Test/internal merge tree client #22697

anthony-murphy · 2024-10-01T01:38:40Z

in progress

…to prototype-deprecate-client

…to test/internal-merge-tree-client

msfluid-bot · 2024-10-01T02:23:18Z

⯆ @fluid-example/bundle-size-tests: -165 Bytes

Metric Name	Baseline Size	Compare Size	Size Diff
aqueduct.js	464.08 KB	463.91 KB	⯆ -170 Bytes
azureClient.js	562.24 KB	562.29 KB	⯅ +49 Bytes
connectionState.js	724 Bytes	724 Bytes	■ No change
containerRuntime.js	261.67 KB	261.68 KB	⯅ +14 Bytes
fluidFramework.js	406.75 KB	406.77 KB	⯅ +14 Bytes
loader.js	134.16 KB	134.18 KB	⯅ +14 Bytes
map.js	42.71 KB	42.71 KB	⯅ +7 Bytes
matrix.js	148.54 KB	148.54 KB	⯅ +7 Bytes
odspClient.js	528.18 KB	528.23 KB	⯅ +49 Bytes
odspDriver.js	97.84 KB	97.86 KB	⯅ +21 Bytes
odspPrefetchSnapshot.js	42.81 KB	42.82 KB	⯅ +14 Bytes
sharedString.js	164.73 KB	164.54 KB	⯆ -198 Bytes
sharedTree.js	397.21 KB	397.22 KB	⯅ +7 Bytes
Total Size	3.32 MB	3.32 MB	⯆ -165 Bytes

Baseline commit: 30003a8

Generated by 🚫 dangerJS against a898922

…to test/internal-merge-tree-client

anthony-murphy · 2024-10-16T23:11:44Z

#22827

packages/dds/merge-tree/api-report/merge-tree.legacy.alpha.api.md

…to test/internal-merge-tree-client

jason-ha · 2024-10-18T00:30:35Z

packages/dds/sequence/package.json

@@ -193,6 +193,7 @@
 	},
 	"typeValidation": {
 		"broken": {},
+		"disabled": true,


Why full stop disable? Isn't this exactly the case where we want to list the incompatibilities? Is another option to change entrypoint from "legacy" to "public"?

things like the 2.4 release and other changes create lots of merge conflicts. i don't want to deal with merge conflict right now. once we are closer to the merge point i'll remove it and deal with them

…to test/internal-merge-tree-client

jason-ha

Shouldn't you be able to make many of the test changes before the breaking change?
When we have a well staged deprecation, I would expect to see more removal of things we don't have to test for anymore compared to test adaptation. The latter suggests customers need to adapt too. OR we don't have good coverage on the external facing APIs which also hinders confidence.

packages/dds/sequence/src/intervalCollection.ts

jason-ha · 2024-10-18T00:37:19Z

packages/dds/sequence/src/intervals/sequenceInterval.ts

 import {
-	IIntervalHelpers,
 	ISerializableInterval,
 	ISerializedInterval,
 	IntervalStickiness,
 	IntervalType,
 	endReferenceSlidingPreference,
 	startReferenceSlidingPreference,
+	type IIntervalHelpers,
 } from "./intervalUtils.js";


I thought the sorting would still want type IIntervalHelpers first.
Other pattern that has no such confusion is to have separate import type {} block. We do that more now.

i don't know. i just let the formatter do what it wants.

jason-ha · 2024-10-18T00:38:39Z

packages/dds/sequence/src/intervals/sequenceInterval.ts

+	readonly stickiness: IntervalStickiness;
+}
+
+export class SequenceIntervalClass implements SequenceInterval {


This should have an explicit tsdoc and @internal tag. No complaints from the tooling?

no it shouldn't, it's never exported from the package

anthony-murphy · 2024-10-18T01:00:53Z

Shouldn't you be able to make many of the test changes before the breaking change? When we have a well staged deprecation, I would expect to see more removal of things we don't have to test for anymore compared to test adaptation. The latter suggests customers need to adapt too. OR we don't have good coverage on the external facing APIs which also hinders confidence.

