feat: support dot-path in slicingArguments - #2801
Merged
ysmolski merged 24 commits intoMay 18, 2026
Merged
Conversation
ysmolski
requested review from
Aenimus,
JivusAyrus,
StarpTech,
thisisnithin and
wilsonrivera
as code owners
April 28, 2026 09:28
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
There was a problem hiding this comment.
Claude Code Review
This repository is configured for manual code reviews. Comment @claude review to trigger a review and subscribe this PR to future pushes, or @claude review once for a one-time review.
Tip: disable this comment in your organization's Code Review settings.
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
🚧 Files skipped from review as they are similar to previous changes (1)
WalkthroughThis PR extends the ChangesNested-Path Slicing Arguments Support
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Possibly related PRs
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
Warning Review ran into problems🔥 ProblemsGit: Failed to clone repository. Please run the Comment |
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #2801 +/- ##
==========================================
+ Coverage 1.49% 46.50% +45.00%
==========================================
Files 315 1085 +770
Lines 44219 146198 +101979
Branches 451 9413 +8962
==========================================
+ Hits 663 67990 +67327
- Misses 43252 76468 +33216
- Partials 304 1740 +1436
🚀 New features to boost your workflow:
|
Router-nonroot image scan passed✅ No security vulnerabilities found in image: |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@composition/src/v1/normalization/normalization-factory.ts`:
- Around line 2665-2670: The code currently uses only the first segment
(firstSegment) when reporting a missing root argument, which loses the full SDL
path; change the error creation to pass the full configured argument path (e.g.,
the original slicing argument string or segments.join(".")/slicingArg) into
listSizeInvalidSlicingArgumentErrorMessage instead of firstSegment so the
diagnostic shows the full path; update the call site where argData is checked
(the block referencing segments, firstSegment, argData, errorMessages, and
listSizeInvalidSlicingArgumentErrorMessage) to supply the full path string and
ensure any helpers expecting just the name are adjusted to accept the dotted
path.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 3479e173-63e7-462f-aee0-ad2ba57ad429
📥 Commits
Reviewing files that changed from the base of the PR and between f5929be and 28c3064852a3a0851bc10d6f3aa1a9dfa1e9f1c8.
📒 Files selected for processing (4)
composition-go/index.global.jscomposition/src/errors/errors.tscomposition/src/v1/normalization/normalization-factory.tscomposition/tests/v1/directives/listSize.test.ts
Base automatically changed from
yury/eng-8700-router-support-cost-on-the-arguments-of-directives
to
main
April 29, 2026 11:28
ysmolski
requested review from
a team,
Noroth,
SkArchon,
devsergiy and
endigma
as code owners
April 29, 2026 11:28
ysmolski
force-pushed
the
yury/eng-9338-router-support-dot-path-in-slicingarguments
branch
from
May 6, 2026 13:49
c9873ed to
6bf922d
Compare
Contributor
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
composition/src/errors/errors.ts (1)
1789-1797:⚠️ Potential issue | 🟡 Minor | ⚡ Quick winGeneralize the non-
Intmessage for nested paths.With dot-path support, this can now describe a nested input field rather than a top-level argument, so the current
"references an argument of type"wording is inaccurate for valid new inputs likeinput.pagination.first.Suggested wording
export function listSizeSlicingArgumentNotIntErrorMessage( directiveCoords: DirectiveArgumentCoords, argumentName: ArgumentName, actualType: TypeName, ): string { return ( - ` The "slicingArguments" value "${argumentName}" on "${directiveCoords}" references an argument of type` + + ` The "slicingArguments" value "${argumentName}" on "${directiveCoords}" references a value of type` + ` "${actualType}", but slicing arguments must be of type "Int" or "Int!".` ); }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@composition/src/errors/errors.ts` around lines 1789 - 1797, The error string in listSizeSlicingArgumentNotIntErrorMessage is specific to top-level arguments ("references an argument of type") but must cover nested dot-paths (input fields) as well; update the returned message to use a generalized phrase like "references a value of type" or "references an input value of type" so the function listSizeSlicingArgumentNotIntErrorMessage(directiveCoords, argumentName, actualType) correctly describes both arguments and nested input fields when actualType is not Int/Int!.
