ServiceWorker static routing API #1701

yoshisatoyanagisawa · 2024-01-15T06:39:54Z

This API allows developers to configure the routing, and allows them to offload simple things ServiceWorkers do. If the condition matches, the navigation happens without starting ServiceWorkers or executing JavaScript, which allows web pages to avoid performance penalties due to ServiceWorker interceptions.

Explainer: https://github.com/WICG/service-worker-static-routing-api
TAG review: w3ctag/design-reviews#863

💥 Error: 503 Service Unavailable 💥

PR Preview failed to build. (Last tried on Feb 22, 2024, 12:13 AM UTC).

More

PR Preview relies on a number of web services to run. There seems to be an issue with the following one:

🚨 HTML Diff Service - The HTML Diff Service is used to create HTML diffs of the spec changes suggested in a pull request.

🔗 [Related URL]([object Object])

<html><body><h1>503 Service Unavailable</h1>
No server is available to handle this request.
</body></html>

If you don't have enough information above to solve the error by yourself (or to understand to which web service the error is related to, if any), please file an issue.

I used to work on the specification update to support the ServiceWorker static routing API (https://github.com/WICG/service-worker-static-routing-api) w3c#1686 However, I accidentally closed it by force-sync to the ServiceWorker specification's repository HEAD. This CL is for reviving it.

Just trying to add workflows rule to run CI in static_routing_api repository.

revert

Considering the future update of the condition, we should mark the case handled as one of conditions. Also, we need to make it extensible.

It was flipped by mistake.

Revive of PR w3c#1686.

mkruisselbrink

just a couple of hopefully minor points/questions, otherwise this looks good to me.

mkruisselbrink · 2024-02-20T23:34:55Z

docs/index.bs

@@ -1077,7 +1080,9 @@ spec: storage; urlPrefix: https://storage.spec.whatwg.org/
      };
    </pre>

-    A {{ServiceWorkerGlobalScope}} object represents the global execution context of a [=/service worker=]. A {{ServiceWorkerGlobalScope}} object has an associated <dfn export for="ServiceWorkerGlobalScope">service worker</dfn> (a [=/service worker=]). A {{ServiceWorkerGlobalScope}} object has an associated <dfn for="ServiceWorkerGlobalScope">force bypass cache for import scripts flag</dfn>. It is initially unset.
+    A {{ServiceWorkerGlobalScope}} object represents the global execution context of a [=/service worker=]. A {{ServiceWorkerGlobalScope}} object has an associated <dfn export for="ServiceWorkerGlobalScope">service worker</dfn> (a [=/service worker=]). A {{ServiceWorkerGlobalScope}} object has an associated <dfn for="ServiceWorkerGlobalScope">force bypass cache for import scripts flag</dfn>. A {{ServiceWorkerGlobalScope}} object has an associated <dfn for="ServiceWorkerGlobalScope">race response map</dfn> which is an [=ordered map=] where the [=map/keys=] are [=/requests=] and the [=map/values=] are [=race response=]. It is initially unset.


nit: could you re-format this so every "A {{ServiceWorkerGlobalScope}} object has" sentence starts on a new line? And maybe even have blank lines between them so they also start on new lines/paragraphs in the generated HTML. I think that might be more readable.

Done. Also moved It is initially unset. to the correct field.

docs/index.bs

mkruisselbrink · 2024-02-20T23:57:18Z

docs/index.bs

+          1. Else if |source| is {{RouterSourceEnum/"race-network-and-fetch-handler"}}, and |request|'s [=request/method=] is \`<code>GET</code>\` then:
+              1. Let |queue| be an empty [=queue=] of [=/response=].
+              1. Let |raceFetchController| be null.
+              1. Set |raceResponse|'s [=race response/value=] to "<code>pending</code>".


Where did |raceResponse| come from? Should this be Let |raceResponse| be a [=race response=] whose [=race response/value=] is "<code>pending</code>".?

|raceResponse| is declared in L3156.

