Repository navigation
Conversation
Update website/docs for behavior that differs between v2.4.0-rc.1 and v2.4.0-rc.2: acceleration settings applied on restart, deletion_mode auto, snapshots and the startup check under runtime.state, results cache aging on TTL, non-deterministic queries, Server-Timing, Flight error codes and TLS handshake timeout, HTTP rate control across shared origins and instances, Iceberg Hadoop metadata, NSQL reasoning_effort, /v1/decisions, and rrf() named arguments. Sync the OpenAPI spec for rc.2 API changes and regenerate the affected API pages. Remove the v2.4.0-rc.1 migration and history notes so the pages describe current behavior only.
Contributor
✅ Pull with Spice PassedPassing checks:
|
Contributor
🔍 Pull with Spice FailedPassing checks:
Failed checks:
Please address these issues and update your pull request. |
Contributor
|
🚀 deployed to https://34318dc4.spiceai-org-website.pages.dev |
Contributor
There was a problem hiding this comment.
🟡 Changes recommended
The namespace response schema contradicts runtime serialization and needs correction across generated API representations.
4 open findings
What changed in this PR
Updates the website documentation and generated API reference for v2.4.0-rc.2 behavior.
Changes:
- Documents acceleration, caching, runtime state, connector, and API changes.
- Adds Decisions API reference pages and refreshes affected schemas.
- Removes rc.1-specific migration and history notes.
| File | Description |
|---|---|
| website/docs/troubleshooting/index.md | Documents Flight error codes. |
| website/docs/reference/sql/search.md | Clarifies search limits and nested arguments. |
| website/docs/reference/sql/prepared_statements.md | Explains authenticated session ownership. |
| website/docs/reference/spicepod/runtime.md | Updates rate control and state storage guidance. |
| website/docs/features/views/index.md | Documents startup dependencies. |
| website/docs/features/observability/index.md | Updates cache metrics. |
| website/docs/features/large-language-models/tools.md | Describes Responses API tool rounds. |
| website/docs/features/large-language-models/parameter_overrides.md | Documents NSQL reasoning effort. |
| website/docs/features/large-language-models/mcp.md | Removes historical environment note. |
| website/docs/features/large-language-models/decisions.md | Updates decision behavior and task history. |
| website/docs/features/distributed-query/index.md | Documents missing-graph recovery failure. |
| website/docs/features/data-acceleration/snapshots.md | Documents runtime-state snapshot storage. |
| website/docs/features/data-acceleration/partitioning.md | Explains automatic partition rebuilds. |
| website/docs/features/data-acceleration/data-refresh.md | Documents acceleration creation retries. |
| website/docs/features/caching/index.md | Revises cache eligibility, expiration, and invalidation. |
| website/docs/components/tools/mcp.md | Removes historical environment note. |
| website/docs/components/data-connectors/sharepoint.md | Clarifies folder rows and nullable fields. |
| website/docs/components/data-connectors/iceberg.md | Updates metadata discovery guidance. |
| website/docs/components/data-connectors/huggingface.md | Documents retries and resumed downloads. |
| website/docs/components/data-connectors/https/deployment.md | Updates shared-origin configuration guidance. |
| website/docs/components/data-accelerators/cayenne/performance.md | Clarifies unsupported tuning parameters. |
| website/docs/components/data-accelerators/cayenne/index.md | Updates deletion strategies and settings changes. |
| website/docs/components/data-accelerators/cayenne/deployment.md | Updates compaction troubleshooting. |
| website/docs/components/catalogs/iceberg.md | Clarifies Hadoop metadata discovery. |
| website/docs/api/tls/index.md | Documents Flight handshake timeout. |
| website/docs/api/overview.md | Explains server timing headers and trailers. |
| website/docs/api/HTTP/sidebar.ts | Updates API navigation. |
| website/docs/api/HTTP/post-sql.StatusCodes.json | Adds server timing documentation. |
| website/docs/api/HTTP/post-sql.api.mdx | Refreshes embedded SQL operation. |
| website/docs/api/HTTP/post-nsql.RequestSchema.json | Adds reasoning effort schema. |
| website/docs/api/HTTP/post-nsql.api.mdx | Refreshes embedded NSQL operation. |
| website/docs/api/HTTP/post-decisions.StatusCodes.json | Adds decision response schemas. |
| website/docs/api/HTTP/post-decisions.RequestSchema.json | Adds decision request schema. |
| website/docs/api/HTTP/post-decisions.api.mdx | Adds Decisions API page. |
| website/docs/api/HTTP/post-chat-completions.StatusCodes.json | Updates response types and errors. |
| website/docs/api/HTTP/post-chat-completions.RequestSchema.json | Updates OpenAI request types. |
| website/docs/api/HTTP/post-chat-completions.api.mdx | Refreshes embedded chat operation. |
| website/docs/api/HTTP/get-table.StatusCodes.json | Revises Iceberg table responses. |
| website/docs/api/HTTP/get-table.api.mdx | Documents table loading restrictions. |
| website/docs/api/HTTP/get-namespace.StatusCodes.json | Adds namespace response body schema. |
| website/docs/api/HTTP/get-namespace.api.mdx | Updates namespace loading reference. |
| website/docs/api/HTTP/get-iceberg-namespaces.StatusCodes.json | Updates Iceberg error types. |
| website/docs/api/HTTP/get-iceberg-namespaces.api.mdx | Refreshes embedded namespace-list operation. |
🧠 Review effort: Balanced
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
2 tasks done
This branch was successfully deployed
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
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.
