fix(form): namespace schema $id by content to avoid stale validator#4067
fix(form): namespace schema $id by content to avoid stale validator#4067
Conversation
|
Caution Review failedPull request was closed or merged during review No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (2)
📝 WalkthroughWalkthroughThe form component now preprocesses schemas before validation and rendering by replacing the ChangesSchema Identity Resolution
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related issues
Possibly related PRs
Suggested labels
Suggested reviewers
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Tip 💬 Introducing Slack Agent: The best way for teams to turn conversations into code.Slack Agent is built on CodeRabbit's deep understanding of your code, so your team can collaborate across the entire SDLC without losing context.
Built for teams:
One agent for your entire SDLC. Right inside Slack. Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
Documentation has been published to https://lundalogik.github.io/lime-elements/versions/PR-4067/ |
FredrikWallstrom
force-pushed
the
fix/form-stale-validator-on-id-collision
branch
2 times, most recently
from
591da52 to
8948275
Compare
Ajv8 caches compiled validators by `$id`. When a consumer reuses the same `$id` (typical for server-supplied schemas) with different content between renders — e.g. by filtering its `properties` based on user input — Ajv returns the validator it compiled for the first variant, which silently rejects the new variant's properties as additional or otherwise reports stale errors. Wrap the schema in a copy whose `$id` becomes `<original-$id>-<fingerprint>`, where the fingerprint is a short hash of the stringified content (sans `$id`). Distinct content gets a distinct cache entry; identical content still hits the cache. The fingerprint — rather than the raw content — keeps `$id` short and URI-safe, since the validator parses it as a URI and chokes on JSON-shaped characters.
FredrikWallstrom
force-pushed
the
fix/form-stale-validator-on-id-collision
branch
from
8948275 to
8652ac6
Compare
| let fingerprint = 5381; | ||
| for (let i = 0; i < json.length; i++) { | ||
| fingerprint = | ||
| ((fingerprint << 5) + fingerprint + json.codePointAt(i)) & | ||
| 0xff_ff_ff_ff; | ||
| } | ||
| const contentId = (fingerprint >>> 0).toString(36); |
There was a problem hiding this comment.
I tried to keep it simple and just use JSON.stringify (or URLEncode) as $id, but unfortunately the $id became too big, which caused Lime Admin to crash on big schemas.
There was a problem hiding this comment.
So the ID would be a JSON encoded string of the schema itself? 😄
There was a problem hiding this comment.
yeah, that actually worked (but not on big schemas though).
|
🎉 This PR is included in version 39.23.1 🎉 The release is available on: Your semantic-release bot 📦🚀 |
3 participants
Summary
When a consumer renders
limel-formwith a schema that reuses the same$idwhile its content changes between renders (e.g. a server schema whosepropertiesget filtered based on user input), the form'svalidateevent would fire with errors that don't match the schema's actual content — most visibly, fields would be reported as "additional properties" even though they're declared inproperties.The root cause is that the validator caches compiled validators by
$id. The first variant compiled under a given$idis reused for every later variant with the same$id, so swapping the schema's content silently has no effect on validation.This fix wraps the schema in a copy whose
$idbecomes<original-$id>-<fingerprint>, where the fingerprint is a short hash of the stringified content (sans$id), before handing it to RJSF and the validator. Distinct content produces a distinct cache entry; identical content still hits the cache. The fingerprint — rather than the raw content — keeps$idshort and URI-safe, since the validator parses it as a URI and chokes on JSON-shaped characters. The consumer's original schema is left untouched.Fixes Lundalogik/crm-client#971.
Test plan
form.e2e.tsxcovers the scenario (same$id, schema gains a required property, value updates to satisfy it).Summary by CodeRabbit
Bug Fixes
Tests