ktx/packages/cli/test/context/ingest/repo-fetch.test.ts

228 lines
9.2 KiB
TypeScript
Raw Permalink Normal View History

2026-05-10 23:12:26 +02:00
import { access, mkdir, mkdtemp, readFile, rm, writeFile } from 'node:fs/promises';
import { tmpdir } from 'node:os';
import { join } from 'node:path';
import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
import { makeLocalGitRepo } from '../test/make-local-git-repo.js';
test: split cli tests from source tree (#216) * feat(cli): define full warehouse dialect contract * test(cli): keep dialect edge tests focused * fix(cli): stabilize dialect contract foundation * refactor(connectors): own read-only query preparation * refactor(connectors): resolve dialects through registry * refactor(connectors): keep concrete dialect classes internal * chore(workspace): enforce dialect import boundary * refactor(cli): resolve relationship dialect at scan boundary * refactor(cli): use dialect display parsing for entity details * refactor(cli): use dialect display parsing for warehouse catalog * refactor(cli): use dialect SQL in relationship workflows * test(cli): verify solid dialect scan workflow closure * test: split cli tests from source tree * refactor(cli): standardize BigQuery scope listing * feat(sqlite): implement connector scope listing * test(connectors): cover required table listing * feat(cli): add warehouse driver registry * refactor(setup): route scope discovery through driver registry * refactor(cli): route local query execution through driver registry * refactor(historic-sql): route dialect support through driver registry * refactor(cli): test warehouse connections through driver registry * fix(cli): close driver registry type export gaps * Improve setup daemon diagnostics * refactor(setup): centralize rail-prefixed diagnostics + query-history fallback Extract errorMessage, writePrefixedLines, and flushPrefixedBufferedCommandOutput into clack.ts so the setup wizard, managed daemons, and embedding/agent steps share one rail-formatted writer. setup-databases.ts also adds a "disable query history and retry" option when the schema-context build fails and query history is the likely culprit, surfaced via a new failed-query-history-unavailable status. * fix(cli): carry catalog through the picker so BigQuery/Snowflake/SQL Server scope filters match The setup picker's KtxTableListEntry was a 2-level { schema, name }, so qualifiedTableId always wrote db.name into enabled_tables. When BigQuery, Snowflake, or SQL Server later ran fast ingest, their introspect step filtered the scope set with scopedTableNames(scope, { catalog: projectId|database, db }) — catalog was non-null on the introspect side but null in the scope refs, so every entry was rejected, the live-database adapter staged zero table files, and detect() failed with 'Adapter "live-database" did not recognize fetched source output'. Align the picker boundary with the canonical 3-level KtxTableRef: - Add catalog: string | null to KtxTableListEntry. - BigQuery/Snowflake/SQL Server listTables populate catalog from the resolved projectId / database; Postgres/MySQL/ClickHouse/SQLite set null. - qualifiedTableId emits catalog.schema.name when catalog is non-null (resolveEnabledTables already accepts the 3-part shape) and schemasFromEnabledTables now goes through parseDottedTableEntry so it recovers the schema correctly from both 2-part and 3-part entries. - Export parseDottedTableEntry from enabled-tables.ts (@internal) for picker reuse. Update listTables expectations in all seven connector tests and the setup / picker test fixtures. Add a picker regression test that covers the catalog-bearing round-trip (save + refine). * fix(cli): allow debug telemetry under opt-out env
2026-05-26 08:49:05 +02:00
const FIXTURE_ROOT = join(__dirname, '../../fixtures/lookml/single-model');
2026-05-10 23:12:26 +02:00
async function loadRepoFetch() {
