ktx/packages/llm/src/message-builder.test.ts
Andrey Avtomonov cb8902f1e5
fix(context): merge overlay columns onto manifest columns by name (#94)
* fix(context): merge overlay columns onto manifest columns by name

composeOverlay was appending overlay columns to the manifest column list,
producing duplicate entries when dbt/metabase overlays declared a column
just to attach descriptions. The duplicates carried no `type`, so the
pydantic SourceDefinition rejected them at semantic-query time and broke
`ktx sl query` for every overlay-backed measure. Now overlay columns
match base columns by name (case-insensitive): same-name entries merge
onto the manifest (overlay fields win, type/role fall back to the base,
descriptions merge per source key) and only new names append.

* refactor(sl): split overlay columns from column_overrides and enforce TS/Python wire contract

Overlay sources now have two distinct collections: `columns:` for computed
columns (requiring `expr` + `type`) and `column_overrides:` for metadata
patches to inherited manifest columns. Composing or loading an overlay that
mixes the two — or references an unknown column — fails with a typed error.

Introduce `ResolvedSemanticLayerSource` / `resolvedSourceSchema` /
`toResolvedWire` as the strict shape sent to the Python engine, and add a
schema contract test that diffs Zod against the Pydantic JSON schema dumped
by `python -m semantic_layer dump-schema`. `SourceDefinition` is now
`extra="forbid"` on the Python side.

`loadAllSources` surfaces per-file load errors instead of swallowing them,
so validation/query paths can report manifest shard parse failures.

* fix(context): make scan description generation resilient and quiet

A transient sampleTable failure during ingest used to take out every
table in a connection: generateTableDescription returned a hardcoded
'Table not found' string into descriptions.ai, and KtxDescriptionGenerator
was constructed without a logger, so the failure left no trail anywhere.

- sampleTable / sampleColumn calls retry 3x with 200/400/800ms backoff,
  honouring KtxScanContext.signal via a new KtxAbortedError.
- On retry exhaustion or missing capability, table generation falls back
  to a metadata-only prompt built from column name / native type / comment
  / rawDescriptions. The column path follows the same rule -- call the
  LLM when any of samples or rawDescriptions are available; skip only
  when both are absent.
- Logger is now threaded from KtxScanContext into the generator. Failures
  emit structured KtxScanWarning entries (new description_fallback_used
  code, plus existing sampling_failed / enrichment_failed /
  connector_capability_missing). ktx scan groups warnings by code so a
  batch of identical failures collapses to one summary line plus sample.
- Returns null on failure instead of the 'Table not found' sentinel; the
  manifest writer's existing guard already skips empty descriptions, so
  schema YAML no longer carries misleading text. SCAN_MANAGED_DESCRIPTION_KEYS
  already strips stale 'ai' on merge, so existing YAML clears on next run.

Also suppress AI SDK v6 'system in messages' warning: pull system messages
out of KtxMessageBuilder.wrapSimple's output via a new splitKtxSystemMessages
helper and pass them top-level to generateText (preserves cacheControl
providerOptions on the SystemModelMessage). Agent-runner's local
splitSystemPromptMessages dedupes onto the shared helper.

* test(docs): align examples-docs assertions with revamped docs

PR #103 (setup/guide doc revamp) reworded several CLI examples and
connection labels; the assertions in scripts/examples-docs.test.mjs
still referenced the pre-revamp wording and were failing in CI on main.
Update the regexes to match the post-revamp content:

- drop the `--json` flag from the sl-query example expectation
- move the `Driver:` / `Status: ok` probe to the connection reference,
  which is where that output now lives (driver id is lowercase
  `postgres`, not the display name `PostgreSQL`)
- drop the obsolete `Install \`uv\`...` troubleshooting line
- accept `<connectionId>` everywhere; the docs no longer use the
  hyphenated `<connection-id>` form
- match the `warehouse` connection id used in the quickstart instead of
  the `postgres-warehouse` id only used in the README and setup ref

* fix(sl): skip TS/Python schema contract test when uv is unavailable

The TypeScript checks CI job does not install uv or Python, so the
module-level `execFileSync('uv', ...)` in schemas.contract.test.ts threw
ENOENT and failed the suite. Wrap the schema dump in a try/catch and
guard the describe block with `describe.skipIf` so the test skips in
environments without uv. Local dev and any CI job that has uv on PATH
still runs the cross-language contract assertion.
2026-05-15 02:11:04 +02:00

