feat: add spreadsheet async imports

This commit is contained in:
Codex
2026-06-29 08:00:17 +08:00
parent 78a26d1df2
commit eee21f6eed
27 changed files with 2972 additions and 267 deletions

View File

@@ -0,0 +1,299 @@
import assert from 'node:assert/strict';
import pg from 'pg';
import { spawn } from 'node:child_process';
const databaseUrl = process.env.DATABASE_URL || 'postgresql://postgres:postgres@127.0.0.1:54322/postgres';
const tenantId = '00000000-0000-0000-0000-000000000001';
const adminUserId = '00000000-0000-0000-0000-000000000102';
const ids = {
region: '00000000-0000-0000-0000-000000000301',
subject: '00000000-0000-0000-0000-000000000501',
category: '00000000-0000-0000-0000-000000000601',
contentEntry: '00000000-0000-0000-0000-000000000611',
contentNodeSchoolTarget: '00000000-0000-0000-0000-000000000614',
questionCollection: '00000000-0000-0000-0000-000000000615',
};
function runWorkerOnce() {
const child = spawn(process.execPath, ['apps/worker/dist/apps/worker/src/index.js', '--once', '--job', 'imports'], {
cwd: process.cwd(),
env: {
...process.env,
DATABASE_URL: databaseUrl,
WORKER_IMPORT_BATCH_SIZE: '5',
WORKER_IMPORT_ID: 'imports-integration-test',
},
stdio: ['ignore', 'pipe', 'pipe'],
windowsHide: true,
});
let output = '';
child.stdout.on('data', chunk => {
output += chunk.toString();
});
child.stderr.on('data', chunk => {
output += chunk.toString();
});
return new Promise((resolve, reject) => {
child.on('error', reject);
child.on('exit', code => {
try {
assert.equal(code, 0, `worker should exit 0\n${output}`);
assert.match(output, /imports batch processed=\d+/, 'worker output should include imports summary');
resolve(output);
} catch (error) {
reject(error);
}
});
});
}
async function cleanup(pool) {
await pool.query(
`
delete from public.question_collection_items
where tenant_id = $1
and question_id in (
select id from public.questions
where tenant_id = $1
and (
legacy_id like 'worker-import-question-%'
or legacy_id like 'integration-import-async-choice-%'
)
)
`,
[tenantId],
);
await pool.query(
`
delete from public.question_versions
where tenant_id = $1
and question_id in (
select id from public.questions
where tenant_id = $1
and (
legacy_id like 'worker-import-question-%'
or legacy_id like 'integration-import-async-choice-%'
)
)
`,
[tenantId],
);
await pool.query(
`
delete from public.questions
where tenant_id = $1
and (
legacy_id like 'worker-import-question-%'
or legacy_id like 'integration-import-async-choice-%'
)
`,
[tenantId],
);
await pool.query(
`
delete from public.audit_logs
where tenant_id = $1
and target_type = 'content_import_job'
and details::text like '%worker-import%'
`,
[tenantId],
);
await pool.query(
`
delete from public.content_import_jobs
where tenant_id = $1
and (
source_name like 'worker-import-%'
or source_name = 'async-question-import.json'
)
`,
[tenantId],
);
}
async function createQueuedQuestionImport(pool) {
const preview = await pool.query(
`
insert into public.content_import_jobs (
tenant_id, created_by, import_type, source_format, status,
source_name, source_hash, target_region_id, target_subject_id,
target_category_id, target_content_node_id, target_collection_id,
dry_run, total_count, valid_count, error_count, warning_count,
summary, raw_payload, normalized_payload, execution_mode, queued_at,
next_attempt_at, parser_metadata
)
values (
$1, $2, 'questions', 'json', 'pending',
'worker-import-questions.json', 'worker-import-source-hash',
$3::uuid, $4::uuid, $5::uuid, $6::uuid, $7::uuid,
false, 1, 1, 0, 0,
$8::jsonb, $9::jsonb, $10::jsonb, 'async', now(), now(), '{}'::jsonb
)
returning id
`,
[
tenantId,
adminUserId,
ids.region,
ids.subject,
ids.category,
ids.contentNodeSchoolTarget,
ids.questionCollection,
JSON.stringify({
target: {
regionId: ids.region,
subjectId: ids.subject,
categoryId: ids.category,
contentNodeId: ids.contentNodeSchoolTarget,
collectionId: ids.questionCollection,
},
importOptions: { allowPartial: false },
source: 'worker-import-integration',
}),
JSON.stringify([
{
legacyId: 'worker-import-question-001',