test: split cli tests from source tree (#216) * feat(cli): define full warehouse dialect contract * test(cli): keep dialect edge tests focused * fix(cli): stabilize dialect contract foundation * refactor(connectors): own read-only query preparation * refactor(connectors): resolve dialects through registry * refactor(connectors): keep concrete dialect classes internal * chore(workspace): enforce dialect import boundary * refactor(cli): resolve relationship dialect at scan boundary * refactor(cli): use dialect display parsing for entity details * refactor(cli): use dialect display parsing for warehouse catalog * refactor(cli): use dialect SQL in relationship workflows * test(cli): verify solid dialect scan workflow closure * test: split cli tests from source tree * refactor(cli): standardize BigQuery scope listing * feat(sqlite): implement connector scope listing * test(connectors): cover required table listing * feat(cli): add warehouse driver registry * refactor(setup): route scope discovery through driver registry * refactor(cli): route local query execution through driver registry * refactor(historic-sql): route dialect support through driver registry * refactor(cli): test warehouse connections through driver registry * fix(cli): close driver registry type export gaps * Improve setup daemon diagnostics * refactor(setup): centralize rail-prefixed diagnostics + query-history fallback Extract errorMessage, writePrefixedLines, and flushPrefixedBufferedCommandOutput into clack.ts so the setup wizard, managed daemons, and embedding/agent steps share one rail-formatted writer. setup-databases.ts also adds a "disable query history and retry" option when the schema-context build fails and query history is the likely culprit, surfaced via a new failed-query-history-unavailable status. * fix(cli): carry catalog through the picker so BigQuery/Snowflake/SQL Server scope filters match The setup picker's KtxTableListEntry was a 2-level { schema, name }, so qualifiedTableId always wrote db.name into enabled_tables. When BigQuery, Snowflake, or SQL Server later ran fast ingest, their introspect step filtered the scope set with scopedTableNames(scope, { catalog: projectId|database, db }) — catalog was non-null on the introspect side but null in the scope refs, so every entry was rejected, the live-database adapter staged zero table files, and detect() failed with 'Adapter "live-database" did not recognize fetched source output'. Align the picker boundary with the canonical 3-level KtxTableRef: - Add catalog: string | null to KtxTableListEntry. - BigQuery/Snowflake/SQL Server listTables populate catalog from the resolved projectId / database; Postgres/MySQL/ClickHouse/SQLite set null. - qualifiedTableId emits catalog.schema.name when catalog is non-null (resolveEnabledTables already accepts the 3-part shape) and schemasFromEnabledTables now goes through parseDottedTableEntry so it recovers the schema correctly from both 2-part and 3-part entries. - Export parseDottedTableEntry from enabled-tables.ts (@internal) for picker reuse. Update listTables expectations in all seven connector tests and the setup / picker test fixtures. Add a picker regression test that covers the catalog-bearing round-trip (save + refine). * fix(cli): allow debug telemetry under opt-out env
2026-05-26 08:49:05 +02:00
return await import('../../../src/context/ingest/repo-fetch.js');
2026-05-10 23:12:26 +02:00
}
describe('repo-fetch', () => {
let tmpRoot: string;
beforeEach(async () => {
tmpRoot = await mkdtemp(join(tmpdir(), 'repo-fetch-'));
vi.resetModules();
test: split cli tests from source tree (#216) * feat(cli): define full warehouse dialect contract * test(cli): keep dialect edge tests focused * fix(cli): stabilize dialect contract foundation * refactor(connectors): own read-only query preparation * refactor(connectors): resolve dialects through registry * refactor(connectors): keep concrete dialect classes internal * chore(workspace): enforce dialect import boundary * refactor(cli): resolve relationship dialect at scan boundary * refactor(cli): use dialect display parsing for entity details * refactor(cli): use dialect display parsing for warehouse catalog * refactor(cli): use dialect SQL in relationship workflows * test(cli): verify solid dialect scan workflow closure * test: split cli tests from source tree * refactor(cli): standardize