that is not possible given the changes. the changes literally rename the class and make it un-exported

jason-ha · 2024-10-18T01:04:23Z

Just having another staging thought. For the cases where you now have a FooClass and Foo is interface, I think you can currently do
const FooClass = Foo;
exporting as needed for internal use. Then there is better clarity where you need to preserve the class internally versus interface and less churn when the time comes.
Unfortunately, I don't know that we can get away with the reverse (const Foo = FooClass;) at least not without more song and dance per api-extractor.

anthony-murphy · 2024-10-18T01:06:57Z

Just having another staging thought. For the cases where you now have a FooClass and Foo is interface, I think you can currently do const FooClass = Foo; exporting as needed for internal use. Then there is better clarity where you need to preserve the class internally versus interface and less churn when the time comes. Unfortunately, I don't know that we can get away with the reverse (const Foo = FooClass;) at least not without more song and dance per api-extractor.

i don't understand this comment. i'm not exporting anything concrete. classes are being replaced with interfaces, nothing but types are exported

jason-ha · 2024-10-18T01:07:38Z

Shouldn't you be able to make many of the test changes before the breaking change? When we have a well staged deprecation, I would expect to see more removal of things we don't have to test for anymore compared to test adaptation. The latter suggests customers need to adapt too. OR we don't have good coverage on the external facing APIs which also hinders confidence.

that is not possible given the changes. the changes literally rename the class and make it un-exported

I am not sure I follow. Either you are doing non-"public" testing and still internally "export"ing the class or you shouldn't need to access the class beyond its instance type (no ctor) for testing.

anthony-murphy · 2024-10-18T01:13:35Z

Shouldn't you be able to make many of the test changes before the breaking change? When we have a well staged deprecation, I would expect to see more removal of things we don't have to test for anymore compared to test adaptation. The latter suggests customers need to adapt too. OR we don't have good coverage on the external facing APIs which also hinders confidence.

that is not possible given the changes. the changes literally rename the class and make it un-exported

I am not sure I follow. Either you are doing non-"public" testing and still internally "export"ing the class or you shouldn't need to access the class beyond its instance type (no ctor) for testing.

the test's are for implementations, they unit test the actual class which is only ever constructed within the package, consumers only ever see the interface. The tests pay the cost of dealing with the rename, and the interface get the original name, so customers don't have to change anything.

msfluid-bot

Code Coverage Summary

↓ packages.dds.merge-tree.src:
Line Coverage Change: -0.01% Branch Coverage Change: -0.01%

Metric Name	Baseline coverage	PR coverage	Coverage Diff
Branch Coverage	94.90%	94.89%	↓ -0.01%
Line Coverage	97.08%	97.07%	↓ -0.01%

↑ packages.dds.sequence.src:
Line Coverage Change: 0.07% Branch Coverage Change: -0.05%

Metric Name	Baseline coverage	PR coverage	Coverage Diff
Branch Coverage	89.47%	89.42%	↓ -0.05%
Line Coverage	89.74%	89.81%	↑ 0.07%

↑ packages.dds.sequence.src.intervalIndex:
Line Coverage Change: 0.03% Branch Coverage Change: No change

Metric Name	Baseline coverage	PR coverage	Coverage Diff
Branch Coverage	92.55%	92.55%	→ No change
Line Coverage	87.31%	87.34%	↑ 0.03%

↑ packages.dds.sequence.src.intervals:
Line Coverage Change: 0.03% Branch Coverage Change: No change

Metric Name	Baseline coverage	PR coverage	Coverage Diff
Branch Coverage	89.47%	89.47%	→ No change
Line Coverage	96.22%	96.25%	↑ 0.03%

Baseline commit: 30003a8
Baseline build: 303240
Happy Coding!!

Code coverage comparison check passed!!

