LGTM3

On Fri, Jul 26, 2024, 8:42 AM Rick Byers <[email protected]> wrote:

> I'm a bit worried about primarily chromium-sites or Android WebView apps
> who might have used this by accident instead of innerHTML. I took a look at
> WebView-specific UseCounters and found them to be less than half of the
> Chrome Android values (~0.0007% vs. 0.0002%), so I expect that just
> represents general web browsing rather than any significant
> WebView-specific usage.
>
> So seems worth trying to me (and disabling if there are non-trivial
> reports of issues). LGTM2
>
> On Fri, Jul 26, 2024 at 5:58 AM Yoav Weiss (@Shopify) <
> [email protected]> wrote:
>
>> LGTM1 to remove in M129, given the usecounter numbers
>>
>> On Wed, Jul 10, 2024 at 11:35 PM Mason Freed <[email protected]>
>> wrote:
>>
>>> I'd like to check back in on this deprecation. The drop in usage from
>>> May continued into June and July, with current numbers hovering around
>>> 0.003%:
>>> https://chromestatus.com/metrics/feature/timeline/popularity/3874. I
>>> added deprecation messages in M127, but that's not even in stable yet. My
>>> guess would be that usage will continue to drop as 127 goes to stable.
>>>
>>> Given the above, I'd like to re-request permission to remove the feature
>>> in M129, assuming all of the trends continue to look good.
>>>
>>> Thanks,
>>> Mason
>>>
>>>
>>> On Tuesday, June 4, 2024 at 12:28:54 AM UTC-7 Daniel Bratell wrote:
>>>
>>>> LGTM3 for the deprecation in 127. I'd like to hold off on stamping the
>>>> removal approval until later but threatening (well, targetting) removal in
>>>> 129 seems ok.
>>>>
>>>> /Daniel
>>>> On 2024-05-31 02:44, Mason Freed wrote:
>>>>
>>>> Thanks for the LGTMs!
>>>>
>>>> On Wed, May 29, 2024 at 7:54 AM Daniel Bratell <[email protected]>
>>>> wrote:
>>>>
>>>>> Deprecating seems fine, especially since it's a relatively new API and
>>>>> less likely to be used on non-maintained sites, but removal seems a bit
>>>>> premature even if done slowly. Would it be better to let it bake a few
>>>>> milestones and see if a scary deprecation message threatening removal 
>>>>> after
>>>>> the summer drives the usage down?
>>>>>
>>>> I'm starting to get the feeling that this would be a good idea. The
>>>> Enterprise folks on the chromestatus gate also asked for something similar
>>>> - 3 milestones of warning period before the deprecation.
>>>>
>>>> I'm inclined to just do that - it feels safer, and I don't think
>>>> there's a huge rush to turn this off. Especially with usage going in the
>>>> right direction, instead of further up.
>>>>
>>>> So I've modified the chromestatus
>>>> <https://chromestatus.com/feature/5081733588582400> to show a shipping
>>>> milestone of M129. Does that work? And I'll add a big console error
>>>> starting in M127.
>>>>
>>>>
>>>> On Wed, May 29, 2024 at 1:22 AM Yoav Weiss (@Shopify) <
>>>> [email protected]> wrote:
>>>>
>>>>> Your questions prompted me to take a closer look at the sample sites
>>>>>>> still hitting the use counter. I took a close look at the first 10 
>>>>>>> entries
>>>>>>> listed, and I think I found perhaps where the big drop came from. Of 
>>>>>>> those
>>>>>>> ten sites, seven do not use getInnerHTML or getHTML at all. Likely not
>>>>>>> coincidentally, all seven are Shopify sites. My guess would be that 
>>>>>>> Shopify
>>>>>>> very recently removed its usage of getInnerHTML?
>>>>>>>
>>>>>>
>>>>> FWIW, internal code search brought up nothing. It's possible that this
>>>>> is a 3P app
>>>>> <https://shopify.dev/docs/apps/build/online-store/theme-app-extensions> 
>>>>> that
>>>>> changed their use. (or that I'm simply failing to find the relevant change
>>>>> :D)
>>>>>
>>>>
>>>> Thanks for checking that! It's always hard to figure out what happened
>>>> after the fact like this.
>>>>
>>>>
>>>>> The plan to ramp down usage is a good one, although as we previously
>>>>>> discussed in a different thread, it may cause some debugging challenges 
>>>>>> for
>>>>>> developers. It is worthwhile to also reach out to some of the developers
>>>>>> whose sites you noticed would throw an exception -- just an FYI email 
>>>>>> that
>>>>>> this feature is being removed. Given the fairly low usage, readily
>>>>>> available fixes (via getHTML() or possibly innerHTML) and your commitment
>>>>>> to monitor for breakages, this looks good to me.