BigQuery scope listing * feat(sqlite): implement connector scope listing * test(connectors): cover required table listing * feat(cli): add warehouse driver registry * refactor(setup): route scope discovery through driver registry * refactor(cli): route local query execution through driver registry * refactor(historic-sql): route dialect support through driver registry * refactor(cli): test warehouse connections through driver registry * fix(cli): close driver registry type export gaps * Improve setup daemon diagnostics * refactor(setup): centralize rail-prefixed diagnostics + query-history fallback Extract errorMessage, writePrefixedLines, and flushPrefixedBufferedCommandOutput into clack.ts so the setup wizard, managed daemons, and embedding/agent steps share one rail-formatted writer. setup-databases.ts also adds a "disable query history and retry" option when the schema-context build fails and query history is the likely culprit, surfaced via a new failed-query-history-unavailable status. * fix(cli): carry catalog through the picker so BigQuery/Snowflake/SQL Server scope filters match The setup picker's KtxTableListEntry was a 2-level { schema, name }, so qualifiedTableId always wrote db.name into enabled_tables. When BigQuery, Snowflake, or SQL Server later ran fast ingest, their introspect step filtered the scope set with scopedTableNames(scope, { catalog: projectId|database, db }) — catalog was non-null on the introspect side but null in the scope refs, so every entry was rejected, the live-database adapter staged zero table files, and detect() failed with 'Adapter "live-database" did not recognize fetched source output'. Align the picker boundary with the canonical 3-level KtxTableRef: - Add catalog: string | null to KtxTableListEntry. - BigQuery/Snowflake/SQL Server listTables populate catalog from the resolved projectId / database; Postgres/MySQL/ClickHouse/SQLite set null. - qualifiedTableId emits catalog.schema.name when catalog is non-null (resolveEnabledTables already accepts the 3-part shape) and schemasFromEnabledTables now goes through parseDottedTableEntry so it recovers the schema correctly from both 2-part and 3-part entries. - Export parseDottedTableEntry from enabled-tables.ts (@internal) for picker reuse. Update listTables expectations in all seven connector tests and the setup / picker test fixtures. Add a picker regression test that covers the catalog-bearing round-trip (save + refine). * fix(cli): allow debug telemetry under opt-out env
2026-05-26 08:49:05 +02:00
vi.doUnmock('../../../src/context/ingest/git-env.js');
2026-05-10 23:12:26 +02:00
});
afterEach(async () => {
vi.restoreAllMocks();
vi.resetModules();
test: split cli tests from source tree (#216) * feat(cli): define full warehouse dialect contract * test(cli): keep dialect edge tests focused * fix(cli): stabilize dialect contract foundation * refactor(connectors): own read-only query preparation * refactor(connectors): resolve dialects through registry * refactor(connectors): keep concrete dialect classes internal * chore(workspace): enforce dialect import boundary * refactor(cli): resolve relationship dialect at scan boundary * refactor(cli): use dialect display parsing for entity details * refactor(cli): use dialect display parsing for warehouse catalog * refactor(cli): use dialect SQL in relationship workflows * test(cli): verify solid dialect scan workflow closure * test: split cli tests from source tree * refactor(cli): standardize BigQuery scope listing * feat(sqlite): implement connector scope listing * test(connectors): cover required table listing * feat(cli): add warehouse driver registry * refactor(setup): route scope discovery through driver registry * refactor(cli): route local query execution through driver registry * refactor(historic-sql): route dialect support through driver registry * refactor(cli): test warehouse connections through driver registry * fix(cli): close driver registry type export gaps * Improve setup daemon diagnostics * refactor(setup): centralize rail-prefixed diagnostics + query-history fallback Extract errorMessage, writePrefixedLines, and