…to test/internal-merge-tree-client

github-actions · 2024-10-30T17:31:58Z

🔗 No broken links found! ✅

Your attention to detail is admirable.

linkcheck output


> [email protected] ci:linkcheck /home/runner/work/FluidFramework/FluidFramework/docs
> start-server-and-test ci:start 1313 linkcheck:full

1: starting server using command "npm run ci:start"
and when url "[ 'http://127.0.0.1:1313' ]" is responding with HTTP status code 200
running tests using command "npm run linkcheck:full"


> [email protected] ci:start
> http-server ./public --port 1313 --silent


> [email protected] linkcheck:full
> npm run linkcheck:fast -- --external


> [email protected] linkcheck:fast
> linkcheck http://localhost:1313 --skip-file skipped-urls.txt --external

Crawling...

Stats:
  440714 links
    3397 destination URLs
       2 URLs ignored
       0 warnings
       0 errors

anthony-murphy · 2024-10-30T23:55:58Z

packages/dds/merge-tree/api-report/merge-tree.legacy.alpha.api.md

@@ -87,104 +78,6 @@ export abstract class BaseSegment implements ISegment {
    wasMovedOnInsert?: boolean | undefined;
 }

-// @alpha @deprecated (undocumented)
-export class Client extends TypedEventEmitter<IClientEvents> {


@tylerbutler do you know why these type removals don't break the type tests? i would expect those to need suppression, but they don't for some reason.

i guess it makes sense if the look at the imports

import type * as old from "@fluidframework/merge-tree-previous/internal"; import type * as current from "../../index.js";

as the types still exist off internal, just not legacy, so are still available off import type * as current from "../../index.js";

however, i would expect this to be a breaking change, as something that was exported at one layer, is no longer exported.

@CraigMacomber as type test collaborator.
I think there was a task out there to test against the entrypoints directly. I think that may have been mutated to read the tags so that all pulled into a single file without any duplication. I still think that using the entrypoints directly has merit. Especially relevant if we hand craft any entrypoints, which would allow deviation from tags.

jason-ha

I looked over this more and tried to separate more work that can be done right now as I had been mentioning in other comments. From that I produced #22948. Have a look.
I also have the remainder of the changes in a stash. Final result is not identical to this as I made some fixes.
I think the ideal process for these overall changes are:

improvement(client): deprecation cleanup #22948
just breaks (blocked on 2.10 opening)
more taking advantage of the breaks (change internals)

The healthiest way to maintain 2 & 3 over time is likely to be having branch with two commits and rebasing over main when desired.

packages/dds/sequence/src/sequenceDeltaEvent.ts

jason-ha · 2024-10-31T09:59:14Z

packages/dds/merge-tree/api-report/merge-tree.legacy.alpha.api.md

@@ -87,104 +78,6 @@ export abstract class BaseSegment implements ISegment {
    wasMovedOnInsert?: boolean | undefined;
 }

-// @alpha @deprecated (undocumented)
-export class Client extends TypedEventEmitter<IClientEvents> {


@CraigMacomber as type test collaborator.
I think there was a task out there to test against the entrypoints directly. I think that may have been mutated to read the tags so that all pulled into a single file without any duplication. I still think that using the entrypoints directly has merit. Especially relevant if we hand craft any entrypoints, which would allow deviation from tags.

packages/dds/sequence/api-report/sequence.legacy.alpha.api.md

anthony-murphy added 6 commits September 17, 2024 09:32

Remove Exposure of Client

2cc6221

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

2129908

…to prototype-deprecate-client

update type tests

1c687be

fix tests

8d7ef70

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

620f950

…to test/internal-merge-tree-client

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

00be065

…to test/internal-merge-tree-client

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

2ae3b9e

…to test/internal-merge-tree-client

anthony-murphy force-pushed the test/internal-merge-tree-client branch from 7e8d1c7 to 2ae3b9e Compare October 2, 2024 19:56

anthony-murphy and others added 10 commits October 3, 2024 13:37

Merge branch 'main' into test/internal-merge-tree-client

19691c2

Merge branch 'main' into test/internal-merge-tree-client

8b0ecb3

Merge branch 'main' into test/internal-merge-tree-client

e374f9a

Merge branch 'main' into test/internal-merge-tree-client

1cb5de6

Merge branch 'main' into test/internal-merge-tree-client

488b0d2

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

441a6a9

…to test/internal-merge-tree-client

remove some deprecations

36ae191

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

bc72f79

…to test/internal-merge-tree-client

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

20eb3df

…to test/internal-merge-tree-client

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

f1d2579

…to test/internal-merge-tree-client

jason-ha reviewed Oct 16, 2024

View reviewed changes

packages/dds/merge-tree/api-report/merge-tree.legacy.alpha.api.md Outdated Show resolved Hide resolved