146 lines
5.5 KiB
TypeScript

import type { ModelMessage } from 'ai';
import { describe, expect, it } from 'vitest';
import { KtxMessageBuilder, splitKtxSystemMessages } from './message-builder.js';
import { createKtxLlmProvider } from './model-provider.js';
function makeBuilder(overrides: Parameters<typeof createKtxLlmProvider>[0]['promptCaching'] = {}) {
const provider = createKtxLlmProvider({
backend: 'gateway',
gateway: { baseURL: 'https://gateway.test' },
modelSlots: { default: 'anthropic/claude-sonnet-4-6' },
promptCaching: { enabled: true, ...overrides },
});
return new KtxMessageBuilder(provider);
}
describe('KtxMessageBuilder.build', () => {
it('caches static system, last sorted tool, and last history message', () => {
const builder = makeBuilder();
const out = builder.build({
parts: { staticSystem: 'STATIC', dynamicSystem: 'DYNAMIC' },
history: [
{ role: 'user', content: 'first' },
{ role: 'assistant', content: [{ type: 'text', text: 'reply A' }, { type: 'text', text: 'reply B' }] } as ModelMessage,
],
currentMessage: { role: 'user', content: 'now' },
tools: {
zoo: { description: 'z' },
apple: { description: 'a' },
},
model: 'anthropic/claude-sonnet-4-6',
});
expect(out.messages[0]).toMatchObject({
role: 'system',
content: 'STATIC',
providerOptions: { anthropic: { cacheControl: { type: 'ephemeral', ttl: '1h' } } },
});
expect(out.messages[1]).toMatchObject({ role: 'system', content: 'DYNAMIC' });
expect((out.messages[1] as { providerOptions?: unknown }).providerOptions).toBeUndefined();
expect((out.messages[3] as { content: Array<{ providerOptions?: unknown }> }).content[1].providerOptions).toEqual({
anthropic: { cacheControl: { type: 'ephemeral', ttl: '5m' } },
});
expect(Object.keys(out.tools)).toEqual(['apple', 'zoo']);
expect((out.tools.zoo as { providerOptions?: unknown }).providerOptions).toEqual({
anthropic: { cacheControl: { type: 'ephemeral', ttl: '1h' } },
});
});
it('wraps leading user context onto currentMessage as a system reminder part', () => {
const builder = makeBuilder();
const out = builder.build({
parts: { staticSystem: 'STATIC', leadingUserContext: 'current_date: 2026-05-04' },
history: [],
currentMessage: { role: 'user', content: 'question' },
tools: {},
model: 'anthropic/claude-sonnet-4-6',
});
expect(out.messages[out.messages.length - 1]).toMatchObject({
role: 'user',
content: [
{ type: 'text', text: '<system-reminder>\ncurrent_date: 2026-05-04\n</system-reminder>' },
{ type: 'text', text: 'question' },
],
});
});
it('omits cache markers for non-Anthropic protocol models', () => {
const builder = makeBuilder();
const out = builder.wrapSimple({
system: 'SYS',
messages: [{ role: 'user', content: 'q' }],
tools: { z: {} },
model: 'gpt-5',
});
expect((out.messages[0] as { providerOptions?: unknown }).providerOptions).toBeUndefined();
expect((out.tools.z as { providerOptions?: unknown }).providerOptions).toBeUndefined();
});
it('clamps every TTL to 5m for Vertex when vertexFallbackTo5m is enabled', () => {
const provider = createKtxLlmProvider({
backend: 'vertex',
vertex: { project: 'ktx-test', location: 'us-east5' },
modelSlots: { default: 'claude-sonnet-4-6' },
promptCaching: {
enabled: true,
systemTtl: '1h',
toolsTtl: '1h',
historyTtl: '1h',
vertexFallbackTo5m: true,
},
});
const builder = new KtxMessageBuilder(provider);
const out = builder.build({
parts: { staticSystem: 'STATIC' },
history: [{ role: 'user', content: 'history' }],
currentMessage: { role: 'user', content: 'now' },
tools: { z: {} },
model: 'claude-sonnet-4-6',
});
expect((out.messages[0] as { providerOptions: any }).providerOptions.anthropic.cacheControl.ttl).toBe('5m');
expect((out.messages[1] as { content: Array<{ providerOptions: any }> }).content[0].providerOptions.anthropic.cacheControl.ttl).toBe(
'5m',
);
expect((out.tools.z as { providerOptions: any }).providerOptions.anthropic.cacheControl.ttl).toBe('5m');
});
});
describe('splitKtxSystemMessages', () => {
it('returns undefined system when no system messages are present', () => {
const split = splitKtxSystemMessages([
{ role: 'user', content: 'hello' },
{ role: 'assistant', content: 'hi' },
]);
expect(split.system).toBeUndefined();
expect(split.messages).toHaveLength(2);
});
it('returns a single system message object when one system message is present, preserving providerOptions', () => {
const systemMessage = {
role: 'system' as const,
content: 'You are helpful.',
providerOptions: { anthropic: { cacheControl: { type: 'ephemeral' } } },
};
const split = splitKtxSystemMessages([systemMessage, { role: 'user', content: 'hello' }]);
expect(split.system).toBe(systemMessage);
expect(split.messages).toEqual([{ role: 'user', content: 'hello' }]);
});
it('returns an array of system messages when multiple are present, in order', () => {
const split = splitKtxSystemMessages([
{ role: 'system', content: 'cached' },
{ role: 'system', content: 'fresh' },
{ role: 'user', content: 'hello' },
]);
expect(Array.isArray(split.system)).toBe(true);
expect(split.system).toHaveLength(2);
expect(split.messages).toEqual([{ role: 'user', content: 'hello' }]);
});
});