flushPrefixedBufferedCommandOutput into clack.ts so the setup wizard, managed daemons, and embedding/agent steps share one rail-formatted writer. setup-databases.ts also adds a "disable query history and retry" option when the schema-context build fails and query history is the likely culprit, surfaced via a new failed-query-history-unavailable status. * fix(cli): carry catalog through the picker so BigQuery/Snowflake/SQL Server scope filters match The setup picker's KtxTableListEntry was a 2-level { schema, name }, so qualifiedTableId always wrote db.name into enabled_tables. When BigQuery, Snowflake, or SQL Server later ran fast ingest, their introspect step filtered the scope set with scopedTableNames(scope, { catalog: projectId|database, db }) — catalog was non-null on the introspect side but null in the scope refs, so every entry was rejected, the live-database adapter staged zero table files, and detect() failed with 'Adapter "live-database" did not recognize fetched source output'. Align the picker boundary with the canonical 3-level KtxTableRef: - Add catalog: string | null to KtxTableListEntry. - BigQuery/Snowflake/SQL Server listTables populate catalog from the resolved projectId / database; Postgres/MySQL/ClickHouse/SQLite set null. - qualifiedTableId emits catalog.schema.name when catalog is non-null (resolveEnabledTables already accepts the 3-part shape) and schemasFromEnabledTables now goes through parseDottedTableEntry so it recovers the schema correctly from both 2-part and 3-part entries. - Export parseDottedTableEntry from enabled-tables.ts (@internal) for picker reuse. Update listTables expectations in all seven connector tests and the setup / picker test fixtures. Add a picker regression test that covers the catalog-bearing round-trip (save + refine). * fix(cli): allow debug telemetry under opt-out env
2026-05-26 08:49:05 +02:00
vi.doUnmock('../../../src/context/ingest/git-env.js');
2026-05-10 23:12:26 +02:00
await rm(tmpRoot, { recursive: true, force: true });
});
it('builds authenticated URLs for GitHub, GitLab, generic hosts, empty tokens, and file URLs', async () => {
const { buildAuthenticatedUrl } = await loadRepoFetch();
expect(buildAuthenticatedUrl('https://github.com/acme/repo.git', null)).toBe(
'https://github.com/acme/repo.git',
);
expect(buildAuthenticatedUrl('file:///tmp/repo', 'tok')).toBe('file:///tmp/repo');
expect(buildAuthenticatedUrl('https://github.com/acme/repo.git', 'tok')).toBe(
'https://x-token-auth:tok@github.com/acme/repo.git', // pragma: allowlist secret
);
expect(buildAuthenticatedUrl('https://gitlab.com/acme/repo.git', 'tok')).toBe(
'https://oauth2:tok@gitlab.com/acme/repo.git', // pragma: allowlist secret
);
expect(buildAuthenticatedUrl('https://git.example.com/acme/repo.git', 'tok')).toBe(
'https://token:tok@git.example.com/acme/repo.git', // pragma: allowlist secret
);
});
it('sanitizes both URL password patterns and literal token text', async () => {
const { sanitizeRepoError } = await loadRepoFetch();
const message = sanitizeRepoError(
new Error('failed https://token:supersecret@git.example.com/acme/repo.git because supersecret expired'), // pragma: allowlist secret
'supersecret',
);
expect(message).toContain('https://token:***@git.example.com/acme/repo.git');
expect(message).not.toContain('supersecret');
});
it('validates required and URL-shaped repository configuration', async () => {
const { RepoConfigError, validateRepoConfig } = await loadRepoFetch();
expect(() => validateRepoConfig({ repoUrl: '' })).toThrow(RepoConfigError);
expect(() => validateRepoConfig({ repoUrl: 'not a url' })).toThrow(RepoConfigError);
expect(() => validateRepoConfig({ repoUrl: 'file:///tmp/repo' })).not.toThrow();
});
it('clones a local repo and returns the full 40-character commit hash', async () => {
const { cloneOrPull } = await loadRepoFetch();
const repo = await makeLocalGitRepo(FIXTURE_ROOT, join(tmpRoot, 'origin'));
const cacheDir = join(tmpRoot, 'cache', 'conn-1');
const result = await cloneOrPull({
repoUrl: repo.repoUrl,
cacheDir,
branch: 'main',
});
expect(result.commitHash).toMatch(/^[0-9a-f]{40}$/);