type: 'choice',
content: '异步导入题worker 应该复用哪套导入规则?',
options: ['自己重写', '复用后端导入 executor', '前端直写数据库', '跳过校验'],
correctOptionIndices: [1],
explanation: 'worker 和 API 必须复用同一套后端导入规则。',
difficulty: 2,
tags: ['worker-import'],
},
]),
JSON.stringify([
{
legacyId: 'worker-import-question-001',
type: 'choice',
typeLabel: null,
content: '异步导入题worker 应该复用哪套导入规则?',
options: ['自己重写', '复用后端导入 executor', '前端直写数据库', '跳过校验'],
correctOptionIndex: 1,
correctOptionIndices: [1],
answerText: null,
explanation: 'worker 和 API 必须复用同一套后端导入规则。',
difficulty: 2,
tags: ['worker-import'],
mediaUrl: null,
subQuestions: [],
codeLang: null,
codeTemplate: null,
examMarkers: {},
sourceHash: 'worker-import-question-hash-001',
},
]),
],
);
const jobId = preview.rows[0].id;
await pool.query(
`
insert into public.content_import_items (
tenant_id, job_id, row_no, external_id, status, target_type,
source_payload, normalized_payload, content_hash, issues_count
)
values ($1, $2, 1, 'worker-import-question-001', 'valid', 'question', $3::jsonb, $4::jsonb, 'worker-import-question-hash-001', 0)
`,
[
tenantId,
jobId,
JSON.stringify({
legacyId: 'worker-import-question-001',
content: '异步导入题worker 应该复用哪套导入规则?',
}),
JSON.stringify({
legacyId: 'worker-import-question-001',
type: 'choice',
typeLabel: null,
content: '异步导入题worker 应该复用哪套导入规则?',
options: ['自己重写', '复用后端导入 executor', '前端直写数据库', '跳过校验'],
correctOptionIndex: 1,
correctOptionIndices: [1],
answerText: null,
explanation: 'worker 和 API 必须复用同一套后端导入规则。',
difficulty: 2,
tags: ['worker-import'],
mediaUrl: null,
subQuestions: [],
codeLang: null,
codeTemplate: null,
examMarkers: {},
sourceHash: 'worker-import-question-hash-001',
}),
],
);
return jobId;
}
async function main() {
const pool = new pg.Pool({ connectionString: databaseUrl });
try {
await cleanup(pool);
const jobId = await createQueuedQuestionImport(pool);
const output = await runWorkerOnce();
assert.match(output, /completed=1/, 'worker should complete exactly the queued import job after cleanup');
const job = await pool.query(
`
select status, execution_mode, inserted_count, updated_count, skipped_count,
locked_at, locked_by, attempt_count, error_message
from public.content_import_jobs
where tenant_id = $1 and id = $2
`,
[tenantId, jobId],
);
assert.equal(job.rows[0]?.status, 'completed', 'queued import job should be completed');
assert.equal(job.rows[0]?.execution_mode, 'async', 'job should keep async execution mode');
assert.equal(Number(job.rows[0]?.inserted_count), 1, 'worker should insert one question');
assert.equal(job.rows[0]?.locked_at, null, 'completed job should release lock');
assert.equal(job.rows[0]?.locked_by, null, 'completed job should clear lock owner');
assert.equal(Number(job.rows[0]?.attempt_count), 1, 'worker should record one attempt');
assert.equal(job.rows[0]?.error_message, null, 'completed job should not retain error message');
const question = await pool.query(
`
select q.id, v.content
from public.questions q
join public.question_versions v on v.id = q.current_version_id
where q.tenant_id = $1 and q.legacy_id = 'worker-import-question-001'
limit 1
`,
[tenantId],
);
assert.equal(question.rows[0]?.content, '异步导入题worker 应该复用哪套导入规则?', 'worker should import question content');
const collectionItem = await pool.query(
`
select 1
from public.question_collection_items
where tenant_id = $1 and collection_id = $2 and question_id = $3
limit 1
`,
[tenantId, ids.questionCollection, question.rows[0]?.id],
);
assert.equal(collectionItem.rowCount, 1, 'worker should bind imported question to collection');
const audit = await pool.query(
`
select action
from public.audit_logs
where tenant_id = $1 and target_type = 'content_import_job' and target_id = $2
order by created_at desc
limit 1
`,
[tenantId, jobId],
);
assert.equal(audit.rows[0]?.action, 'content.import.questions.completed', 'worker import should write completion audit');
console.log('Import worker integration test complete.');
} finally {
await cleanup(pool).catch(() => {});
await pool.end();
}
}
main().catch(error => {
console.error(error);
process.exit(1);
});