anthony-murphy added 2 commits October 17, 2024 10:19

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

6444ced

…to test/internal-merge-tree-client

Add IMergeTreeOptionsInternal

887dfca

jason-ha reviewed Oct 18, 2024

View reviewed changes

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

f096b9c

…to test/internal-merge-tree-client

jason-ha reviewed Oct 18, 2024

View reviewed changes

msfluid-bot reviewed Oct 18, 2024

View reviewed changes

anthony-murphy added 4 commits October 25, 2024 11:06

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

966a8cd

…to test/internal-merge-tree-client

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

ac52ee6

…to test/internal-merge-tree-client

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

c32cefb

…to test/internal-merge-tree-client

Merge branch 'main' of https://github.com/microsoft/FluidFramework in…

a898922

…to test/internal-merge-tree-client

anthony-murphy commented Oct 30, 2024

View reviewed changes

jason-ha reviewed Oct 31, 2024

View reviewed changes

anthony-murphy added 3 commits October 31, 2024 09:55

fix some imports

9b443b7

fix up event typing

636b4de

improve the sequence interval interface

25c42b4

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Test/internal merge tree client #22697

Test/internal merge tree client #22697

anthony-murphy commented Oct 1, 2024 •

edited

Loading

msfluid-bot commented Oct 1, 2024 •

edited

Loading

anthony-murphy commented Oct 16, 2024 •

edited

Loading

jason-ha Oct 18, 2024

anthony-murphy Oct 18, 2024 •

edited

Loading

jason-ha left a comment

jason-ha Oct 18, 2024

anthony-murphy Oct 18, 2024

jason-ha Oct 18, 2024

anthony-murphy Oct 18, 2024

anthony-murphy commented Oct 18, 2024

jason-ha commented Oct 18, 2024

anthony-murphy commented Oct 18, 2024 •

edited

Loading

jason-ha commented Oct 18, 2024

anthony-murphy commented Oct 18, 2024 •

edited

Loading

msfluid-bot left a comment •

edited

Loading

github-actions bot commented Oct 30, 2024

anthony-murphy Oct 30, 2024

anthony-murphy Oct 30, 2024 •

edited

Loading

jason-ha Oct 31, 2024

jason-ha left a comment

jason-ha Oct 31, 2024

Test/internal merge tree client #22697

Are you sure you want to change the base?

Test/internal merge tree client #22697

Conversation

anthony-murphy commented Oct 1, 2024 • edited Loading

msfluid-bot commented Oct 1, 2024 • edited Loading

anthony-murphy commented Oct 16, 2024 • edited Loading

Choose a reason for hiding this comment

anthony-murphy Oct 18, 2024 • edited Loading

Choose a reason for hiding this comment

jason-ha 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

anthony-murphy commented Oct 18, 2024

jason-ha commented Oct 18, 2024

anthony-murphy commented Oct 18, 2024 • edited Loading

jason-ha commented Oct 18, 2024

anthony-murphy commented Oct 18, 2024 • edited Loading

msfluid-bot left a comment • edited Loading

Choose a reason for hiding this comment

Code Coverage Summary

Code coverage comparison check passed!!

github-actions bot commented Oct 30, 2024

linkcheck output

Choose a reason for hiding this comment

anthony-murphy Oct 30, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jason-ha left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

anthony-murphy commented Oct 1, 2024 •

edited

Loading

msfluid-bot commented Oct 1, 2024 •

edited

Loading

anthony-murphy commented Oct 16, 2024 •

edited

Loading

anthony-murphy Oct 18, 2024 •

edited

Loading

anthony-murphy commented Oct 18, 2024 •

edited

Loading

anthony-murphy commented Oct 18, 2024 •

edited

Loading

msfluid-bot left a comment •

edited

Loading

anthony-murphy Oct 30, 2024 •

edited

Loading