🧹 Nitpick comments (2)
composition/src/v1/normalization/normalization-factory.ts (1)
2669-2670: Gate dottedslicingArgumentson router support.These paths are now serialized into
costs.listSizesunchanged. If composition ships before router support, older routers can ignore them and start rejecting requests for fields that still require a slicing argument. Consider enforcing this with a compatibility check, feature gate, or rollout guard instead of relying on release ordering alone.Also applies to: 2758-2759
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@composition/src/v1/normalization/normalization-factory.ts` around lines 2669 - 2670, Paths are being unconditionally added to listSizeConfig.slicingArguments and slicingArgChainByPath (slicingArgPath / slicingArgChainByPath), which causes costs.listSizes to serialize slicing arguments even when downstream routers may not support them; add a compatibility gate before pushing/setting these values: detect router support (e.g., via a feature flag or a runtime check like routerSupportsSlicingArgs(costs) or enableRouterSlicingArgs) and only call listSizeConfig.slicingArguments.push(slicingArgPath) and slicingArgChainByPath.set(slicingArgPath, chain) when the gate returns true; apply the same guarded change for the identical block around lines where slicing arguments are added (the other occurrence at the reported range 2758-2759).composition/tests/v1/directives/listSize.test.ts (1)
655-676: ⚡ Quick winAdd a normalized-SDL assertion for a nested-path happy path.
The new success cases only verify
costs.listSizes, so a regression in emitted composition output would still pass. Please mirror the earlier normalization checks with at least one assertion thatschemaToSortedNormalizedString(schema)preserves@listSize(slicingArguments: ["input.pagination.first"])—ideally also for the mixed flat+nested case.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@composition/tests/v1/directives/listSize.test.ts` around lines 655 - 676, Add assertions that verify the normalized SDL still emits the `@listSize` directive with the nested slicingArguments: after the existing checks in the "nested-path slicingArguments tests" (in the tests using subgraphWithDeepNestedSlicingArg and subgraphWithMixedSlicingArgs / ROUTER_COMPATIBILITY_VERSION_ONE), call schemaToSortedNormalizedString(schema) on the returned schema and assert the string contains '@listSize(slicingArguments: ["input.pagination.first"])' for the deep-nested case and contains the corresponding combined form (e.g. '@listSize(slicingArguments: ["limit","input.pagination.first"])') for the mixed case; keep these assertions alongside the existing checks that reference costs.listSizes.get('Query.search') so the test covers both runtime cost normalization and the emitted normalized SDL.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@composition/src/errors/errors.ts`:
- Around line 1789-1797: The error string in
listSizeSlicingArgumentNotIntErrorMessage is specific to top-level arguments
("references an argument of type") but must cover nested dot-paths (input
fields) as well; update the returned message to use a generalized phrase like
"references a value of type" or "references an input value of type" so the
function listSizeSlicingArgumentNotIntErrorMessage(directiveCoords,
argumentName, actualType) correctly describes both arguments and nested input
fields when actualType is not Int/Int!.
---
Nitpick comments:
In `@composition/src/v1/normalization/normalization-factory.ts`:
- Around line 2669-2670: Paths are being unconditionally added to
listSizeConfig.slicingArguments and slicingArgChainByPath (slicingArgPath /
slicingArgChainByPath), which causes costs.listSizes to serialize slicing
arguments even when downstream routers may not support them; add a compatibility
gate before pushing/setting these values: detect router support (e.g., via a
feature flag or a runtime check like routerSupportsSlicingArgs(costs) or
enableRouterSlicingArgs) and only call
listSizeConfig.slicingArguments.push(slicingArgPath) and
slicingArgChainByPath.set(slicingArgPath, chain) when the gate returns true;
apply the same guarded change for the identical block around lines where slicing
arguments are added (the other occurrence at the reported range 2758-2759).
In `@composition/tests/v1/directives/listSize.test.ts`:
- Around line 655-676: Add assertions that verify the normalized SDL still emits
the `@listSize` directive with the nested slicingArguments: after the existing
checks in the "nested-path slicingArguments tests" (in the tests using
subgraphWithDeepNestedSlicingArg and subgraphWithMixedSlicingArgs /
ROUTER_COMPATIBILITY_VERSION_ONE), call schemaToSortedNormalizedString(schema)
on the returned schema and assert the string contains
'@listSize(slicingArguments: ["input.pagination.first"])' for the deep-nested
case and contains the corresponding combined form (e.g.
'@listSize(slicingArguments: ["limit","input.pagination.first"])') for the mixed
case; keep these assertions alongside the existing checks that reference
costs.listSizes.get('Query.search') so the test covers both runtime cost
normalization and the emitted normalized SDL.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: af468a49-15fd-4b85-baa3-b5b70aea3fd3
📥 Commits
Reviewing files that changed from the base of the PR and between 28c3064852a3a0851bc10d6f3aa1a9dfa1e9f1c8 and 6bf922d.