Let |raceResponse| be a [=race response=] whose [=race response/value=] is null.

Ah yes, but does it need to be defined at that level? It's only ever non-null in this "race-network-and-fetch-handler" branch of the algorithm, so it could be entirely local to this branch?

In my understanding we can't. |raceResponse| is used in L3247, which is outside of this level.

Return the result of [=Create Fetch Event and Dispatch=] with |request|, |registration|, |useHighResPerformanceTimers|, |timingInfo|, |workerRealm|, |reservedClient|, |preloadResponse|, and |raceResponse|.

But while it is used there, afaik it is always null? In which case that line might be clearer just directly passing null anyway.

mkruisselbrink · 2024-02-20T23:58:41Z

docs/index.bs

+              1. Let |queue| be an empty [=queue=] of [=/response=].
+              1. Let |raceFetchController| be null.
+              1. Set |raceResponse|'s [=race response/value=] to "<code>pending</code>".
+              1. Run the following substeps [=in parallel=], but [=abort when=] |controller|'s [=fetch controller/state=] is "<code>terminated</code>" or "<code>aborted</code>":


(mostly an FYI, if #1705 lands first, |controller| will need to be updated to |fetchController|. If this PR lands first, I'll need to update the other PR to also change this here).

Ack. I'll update it once the other PR lands firsts.

docs/index.bs

mkruisselbrink · 2024-02-21T00:06:29Z

docs/index.bs

+      1. Let |shouldSoftUpdate| be true if any of the following are true, and false otherwise:
+          * |request| is a [=non-subresource request=].
+          * |request| is a [=subresource request=] and |registration| is [=stale=].
+      1. Let |raceResponseMap| be |activeWorker|'s [=service worker/global object=]'s [=race response map=].


this should move till after the run Service Worker step below as well to make sure global object has been initialized.

Removed |raceResponseMap| at this point, and directly call |activeWorker|'s [=service worker/global object=]'s [=race response map=] after the global object has been initialized.

Co-authored-by: Marijn Kruisselbrink <mek@chromium.org>

… unset` phrase (#14) * Remove |raceResponseMap| * Re-format {{ServiceWorkerGlobalScope}} part and move `It is initially unset` phrase

Upon the review request, this CL makes |activeWorker| not modified in the Setup ServiceWorkerGlobalScope algorithm. It is the caller's responsibility to update the active service worker's global scope. With this change, an effemeral ServiceWorkerGlobalScope can be used for a cache source in the static routing API.

docs/index.bs

mkruisselbrink · 2024-02-21T21:13:13Z

docs/index.bs

+      :: |workerRealm|, a [=relevant realm=] of the [=service worker/global object=]
+      :: |reservedClient|, a [=request/reserved client=]
+      :: |preloadResponse|, a [=promise=]
+      :: |raceResponse|, a [=race response=]


nit: add "or null", since sometimes you pass in null (and handle null)

mkruisselbrink · 2024-02-21T21:16:55Z

docs/index.bs

+      : Output
+      :: a [=/response=] or null
+
+      1. If |request|'s [=request/reserved client=] is null, return null.


nit: maybe add a Note explaining that this check ensures that we're only dealing with non-subresrouce requests, and thus looking up a registration using the request URL is correct.

Ah, I was wrong. We don't want to limit this algorithm for non-subresource requests. What we'd like to achieve here is to get [=active worker=] from [=/request=], regardless of non-subresource requests or subresource requests.

Updated to get the active worker from subreosurce requests, but perhaps we can just use |request|'s [=request/client=]'s [=active service worker=] for both subresource and non-subresource requests?

What you have now looks reasonable to me. I think using request's client's active service worker for non-subresource requests would be incorrect. Maybe at some point we should abstract out a "give me the right registration for this request" algorithm to be shared between Handle Fetch and this algorithm, but for now what you have seems fine too.

Thank you for the clarification!