>>>>>>
>>>>>
>>>> I might want to tweak this plan now that we're extending by 2/3
>>>> milestones. Maybe now (after console warnings) it makes more sense to just
>>>> disable in one shot?
>>>>
>>>> I'll try to reach out to the sites I noticed.
>>>>
>>>> Thanks,
>>>> Mason
>>>>
>>>>
>>>>
>>>>
>>>>> /Daniel
>>>>> On 2024-05-29 10:22, Yoav Weiss (@Shopify) wrote:
>>>>>
>>>>> LGTM2
>>>>>
>>>>> On Tue, May 28, 2024 at 11:10 PM Vladimir Levin <[email protected]>
>>>>> wrote:
>>>>>
>>>>>>
>>>>>>
>>>>>> On Tue, May 28, 2024 at 12:30 PM Mason Freed <[email protected]>
>>>>>> wrote:
>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Mon, May 27, 2024 at 8:15 AM Vladimir Levin <[email protected]>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Interoperability and Compatibility
>>>>>>>>>
>>>>>>>>> The use counter for getInnerHTML() (
>>>>>>>>> https://chromestatus.com/metrics/feature/timeline/popularity/3874)
>>>>>>>>> peaked at 0.05% of page loads using this function as of January 2024, 
>>>>>>>>> and
>>>>>>>>> dropped precipitously toward 0.01% in May, 2024. This is presumably 
>>>>>>>>> due to
>>>>>>>>> the shipment of its replacement, getHTML().
>>>>>>>>>
>>>>>>>>
>>>>>>>> It's great to see the numbers reduce significantly. If the numbers
>>>>>>>> are being migrated to getHTML() though I would have expected
>>>>>>>> https://chromestatus.com/metrics/feature/timeline/popularity/4781
>>>>>>>> to grow by ~0.04 percentage points, but that one is still significantly
>>>>>>>> lower (although growing). Is it possible that June 1 numbers would 
>>>>>>>> show a
>>>>>>>> better balance? Do you by any chance know when the next data point is
>>>>>>>> expected to be visible on chromestatus?
>>>>>>>>
>>>>>>>> I'm also assume people are using a readily available replacement as
>>>>>>>> opposed to just not using getInnerHTML, but it would be nice if number
>>>>>>>> supported that
>>>>>>>>
>>>>>>>
>>>>>>> Great questions. So AFAIK the use counter plot for the current month
>>>>>>> is a continuous aggregation. I.e. the 0.0168% I see today (May 28) is 
>>>>>>> as of
>>>>>>> the 28th, and will change tomorrow (slightly). Given that we're almost 
>>>>>>> to
>>>>>>> the end of the month, I wouldn't expect a ton of shift. So I think you
>>>>>>> might be right that this isn't actually a shift to getHTML, but just a
>>>>>>> shift away from getInnerHTML. See more below.
>>>>>>>
>>>>>>> Your questions prompted me to take a closer look at the sample sites
>>>>>>> still hitting the use counter. I took a close look at the first 10 
>>>>>>> entries
>>>>>>> listed, and I think I found perhaps where the big drop came from. Of 
>>>>>>> those
>>>>>>> ten sites, seven do not use getInnerHTML or getHTML at all. Likely not
>>>>>>> coincidentally, all seven are Shopify sites. My guess would be that 
>>>>>>> Shopify
>>>>>>> very recently removed its usage of getInnerHTML?
>>>>>>>
>>>>>>
>>>>> FWIW, internal code search brought up nothing. It's possible that this
>>>>> is a 3P app
>>>>> <https://shopify.dev/docs/apps/build/online-store/theme-app-extensions>
>>>>> that changed their use. (or that I'm simply failing to find the relevant
>>>>> change :D)
>>>>>
>>>>>
>>>>>>> The real issue is that the remaining three sites *do* still use
>>>>>>> getInnerHTML, and all three throw exceptions when the feature is 
>>>>>>> disabled.