await expect(readFile(join(cacheDir, 'orders.model.lkml'), 'utf-8')).resolves.toMatch(/connection:/);
});
it('pulls an existing cache and returns the new origin commit hash', async () => {
const { cloneOrPull } = await loadRepoFetch();
const repo = await makeLocalGitRepo(FIXTURE_ROOT, join(tmpRoot, 'origin'));
const cacheDir = join(tmpRoot, 'cache', 'conn-pull');
const first = await cloneOrPull({ repoUrl: repo.repoUrl, cacheDir, branch: 'main' });
await repo.writeFile('views/orders.view.lkml', 'view: orders { sql_table_name: public.orders_v2 ;; }\n');
const secondCommit = await repo.commit('bump lookml view');
const second = await cloneOrPull({ repoUrl: repo.repoUrl, cacheDir, branch: 'main' });
expect(second.commitHash).toBe(secondCommit);
expect(second.commitHash).not.toBe(first.commitHash);
await expect(readFile(join(cacheDir, 'views', 'orders.view.lkml'), 'utf-8')).resolves.toMatch(/orders_v2/);
});
it('falls back to a fresh clone when the existing cache diverges locally', async () => {
const { cloneOrPull } = await loadRepoFetch();
test: split cli tests from source tree (#216) * feat(cli): define full warehouse dialect contract * test(cli): keep dialect edge tests focused * fix(cli): stabilize dialect contract foundation * refactor(connectors): own read-only query preparation * refactor(connectors): resolve dialects through registry * refactor(connectors): keep concrete dialect classes internal * chore(workspace): enforce dialect import boundary * refactor(cli): resolve relationship dialect at scan boundary * refactor(cli): use dialect display parsing for entity details * refactor(cli): use dialect display parsing for warehouse catalog * refactor(cli): use dialect SQL in relationship workflows * test(cli): verify solid dialect scan workflow closure * test: split cli tests from source tree * refactor(cli): standardize BigQuery scope listing * feat(sqlite): implement connector scope listing * test(connectors): cover required table listing * feat(cli): add warehouse driver registry * refactor(setup): route scope discovery through driver registry * refactor(cli): route local query execution through driver registry * refactor(historic-sql): route dialect support through driver registry * refactor(cli): test warehouse connections through driver registry * fix(cli): close driver registry type export gaps * Improve setup daemon diagnostics * refactor(setup): centralize rail-prefixed diagnostics + query-history fallback Extract errorMessage, writePrefixedLines, and flushPrefixedBufferedCommandOutput into clack.ts so the setup wizard, managed daemons, and embedding/agent steps share one rail-formatted writer. setup-databases.ts also adds a "disable query history and retry" option when the schema-context build fails and query history is the likely culprit, surfaced via a new failed-query-history-unavailable status. * fix(cli): carry catalog through the picker so BigQuery/Snowflake/SQL Server scope filters match The setup picker's KtxTableListEntry was a 2-level { schema, name }, so qualifiedTableId always wrote db.name into enabled_tables. When BigQuery, Snowflake, or SQL Server later ran fast ingest, their introspect step filtered the scope set with scopedTableNames(scope, { catalog: projectId|database, db }) — catalog was non-null on the introspect side but null in the scope refs, so every entry was rejected, the live-database adapter staged zero table files, and detect() failed with 'Adapter "live-database" did not recognize fetched source output'. Align the picker boundary with the canonical 3-level KtxTableRef: - Add catalog: string | null to KtxTableListEntry. - BigQuery/Snowflake/SQL Server listTables populate catalog from the resolved projectId / database; Postgres/MySQL/ClickHouse/SQLite set null. - qualifiedTableId emits catalog.schema.name when catalog is non-null (resolveEnabledTables already accepts the 3-part shape) and schemasFromEnabledTables now goes through parseDottedTableEntry so it recovers the schema correctly from both 2-part and 3-part entries. - Export parseDottedTableEntry from enabled-tables.ts (@internal) for picker reuse. Update listTables expectations in all seven connector tests and the setup / picker test fixtures. Add a picker regression test that covers the catalog-bearing round-trip (save + refine). * fix(cli): allow debug telemetry under opt-out env