* Remove |raceResponseMap| * Re-format {{ServiceWorkerGlobalScope}} part and move `It is initially unset` phrase * Fix |url| in Lookup Race Response * Update race handling, lookup race response algorithm

mkruisselbrink

Thanks for your patience, I think this addressed all my concerns.

SHA: 8d4b9df Reason: push, by mkruisselbrink Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Following w3c#1701, the `install` event now uses the `InstallEvent` interface.

SHA: 8d4b9df Reason: push, by pull[bot] Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Following #1701, the `install` event now uses the `InstallEvent` interface.

In w3c/ServiceWorker#1701, the ServiceWorker static routing API uses the match algorithm to match the request URL. However, the original match algorithm accepts a URLPatternInput object, and it does not accept a URL struct. To allow the ServiceWorker static routing API to use the match algorithm, we need to make the match algorithm to accept the URL struct. Fixes: #218.

Updates in the URLPattern specification was requested during the w3c#1701 review. The issues for the request has been resolved: - whatwg/urlpattern#217 - whatwg/urlpattern#218 However, during the specification updates, the algorithm names are also updated, and adjustment is needed.

…tion (#1707) * Update link texts upon updates in the URLPattern specification Updates in the URLPattern specification was requested during the #1701 review. The issues for the request has been resolved: - whatwg/urlpattern#217 - whatwg/urlpattern#218 However, during the specification updates, the algorithm names are also updated, and adjustment is needed. Co-authored-by: Domenic Denicola <d@domenic.me>

yoshisatoyanagisawa and others added 30 commits November 28, 2023 03:37

Update generate.yml

a884283

Just trying to add workflows rule to run CI in static_routing_api repository.

Update generate.yml

b650b57

revert

Introduce changes upon whatwg/urlpattern#199.

bdc2474

Fixed the parse error.

ef050e2

clean up.

39b670f

another dfn fix.

75d848a

Rename of static router rules object.

025eb82

Stop associating the URLPattern and construct URLPattern on-demand.

d90ac0b

Addressed some of comments.

6577640

Update the case without the urlPattern condition.

cd5c864

Considering the future update of the condition, we should mark the case handled as one of conditions. Also, we need to make it extensible.

Fix the urlPattern existence handling.

0162980

It was flipped by mistake.

Remove the "spec: urlpattern; ..." line.

4639e6c

Simplified the if statement on urlPattern condition does not exist.

f4f57ec

Fix typo.

0d4ed51

Merged the simple fixes.

dd0590b

siple fix.

8bfef6f

Rename algorithm names to align with other algorithm names.

0909c9c

Make which service worker to use clear.

2fd1936

Put algorithm after the algorithm.

bef5857

Fixed typo.

1d8cf97

Fix "Parse URL Pattern" argument.

acdd79d

Merge pull request #1 from yoshisatoyanagisawa/yy_workspace

745eeeb

Revive of PR w3c#1686.

Add 'fetch-event' source

f647207

Fix "fetch-event" to fetch

df4fe98

fxi references

538c264

minor fix

be86954

move has-fetch checking to addRoutes

40eaa69

delete redundant files

c0033b2

Rename hasFetch to hasFetchEventSource

d3bad31

Add note.

088bb7b

mkruisselbrink reviewed Feb 21, 2024

View reviewed changes

sisidovski and others added 4 commits February 21, 2024 10:45

Remove |raceResponseMap|

d237b4a

Update docs/index.bs

7791260

Co-authored-by: Marijn Kruisselbrink <mek@chromium.org>

Re-format {{ServiceWorkerGlobalScope}} part and move `It is initially…

16c38c1

… unset` phrase (#14) * Remove |raceResponseMap| * Re-format {{ServiceWorkerGlobalScope}} part and move `It is initially unset` phrase

yoshisatoyanagisawa commented Feb 21, 2024

View reviewed changes

docs/index.bs Outdated Show resolved Hide resolved

Fix |url| in Lookup Race Response (#15)