>>>>>>> I can't perceive anything broken on the site, but the exception isn't a
>>>>>>> good sign. A few interesting tidbits: one of the three does appear to
>>>>>>> (properly) feature-detect getInnerHTML() yet an exception is still 
>>>>>>> thrown
>>>>>>> that might or might not be related. The other two do not feature detect,
>>>>>>> and the exception is clear: "getInnerHTML is not a function". Very
>>>>>>> interestingly, none of the three use getInnerHMTL for anything 
>>>>>>> declarative
>>>>>>> shadow dom related. They seem to just be using it as a way to get the
>>>>>>> innerHTML value. All three seem to be hand-written JS, so it's possible 
>>>>>>> the
>>>>>>> sites were developed on Chrome in the last few years and the developer
>>>>>>> didn't notice that they should have done foo=el.innerHTML instead of
>>>>>>> foo=el.getInnerHTML().
>>>>>>>
>>>>>>> Given that the use counter is very low (0.01%), I'd still like to
>>>>>>> push ahead with this deprecation. The above sites likely represent 
>>>>>>> interop
>>>>>>> problems, since they'll break on other browsers already today. But I'd 
>>>>>>> like
>>>>>>> to revise my plan: instead of going immediately to 100% removal, I'd 
>>>>>>> like
>>>>>>> to use a slow ramp down over time, to monitor for reported breakage.
>>>>>>>
>>>>>>> Thoughts?
>>>>>>>
>>>>>>
>>>>>> The plan to ramp down usage is a good one, although as we previously
>>>>>> discussed in a different thread, it may cause some debugging challenges 
>>>>>> for
>>>>>> developers. It is worthwhile to also reach out to some of the developers
>>>>>> whose sites you noticed would throw an exception -- just an FYI email 
>>>>>> that
>>>>>> this feature is being removed. Given the fairly low usage, readily
>>>>>> available fixes (via getHTML() or possibly innerHTML) and your commitment
>>>>>> to monitor for breakages, this looks good to me.
>>>>>>
>>>>>> LGTM1
>>>>>>
>>>>>>
>>>>>>>
>>>>>>> Thanks,
>>>>>>> Mason
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>> While 0.01% still represents high usage for deprecation, the
>>>>>>>>> numbers were significantly worse for the deprecation of the old
>>>>>>>>> `shadowroot` attribute, and the removal of that feature generated 
>>>>>>>>> zero bug
>>>>>>>>> reports. It is my strong belief that since this feature is only 
>>>>>>>>> shipped in
>>>>>>>>> Chrome, the vast majority of usage is guarded by feature checks. So 
>>>>>>>>> this
>>>>>>>>> deprecation should be safer than it would seem from the numbers. I'd 
>>>>>>>>> like
>>>>>>>>> to remove this feature in M127 in code, with a killswitch (a re-enable
>>>>>>>>> switch really) in case of problems.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> *Gecko*: No signal
>>>>>>>>>
>>>>>>>>> *WebKit*: No signal
>>>>>>>>>
>>>>>>>>> *Web developers*: No signals
>>>>>>>>>
>>>>>>>>> *Other signals*:
>>>>>>>>>
>>>>>>>>> WebView application risks
>>>>>>>>>
>>>>>>>>> Does this intent deprecate or change behavior of existing APIs,
>>>>>>>>> such that it has potentially high risk for Android WebView-based
>>>>>>>>> applications?
>>>>>>>>>
>>>>>>>>> None
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Debuggability
>>>>>>>>>
>>>>>>>>> None
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Will this feature be supported on all six Blink platforms
>>>>>>>>> (Windows, Mac, Linux, ChromeOS, Android, and Android WebView)? Yes
>>>>>>>>>
>>>>>>>>> Is this feature fully tested by web-platform-tests
>>>>>>>>> <https://chromium.googlesource.com/chromium/src/+/main/docs/testing/web_platform_tests.md>
>>>>>>>>> ? Yes
>>>>>>>>>
>>>>>>>>> Flag name on chrome://flags ElementGetInnerHTML
>>>>>>>>>
>>>>>>>>> Finch feature name ElementGetInnerHTML
>>>>>>>>>
>>>>>>>>> Requires code in //chrome? False
>>>>>>>>>
>>>>>>>>> Tracking bug https://crbug.com/1519972
>>>>>>>>>
>>>>>>>>> Estimated milestones
>>>>>>>>> Shipping on desktop 127
>>>>>>>>> Shipping on Android 127
>>>>>>>>> Shipping on WebView 127
>>>>>>>>>
>>>>>>>>> Anticipated spec changes
>>>>>>>>>
>>>>>>>>> Open questions about a feature may be a source of future web
>>>>>>>>> compat or interop issues. Please list open issues (e.g. links to known
>>>>>>>>> github issues in the project for the feature specification) whose
>>>>>>>>> resolution may introduce web compat/interop risk (e.g., changing to 
>>>>>>>>> naming
>>>>>>>>> or structure of the API in a non-backward-compatible way).