📒 Files selected for processing (4)
composition-go/index.global.jscomposition/src/errors/errors.tscomposition/src/v1/normalization/normalization-factory.tscomposition/tests/v1/directives/listSize.test.ts
ysmolski
changed the title
…8-router-support-dot-path-in-slicingarguments
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@composition/src/v1/normalization/normalization-factory.ts`:
- Around line 2754-2758: The current check on listSizeConfig.slicingArguments
uses slicingArgChainByPath.get(...) and rejects the path if chain.some(node =>
node.defaultValue) — instead, for each node in the chain that has a defaultValue
you must try to resolve the remainder of the dotted path through that node's
defaultValue AST and only treat it as a blocking default if that resolution
actually yields a leaf value for the targeted slicing argument; if the ancestor
defaultValue does not contain/resolve the remaining path segments, it should not
trigger listSizeAssumedSizeSlicingArgDefaultErrorMessage(directiveCoords,
slicingArgPath). Implement a resolver that walks the defaultValue AST along the
tail path from that node and only push the error when the resolved value is
present (or explicitly null/defined per current semantics).
- Around line 2625-2639: The current branch conflates "undefined parent type"
and "parent type not an input object", causing a duplicate segmentNotInputObject
error; update the logic around
parentDefinitionDataByTypeName.get(current.namedTypeName) so that: if
unwrapped.kind === Kind.LIST_TYPE keep the existing error; else if
parentTypeData is undefined set isPathInvalid = true but do NOT call
listSizeSlicingArgumentSegmentNotInputObjectErrorMessage; else if
parentTypeData.kind !== Kind.INPUT_OBJECT_TYPE_DEFINITION then push the
listSizeSlicingArgumentSegmentNotInputObjectErrorMessage and set isPathInvalid =
true. Reference symbols: parentDefinitionDataByTypeName, current, unwrapped,
Kind.LIST_TYPE, parentTypeData.kind, Kind.INPUT_OBJECT_TYPE_DEFINITION,
listSizeSlicingArgumentSegmentNotInputObjectErrorMessage, isPathInvalid.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 8f90da03-7acc-4721-84d6-ec00d8609a7f
📒 Files selected for processing (1)
composition/src/v1/normalization/normalization-factory.ts
Aenimus
reviewed
May 7, 2026
…8-router-support-dot-path-in-slicingarguments
ysmolski
changed the title
…8-router-support-dot-path-in-slicingarguments
dkorittki
approved these changes
May 15, 2026
dkorittki
left a comment
dkorittki
left a comment
Contributor
There was a problem hiding this comment.
lgtm. Added some comments for optional changes. Up to you if you want to follow them, it's nothing critical.
…8-router-support-dot-path-in-slicingarguments
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
router/go.mod (1)
46-47:⚠️ Potential issue | 🟠 Major | 🏗️ Heavy lift
router/go.modpins vulnerableotel/sdkversions in multiple locations.Lines 46, 47, 192, and 193 all pin
go.opentelemetry.io/otel/sdk(and its/metricsubmodule) below the required minimum ofv1.40.0. Update all four tov1.40.0or later:Proposed patch
- go.opentelemetry.io/otel/sdk v1.39.0 + go.opentelemetry.io/otel/sdk v1.40.0 - go.opentelemetry.io/otel/sdk/metric v1.39.0 + go.opentelemetry.io/otel/sdk/metric v1.40.0 ... - go.opentelemetry.io/otel/sdk => go.opentelemetry.io/otel/sdk v1.28.0 + go.opentelemetry.io/otel/sdk => go.opentelemetry.io/otel/sdk v1.40.0 - go.opentelemetry.io/otel/sdk/metric => go.opentelemetry.io/otel/sdk/metric v1.28.0 + go.opentelemetry.io/otel/sdk/metric => go.opentelemetry.io/otel/sdk/metric v1.40.0🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@router/go.mod` around lines 46 - 47, Update the pinned versions of the OpenTelemetry SDK modules: replace every occurrence of go.opentelemetry.io/otel/sdk and go.opentelemetry.io/otel/sdk/metric in go.mod with v1.40.0 (or later) so all references meet the minimum required version; ensure you update all instances (both top-level and metric submodule entries) to the same new version and run go mod tidy to refresh the go.sum.