9a0f50d

mkruisselbrink reviewed Feb 21, 2024

View reviewed changes

Update race handling, lookup race response algorithm (#16)

31f3b94

* Remove |raceResponseMap| * Re-format {{ServiceWorkerGlobalScope}} part and move `It is initially unset` phrase * Fix |url| in Lookup Race Response * Update race handling, lookup race response algorithm

mkruisselbrink approved these changes Feb 22, 2024

View reviewed changes

mkruisselbrink merged commit 8d4b9df into w3c:main Feb 22, 2024
2 checks passed

mkruisselbrink mentioned this pull request Feb 22, 2024

ServiceWorker static routing API mozilla/standards-positions#828

Closed

github-actions bot added a commit that referenced this pull request Feb 22, 2024

ServiceWorker static routing API (#1701)

3977dc0

SHA: 8d4b9df Reason: push, by mkruisselbrink Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

sisidovski mentioned this pull request Feb 22, 2024

Add docs for service worker static routing API mdn/mdn#527

Open

5 tasks

yoshisatoyanagisawa deleted the static_routing_api branch February 22, 2024 08:40

tidoust added a commit to tidoust/ServiceWorker that referenced this pull request Feb 22, 2024

Update interface for install event in summary table

81398f7

Following w3c#1701, the `install` event now uses the `InstallEvent` interface.

tidoust mentioned this pull request Feb 22, 2024

Update interface for install event in summary table #1706

Merged

mkruisselbrink pushed a commit that referenced this pull request Feb 22, 2024

Update interface for install event in summary table (#1706)

613f5a2

Following #1701, the `install` event now uses the `InstallEvent` interface.

This was referenced Feb 28, 2024

Make the match algorithm accept the URL struct whatwg/urlpattern#221

Merged

Accept URL as the match algorithm's input. whatwg/urlpattern#218

Closed

yoshisatoyanagisawa mentioned this pull request Mar 4, 2024

Editorial: Update link texts upon updates in the URLPattern specification #1707

Merged

saschanaz mentioned this pull request Mar 10, 2024

api.InstallEvent - does not exist in Firefox nor in Safari mdn/browser-compat-data#22591

Open

tetsuharuohzeki mentioned this pull request Mar 19, 2024

Can we make to avoid to invoke ServiceWorker _by default_ for if user code call event.addRoutes(rules) in install event handler? #1712

Open

sisidovski mentioned this pull request May 16, 2024

Add service worker static routing API information to resource timing w3c/resource-timing#389

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

ServiceWorker static routing API #1701

ServiceWorker static routing API #1701

yoshisatoyanagisawa commented Jan 15, 2024 •

edited by pr-preview bot

mkruisselbrink left a comment

mkruisselbrink Feb 20, 2024

sisidovski Feb 21, 2024

mkruisselbrink Feb 20, 2024

sisidovski Feb 21, 2024

mkruisselbrink Feb 21, 2024

sisidovski Feb 21, 2024

mkruisselbrink Feb 21, 2024

sisidovski Feb 22, 2024

mkruisselbrink Feb 20, 2024

sisidovski Feb 21, 2024

mkruisselbrink Feb 21, 2024

sisidovski Feb 21, 2024

mkruisselbrink Feb 21, 2024

sisidovski Feb 22, 2024

mkruisselbrink Feb 21, 2024 •

edited

sisidovski Feb 22, 2024

mkruisselbrink Feb 22, 2024

sisidovski Feb 22, 2024

mkruisselbrink left a comment

ServiceWorker static routing API #1701

ServiceWorker static routing API #1701

Conversation

yoshisatoyanagisawa commented Jan 15, 2024 • edited by pr-preview bot

💥 Error: 503 Service Unavailable 💥

mkruisselbrink left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mkruisselbrink Feb 21, 2024 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mkruisselbrink left a comment

Choose a reason for hiding this comment

yoshisatoyanagisawa commented Jan 15, 2024 •

edited by pr-preview bot

mkruisselbrink Feb 21, 2024 •

edited