>>>>>>>>> None
>>>>>>>>>
>>>>>>>>> Link to entry on the Chrome Platform Status
>>>>>>>>> https://chromestatus.com/feature/5081733588582400?gate=5088451454304256
>>>>>>>>>
>>>>>>>>> This intent message was generated by Chrome Platform Status
>>>>>>>>> <https://chromestatus.com/>.
>>>>>>>>> --
>>>>>>>>> You received this message because you are subscribed to the Google
>>>>>>>>> Groups "blink-dev" group.
>>>>>>>>> To unsubscribe from this group and stop receiving emails from it,
>>>>>>>>> send an email to [email protected].
>>>>>>>>> To view this discussion on the web visit
>>>>>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAM%3DNeDjZJvRAcpSj2cAWi6uW7yYmDV8HdMkqQjFOS3q%3DidB9fQ%40mail.gmail.com
>>>>>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAM%3DNeDjZJvRAcpSj2cAWi6uW7yYmDV8HdMkqQjFOS3q%3DidB9fQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
>>>>>>>>> .
>>>>>>>>>
>>>>>>>> --
>>>>>>> You received this message because you are subscribed to the Google
>>>>>>> Groups "blink-dev" group.
>>>>>>> To unsubscribe from this group and stop receiving emails from it,
>>>>>>> send an email to [email protected].
>>>>>>> To view this discussion on the web visit
>>>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAM%3DNeDh_myM1eudBh_%3DeY4F9UZN9vZxRa9%2BmV9vKSR9Nh1iHhw%40mail.gmail.com
>>>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAM%3DNeDh_myM1eudBh_%3DeY4F9UZN9vZxRa9%2BmV9vKSR9Nh1iHhw%40mail.gmail.com?utm_medium=email&utm_source=footer>
>>>>>>> .
>>>>>>>
>>>>>> --
>>>>>> You received this message because you are subscribed to the Google
>>>>>> Groups "blink-dev" group.
>>>>>> To unsubscribe from this group and stop receiving emails from it,
>>>>>> send an email to [email protected].
>>>>>> To view this discussion on the web visit
>>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CADsXd2MvNLyYnrF2q%2BhEx8EAoiZsC_ws3Bj3%3DOw7K5m-hRAvDQ%40mail.gmail.com
>>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CADsXd2MvNLyYnrF2q%2BhEx8EAoiZsC_ws3Bj3%3DOw7K5m-hRAvDQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
>>>>>> .
>>>>>>
>>>>> --
>>>>> You received this message because you are subscribed to the Google
>>>>> Groups "blink-dev" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>>> an email to [email protected].
>>>>> To view this discussion on the web visit
>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohSJ_LsKuzR2FezpC4cxyFsb5nUaG65LeidpyfSmFWVGJaQ%40mail.gmail.com
>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohSJ_LsKuzR2FezpC4cxyFsb5nUaG65LeidpyfSmFWVGJaQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
>>>>> .
>>>>>
>>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups "blink-dev" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to [email protected].
>>>>
>>>> To view this discussion on the web visit
>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAM%3DNeDhoYiq3D6HFupxGhQL1SOSKVUFD%3DxQJGXTamyDvPTDr2A%40mail.gmail.com
>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAM%3DNeDhoYiq3D6HFupxGhQL1SOSKVUFD%3DxQJGXTamyDvPTDr2A%40mail.gmail.com?utm_medium=email&utm_source=footer>
>>>> .
>>>>
>>>> --
>> You received this message because you are subscribed to the Google Groups
>> "blink-dev" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> To view this discussion on the web visit
>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohS%2BZW0sf1J4s1qCVQaLKZnQZ7712WXn%3D5v4rZPGb0AgtFQ%40mail.gmail.com
>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohS%2BZW0sf1J4s1qCVQaLKZnQZ7712WXn%3D5v4rZPGb0AgtFQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
>> .
>>
> --
> You received this message because you are subscribed to the Google Groups
> "blink-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To view this discussion on the web visit
> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAFUtAY_2jbdZe6%3DMphTATNLym9C1E2M80hUfa%2BgzucajxKxXDA%40mail.gmail.com
> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAFUtAY_2jbdZe6%3DMphTATNLym9C1E2M80hUfa%2BgzucajxKxXDA%40mail.gmail.com?utm_medium=email&utm_source=footer>
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOMQ%2Bw-EZVng%2BejfrWWhTW6BzwVFXQ6C_S6CMq4EMtxmthbicQ%40mail.gmail.com.

Reply via email to