Summary
Updates
website/docs/for behavior that differs between v2.4.0-rc.1 and v2.4.0-rc.2. Each change maps to a spiceai/spiceai PR in the rc.2 changelog. Behavior that rc.1 already had is not added.Acceleration:
cayenne_deletion_mode: autouses position deletes for small primary-key tables on local disk, which corrects docs(cayenne): cayenne_deletion_mode auto resolves to key on every primary-key table #2354 (feat(cayenne): let deletion_mode auto pick position deletes for small primary-key tables spiceai#15042).runtime.state: snapshots are stored undersnapshots/(feat(snapshots): store snapshots under runtime.state, built by the shared state-location builder spiceai#15117); the startup check uses_spice/format.json(feat(runtime): checkruntime.stateat startup, and record its state format spiceai#15138).Caching: SQL results age out on
item_ttland a refresh no longer invalidates them; the removed metric values are dropped (feat(cache): age SQL results on TTL instead of invalidating them on refresh spiceai#15033). Queries with non-deterministic functions are not cached (fix: Cayenne deletion-index bounds and reload fencing, scan-statistics single-flight, UDF filter pushdown, Server-Timing, results-cache determinism, and view loading spiceai#14993).APIs:
Server-Timingon SQL query responses (fix: Cayenne deletion-index bounds and reload fencing, scan-statistics single-flight, UDF filter pushdown, Server-Timing, results-cache determinism, and view loading spiceai#14993).INVALID_ARGUMENTfor query and data errors (fix(flight): return query and data errors as InvalidArgument, not Internal spiceai#14956).Connectors:
hf://datasets/...) spiceai#14877, fix(connector-huggingface): retry Hub API responses that break off part way, and harden the live-Hub check spiceai#15009).AI and search:
/v1/nsqlreasoning_effort(feat(nsql): accept reasoning_effort on POST /v1/nsql spiceai#14852)./v1/responsestool rounds (fix(models):/v1/responsesruns Spice tools and honorstool_recursion_limitwhen streaming (fixes #14905, #14631) spiceai#14930)./v1/decisionsresponsemodel, the single-labelchoice, and task history (fix(models): allowed_tools shape, provider error statuses, decision response model (fixes #14910) spiceai#14977, fix(decisions): Fixes for /v1/decisions spiceai#15040).vector_searchlimiton chunked and multi-vector columns (fix(search): vector_search returns its top k rows by score on chunked and multi-vector columns (fixes #14819) spiceai#15056).rrf()(fix(search): keep named arguments on search calls nested in rrf() (fixes #14923) spiceai#14975).API reference:
openapi.jsongains only rc.2's API changes, taken from the regenerated spec in Update openapi.json spiceai#15035:/v1/decisions, NSQLreasoning_effort, the Iceberg namespace and table responses, andServer-Timing;The affected API pages are regenerated.
Current state only: removes the v2.4.0-rc.1 migration and history notes from the decisions, MCP, runtime, Cayenne, data refresh, snapshots and Iceberg pages.
Verification
Behavior was run against a v2.4.0-rc.2 build where feasible. Names and defaults were checked against spiceai trunk
37ed679674, and each statement against thev2.4.0-rc.1source. These were read in source and not run:DoGetserver-timingtrailer;npm run buildpasses with no broken links or anchors.