♻️ Duplicate comments (1)
composition/src/v1/normalization/normalization-factory.ts (1)
2627-2641:⚠️ Potential issue | 🟡 Minor | ⚡ Quick winSkip the extra
segmentNotInputObjecterror when the intermediate type is undefined.If
current.namedTypeNameis missing fromparentDefinitionDataByTypeName, normalization already reports that undefined type later. EmittinglistSizeSlicingArgumentSegmentNotInputObjectErrorMessagehere adds a second, misleading@listSizeerror for the same root cause.Suggested adjustment
const unwrapped = current.type.kind === Kind.NON_NULL_TYPE ? current.type.type : current.type; const parentTypeData = this.parentDefinitionDataByTypeName.get(current.namedTypeName); + if (!parentTypeData && !BASE_SCALARS.has(current.namedTypeName)) { + isPathInvalid = true; + break; + } if ( unwrapped.kind === Kind.LIST_TYPE || !parentTypeData || parentTypeData.kind !== Kind.INPUT_OBJECT_TYPE_DEFINITION ) {🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@composition/src/v1/normalization/normalization-factory.ts` around lines 2627 - 2641, The code is emitting listSizeSlicingArgumentSegmentNotInputObjectErrorMessage even when parentDefinitionDataByTypeName lacks an entry for current.namedTypeName; change the conditional so the segment-not-input-object error is only pushed when parentTypeData is present but its kind is not Kind.INPUT_OBJECT_TYPE_DEFINITION (and unwrapped.kind is not Kind.LIST_TYPE), i.e. check parentTypeData existence separately and avoid calling listSizeSlicingArgumentSegmentNotInputObjectErrorMessage when parentTypeData is undefined; update the branch around parentDefinitionDataByTypeName.get(current.namedTypeName), unwrapped, Kind.LIST_TYPE, listSizeSlicingArgumentSegmentNotInputObjectErrorMessage, directiveCoords, slicingArgPath, current.name, printTypeNode(...) and isPathInvalid accordingly.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@router-tests/go.mod`:
- Around line 36-37: Update the pinned OpenTelemetry SDK versions to mitigate
CVE-2026-24051 by changing any instances of go.opentelemetry.io/otel/sdk and
go.opentelemetry.io/otel/sdk/metric from v1.39.0 (and any replace entries at
v1.28.0) to v1.40.0 or later in router-tests/go.mod; ensure you update both the
require block and any replace directives referring to these modules so they
consistently use >= v1.40.0 and then run go mod tidy to refresh go.sum.
---
Outside diff comments:
In `@router/go.mod`:
- Around line 46-47: Update the pinned versions of the OpenTelemetry SDK
modules: replace every occurrence of go.opentelemetry.io/otel/sdk and
go.opentelemetry.io/otel/sdk/metric in go.mod with v1.40.0 (or later) so all
references meet the minimum required version; ensure you update all instances
(both top-level and metric submodule entries) to the same new version and run go
mod tidy to refresh the go.sum.
---
Duplicate comments:
In `@composition/src/v1/normalization/normalization-factory.ts`:
- Around line 2627-2641: The code is emitting
listSizeSlicingArgumentSegmentNotInputObjectErrorMessage even when
parentDefinitionDataByTypeName lacks an entry for current.namedTypeName; change
the conditional so the segment-not-input-object error is only pushed when
parentTypeData is present but its kind is not Kind.INPUT_OBJECT_TYPE_DEFINITION
(and unwrapped.kind is not Kind.LIST_TYPE), i.e. check parentTypeData existence
separately and avoid calling
listSizeSlicingArgumentSegmentNotInputObjectErrorMessage when parentTypeData is
undefined; update the branch around
parentDefinitionDataByTypeName.get(current.namedTypeName), unwrapped,
Kind.LIST_TYPE, listSizeSlicingArgumentSegmentNotInputObjectErrorMessage,
directiveCoords, slicingArgPath, current.name, printTypeNode(...) and
isPathInvalid accordingly.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: d481c4e0-aaaf-4cfd-b3e1-0934ec4a4e19
⛔ Files ignored due to path filters (2)
router-tests/go.sumis excluded by!**/*.sumrouter/go.sumis excluded by!**/*.sum
📒 Files selected for processing (8)
composition/src/errors/errors.tscomposition/src/v1/normalization/normalization-factory.tscomposition/tests/v1/directives/listSize.test.tsdocs-website/router/security/cost-control.mdxrouter-tests/go.modrouter-tests/operations/testdata/introspection/base-graph-schema.jsonrouter-tests/testenv/testdata/configWithEdfs.jsonrouter/go.mod
✅ Files skipped from review due to trivial changes (1)
- docs-website/router/security/cost-control.mdx
🚧 Files skipped from review as they are similar to previous changes (1)
- composition/src/errors/errors.ts
endigma
approved these changes
May 18, 2026
ysmolski
deleted the
yury/eng-9338-router-support-dot-path-in-slicingarguments
branch
May 18, 2026 09:02
This was referenced May 21, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
When the list size comes from an argument nested in an input object,
use a dot-separated path in slicingArguments:
This PR contains composition and router changes Composition passes slices argiments in the the same
fields, but this time it parses and validates the dot-path for each slicingArgument.
One small caveat. We should release this in the composition after the router has support for the same feature.
Otherwise, if a user composes config with dot-path slicingArguments and feeds to the router without support of it, then router will throw the error "400" for queries on fields having
requireOneSlicingArgument == true. It is because router will ignore such dot-path fields and validation won't pass. Right now it is impossible to release router part first because it depends on the changes in composition so I will release it at the same time.I will make a note on the documentation about these features requiring specific versions of composition and router.
Summary by CodeRabbit
@listSizewith clearer, path-specific validation and error messages.