2026-05-26 08:49:05 +02:00
const { createSimpleGit } = await import('../../../src/context/ingest/git-env.js');
2026-05-10 23:12:26 +02:00
const repo = await makeLocalGitRepo(FIXTURE_ROOT, join(tmpRoot, 'origin'));
const cacheDir = join(tmpRoot, 'cache', 'conn-diverged');
await cloneOrPull({ repoUrl: repo.repoUrl, cacheDir, branch: 'main' });
const cacheGit = createSimpleGit(cacheDir);
2026-05-10 23:51:24 +02:00
await cacheGit.addConfig('user.email', 'test@ktx.local');
await cacheGit.addConfig('user.name', 'KTX Test');
2026-05-10 23:12:26 +02:00
await writeFile(join(cacheDir, 'local-only.txt'), 'local commit\n', 'utf-8');
await cacheGit.add('.');
await cacheGit.commit('local-only divergent commit');
await repo.writeFile('views/orders.view.lkml', 'view: orders { sql_table_name: public.orders_remote ;; }\n');
const originCommit = await repo.commit('remote commit');
const result = await cloneOrPull({ repoUrl: repo.repoUrl, cacheDir, branch: 'main' });
expect(result.commitHash).toBe(originCommit);
await expect(access(join(cacheDir, 'local-only.txt'))).rejects.toThrow();
await expect(readFile(join(cacheDir, 'views', 'orders.view.lkml'), 'utf-8')).resolves.toMatch(/orders_remote/);
});
it('falls back to a fresh clone when the cache has a corrupt .git directory', async () => {
const { cloneOrPull } = await loadRepoFetch();
const repo = await makeLocalGitRepo(FIXTURE_ROOT, join(tmpRoot, 'origin'));
const cacheDir = join(tmpRoot, 'cache', 'conn-corrupt');
await mkdir(join(cacheDir, '.git'), { recursive: true });
await writeFile(join(cacheDir, '.git', 'HEAD'), 'garbage\n', 'utf-8');
const result = await cloneOrPull({ repoUrl: repo.repoUrl, cacheDir, branch: 'main' });
expect(result.commitHash).toMatch(/^[0-9a-f]{40}$/);
await expect(readFile(join(cacheDir, 'orders.model.lkml'), 'utf-8')).resolves.toMatch(/connection:/);
});
it('returns a sanitized RepoFetchError when fresh clone fails', async () => {
const { RepoFetchError, cloneOrPull } = await loadRepoFetch();
const repo = await makeLocalGitRepo(FIXTURE_ROOT, join(tmpRoot, 'origin'));
await expect(
cloneOrPull({
repoUrl: repo.repoUrl,
cacheDir: join(tmpRoot, 'cache', 'missing-branch'),
branch: 'missing',
authToken: 'supersecret-token',
}),
).rejects.toThrow(RepoFetchError);
await expect(
cloneOrPull({
repoUrl: repo.repoUrl,
cacheDir: join(tmpRoot, 'cache', 'missing-branch-2'),
branch: 'missing',
authToken: 'supersecret-token',
}),
).rejects.toThrow(expect.objectContaining({ message: expect.not.stringContaining('supersecret-token') }));
});
it('testRepoConnection returns ok true for a local repo and ok false for a missing local repo', async () => {
const { testRepoConnection } = await loadRepoFetch();
const repo = await makeLocalGitRepo(FIXTURE_ROOT, join(tmpRoot, 'origin'));
await expect(testRepoConnection({ repoUrl: repo.repoUrl })).resolves.toEqual({ ok: true });
const failed = await testRepoConnection({ repoUrl: `file://${join(tmpRoot, 'does-not-exist')}` });
expect(failed.ok).toBe(false);
if (!failed.ok) {
expect(failed.error).toEqual(expect.any(String));
}
});
it('cleans up non-existent and existing repository directories idempotently', async () => {
const { cleanupRepoDir } = await loadRepoFetch();
const existing = join(tmpRoot, 'cache', 'to-clean');
await mkdir(join(existing, '.git'), { recursive: true });
await cleanupRepoDir(existing);
await cleanupRepoDir(existing);
await expect(access(existing)).rejects.toThrow();
});
it('sets the remote URL on every pull so token rotation and token removal update cached .git/config', async () => {
const cacheDir = join(tmpRoot, 'cache', 'auth-refresh');
await mkdir(join(cacheDir, '.git'), { recursive: true });
const worktreeGit = {
remote: vi.fn(async () => undefined),
fetch: vi.fn(async () => undefined),
checkout: vi.fn(async () => undefined),
pull: vi.fn(async () => undefined),
log: vi.fn(async () => ({ latest: { hash: 'a'.repeat(40) } })),
};
const rootGit = {
