@mozilla.org.
To view this discussion visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/CADa3RMO6joTJRc6XSZzrD2GXaiCZuSs%3DGYc593evOX51d%3DY50g%40mail.gmail.com.


[dev-platform] Intent to Ship: High-precision coordinates, i.e., fractional coordinates, of PointerEvents in Nightly builds

2024-12-11 Thread Masayuki Nakano
We shipped high-precision coordinates, i.e., fractional coordinates, of 
PointerEvents in Nightly builds.


Summary: `|screenX|`, `screenY`, `clientX`, `clientY`, `pageX`, `pageY`, 
`offsetX` and `offsetY` of DOM events are now changed from `long` to 
`double`. And these values may be fractional values if the event is a 
`PointerEvent` whose type is NOT `click`, `auxclick` nor `contextmenu`.


Bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1680669

Specification: CSS OM 
 
and PointerEvents 


Platform coverage: All

Preference: `dom.event.pointer.fractional_coordinates.enabled`

DevTools bug: N/A

Other browsers: Chrome already supports this. Safari seems supporting it 
only for trusted events and only in preview builds.


web-platform-tests: trusted: 
>, 
untrusted: 
>


--
Masayuki Nakano
Working on DOM, Events, editor and IME handling for Gecko

--
You received this message because you are subscribed to the Google Groups 
"dev-platform@mozilla.org" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dev-platform+unsubscr...@mozilla.org.
To view this discussion visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/cacc3257-5b8b-4ab4-a597-89da968a23d2%40d-toybox.com.


Re: [dev-platform] New crash ping dashboard now available

2024-12-11 Thread 'Alex Franchuk' via dev-platform@mozilla.org
Jeff,

I haven't looked closer at the technical details behind it, but based on 
the list of possible values for 
https://searchfox.org/mozilla-central/rev/8f423e3c79147c65d078589f3666dd25ffe80840/mobile/android/android-components/components/lib/crash/src/main/java/mozilla/components/lib/crash/Crash.kt#174,
 
it appears as if we don't yet support that level of granularity (but I feel 
as if it ought not be too difficult to improve that?).

-Alex

On Wednesday, December 11, 2024 at 1:50:37 PM UTC-5 Jeff Muizelaar wrote:

> I'm excited to see this finally done!
>
> I noticed that there are no GPU process crashes on Android. Any idea why?
>
> -Jeff
>
> On Wed, Dec 11, 2024 at 10:14 AM 'Alex Franchuk' via 
> dev-platform@mozilla.org  wrote:
>
>> I’m announcing the availability of a new crash ping dashboard:
>> https://crash-pings.mozilla.org.
>>
>> Some of you may be familiar with Socorro , 
>> our crash report dashboard. The difference is that crash pings are 
>> default-on with telemetry, and don’t require user input to be sent (because 
>> they do not contain PII: Personally identifiable information), whereas 
>> reports require user input to be sent. We have far less volume of reports 
>> than pings, and they are far more biased. This means that reports are not 
>> the best source of data to, for example, determine top crashers: to get a 
>> comprehensive view of stability, use ping data!
>>
>> This dashboard was adapted and reworked from the existing Telemetry-based 
>> dashboard . It uses 
>> Glean crash ping data, which is now available for all platforms. This 
>> includes Android crashes (native and Java), which we didn’t have in 
>> Telemetry!
>>
>> Our dashboard currently shows a sampling of the received pings in the 
>> past week, where we limit results within groups of release channel and 
>> process type. This ensures each channel and process type has some pings, 
>> but allows us to reasonably process the total volume of data. We are 
>> working on improving symbolication to be able to increase the sample size, 
>> however the current sample size results in less than about 1.5% error in 
>> the 99% confidence interval.
>>
>> The dashboard is still being developed and improved, but we are 
>> announcing it now as it is moderately functional and may prove useful to 
>> you. If you encounter any issues or have suggestions, please submit bugs on 
>> BMO under the Crash Pings
>> :: Website 
>> 
>>  
>> component. 
>>
>> We plan to move the existing automated topcrasher bug flagging to use 
>> this crash ping data rather than the crash report data it currently uses. 
>> Note, however, that this will never be a replacement for Socorro and crash 
>> reports: crash ping data is a good place to start when triaging issues and 
>> assessing stability, but the depth of information available from reports is 
>> essential for debugging issues.
>>
>> Thanks all,
>>
>> Alex Franchuk
>> ​
>>
>> -- 
>>
> You received this message because you are subscribed to the Google Groups "
>> dev-platform@mozilla.org" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to dev-platform+unsubscr...@mozilla.org.
>> To view this discussion visit 
>> https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/24ac77b1-ef3a-4b30-9024-77e2ebcc6772%40mozilla.com
>>  
>> 
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"dev-platform@mozilla.org" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to dev-platform+unsubscr...@mozilla.org.
To view this discussion visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/4342a785-0442-4886-b89f-d9aa26e6743fn%40mozilla.org.