clone: vi.fn(async () => undefined),
};
test: split cli tests from source tree (#216) * feat(cli): define full warehouse dialect contract * test(cli): keep dialect edge tests focused * fix(cli): stabilize dialect contract foundation * refactor(connectors): own read-only query preparation * refactor(connectors): resolve dialects through registry * refactor(connectors): keep concrete dialect classes internal * chore(workspace): enforce dialect import boundary * refactor(cli): resolve relationship dialect at scan boundary * refactor(cli): use dialect display parsing for entity details * refactor(cli): use dialect display parsing for warehouse catalog * refactor(cli): use dialect SQL in relationship workflows * test(cli): verify solid dialect scan workflow closure * test: split cli tests from source tree * refactor(cli): standardize BigQuery scope listing * feat(sqlite): implement connector scope listing * test(connectors): cover required table listing * feat(cli): add warehouse driver registry * refactor(setup): route scope discovery through driver registry * refactor(cli): route local query execution through driver registry * refactor(historic-sql): route dialect support through driver registry * refactor(cli): test warehouse connections through driver registry * fix(cli): close driver registry type export gaps * Improve setup daemon diagnostics * refactor(setup): centralize rail-prefixed diagnostics + query-history fallback Extract errorMessage, writePrefixedLines, and flushPrefixedBufferedCommandOutput into clack.ts so the setup wizard, managed daemons, and embedding/agent steps share one rail-formatted writer. setup-databases.ts also adds a "disable query history and retry" option when the schema-context build fails and query history is the likely culprit, surfaced via a new failed-query-history-unavailable status. * fix(cli): carry catalog through the picker so BigQuery/Snowflake/SQL Server scope filters match The setup picker's KtxTableListEntry was a 2-level { schema, name }, so qualifiedTableId always wrote db.name into enabled_tables. When BigQuery, Snowflake, or SQL Server later ran fast ingest, their introspect step filtered the scope set with scopedTableNames(scope, { catalog: projectId|database, db }) — catalog was non-null on the introspect side but null in the scope refs, so every entry was rejected, the live-database adapter staged zero table files, and detect() failed with 'Adapter "live-database" did not recognize fetched source output'. Align the picker boundary with the canonical 3-level KtxTableRef: - Add catalog: string | null to KtxTableListEntry. - BigQuery/Snowflake/SQL Server listTables populate catalog from the resolved projectId / database; Postgres/MySQL/ClickHouse/SQLite set null. - qualifiedTableId emits catalog.schema.name when catalog is non-null (resolveEnabledTables already accepts the 3-part shape) and schemasFromEnabledTables now goes through parseDottedTableEntry so it recovers the schema correctly from both 2-part and 3-part entries. - Export parseDottedTableEntry from enabled-tables.ts (@internal) for picker reuse. Update listTables expectations in all seven connector tests and the setup / picker test fixtures. Add a picker regression test that covers the catalog-bearing round-trip (save + refine). * fix(cli): allow debug telemetry under opt-out env
2026-05-26 08:49:05 +02:00
vi.doMock('../../../src/context/ingest/git-env.js', () => ({
2026-05-10 23:12:26 +02:00
createSimpleGit: vi.fn((baseDir?: string) => (baseDir ? worktreeGit : rootGit)),
}));
const { cloneOrPull } = await loadRepoFetch();
await cloneOrPull({
repoUrl: 'https://github.com/acme/repo.git',
authToken: 'new-token',
cacheDir,
branch: 'main',
});
await cloneOrPull({
repoUrl: 'https://github.com/acme/repo.git',
authToken: null,
cacheDir,
branch: 'main',
});
expect(worktreeGit.remote).toHaveBeenCalledWith([
'set-url',
'origin',
'https://x-token-auth:new-token@github.com/acme/repo.git', // pragma: allowlist secret
]);
expect(worktreeGit.remote).toHaveBeenCalledWith(['set-url', 'origin', 'https://github.com/acme/repo.git']);
expect(rootGit.clone).not.toHaveBeenCalled();
});
});