a9c51021aa
线上实测:reviewer 用 fable-5 报「There's an issue with the selected model (claude-fable-5)」(该档未对当前账号开通),但 isModelError 未命中此措辞 → 不回退 → 复审失败(verdict=null)。补正则后 fable-5 不可用会自动回退 opus 重试,复审正常。 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
113 lines
5.3 KiB
TypeScript
113 lines
5.3 KiB
TypeScript
import { test } from 'node:test';
|
||
import assert from 'node:assert/strict';
|
||
import { pickModel, isModelError, pickFallbackModel, MODEL_FALLBACK_CHAIN } from '../src/executor/models.js';
|
||
import type { Project, Task } from '../src/model/types.js';
|
||
import type { Complexity } from '../src/model/complexity.js';
|
||
|
||
function fakeTask(complexity: Complexity): Task {
|
||
return { complexity } as Task;
|
||
}
|
||
function fakeProject(model: string | null = null): Project {
|
||
return { model } as Project;
|
||
}
|
||
|
||
/** 临时设置 env,跑完恢复(避免污染其他用例) */
|
||
function withEnv(vars: Record<string, string>, fn: () => void): void {
|
||
const saved = new Map<string, string | undefined>();
|
||
for (const [k, v] of Object.entries(vars)) {
|
||
saved.set(k, process.env[k]);
|
||
process.env[k] = v;
|
||
}
|
||
try {
|
||
fn();
|
||
} finally {
|
||
for (const [k, v] of saved) {
|
||
if (v === undefined) delete process.env[k];
|
||
else process.env[k] = v;
|
||
}
|
||
}
|
||
}
|
||
|
||
const ENV_KEYS = [
|
||
'MAESTRO_MODEL_EASY', 'MAESTRO_MODEL_MEDIUM', 'MAESTRO_MODEL_HARD',
|
||
'MAESTRO_MODEL_REVIEW_EASY', 'MAESTRO_MODEL_REVIEW_MEDIUM', 'MAESTRO_MODEL_REVIEW_HARD',
|
||
];
|
||
|
||
test('pickModel:复杂度映射默认值(executor / reviewer)', () => {
|
||
// 防止外部 env 干扰:先清空再断言默认
|
||
withEnv(Object.fromEntries(ENV_KEYS.map((k) => [k, ''])), () => {
|
||
const p = fakeProject();
|
||
assert.equal(pickModel(fakeTask('easy'), p, 'executor'), 'claude-sonnet-4-6');
|
||
assert.equal(pickModel(fakeTask('medium'), p, 'executor'), 'claude-opus-4-8');
|
||
assert.equal(pickModel(fakeTask('hard'), p, 'executor'), 'claude-opus-4-8');
|
||
// reviewer 三档统一 fable-5(复审是质量关口,用最强档保严格度)
|
||
assert.equal(pickModel(fakeTask('easy'), p, 'reviewer'), 'claude-fable-5');
|
||
assert.equal(pickModel(fakeTask('medium'), p, 'reviewer'), 'claude-fable-5');
|
||
assert.equal(pickModel(fakeTask('hard'), p, 'reviewer'), 'claude-fable-5');
|
||
// planner hard 拆解最烧脑 → fable-5;medium=opus;easy=sonnet
|
||
assert.equal(pickModel(fakeTask('hard'), p, 'planner'), 'claude-fable-5');
|
||
assert.equal(pickModel(fakeTask('medium'), p, 'planner'), 'claude-opus-4-8');
|
||
assert.equal(pickModel(fakeTask('easy'), p, 'planner'), 'claude-sonnet-4-6');
|
||
});
|
||
});
|
||
|
||
test('pickModel:project.model executor 最优先,但 reviewer/conflict 不受 project.model 覆盖', () => {
|
||
withEnv(Object.fromEntries(ENV_KEYS.map((k) => [k, ''])), () => {
|
||
const p = fakeProject('claude-opus-4-6');
|
||
for (const c of ['easy', 'medium', 'hard'] as const) {
|
||
// executor 受 project.model 覆盖
|
||
assert.equal(pickModel(fakeTask(c), p, 'executor'), 'claude-opus-4-6');
|
||
// reviewer 不受 project.model 覆盖,始终用 fable-5(保自审独立性)
|
||
assert.equal(pickModel(fakeTask(c), p, 'reviewer'), 'claude-fable-5');
|
||
// conflict 固定 fable-5(解冲突恒用最强档)
|
||
assert.equal(pickModel(fakeTask(c), p, 'conflict'), 'claude-fable-5');
|
||
}
|
||
});
|
||
});
|
||
|
||
test('pickModel:env 覆盖默认值,executor/reviewer 各用各的 env', () => {
|
||
withEnv({
|
||
MAESTRO_MODEL_EASY: 'claude-haiku-4-5',
|
||
MAESTRO_MODEL_REVIEW_EASY: 'claude-sonnet-4-6', // reviewer env 可单独覆盖
|
||
}, () => {
|
||
const p = fakeProject();
|
||
assert.equal(pickModel(fakeTask('easy'), p, 'executor'), 'claude-haiku-4-5');
|
||
// reviewer easy 被 MAESTRO_MODEL_REVIEW_EASY 覆盖
|
||
assert.equal(pickModel(fakeTask('easy'), p, 'reviewer'), 'claude-sonnet-4-6');
|
||
// reviewer hard 无 env → 默认 fable-5
|
||
assert.equal(pickModel(fakeTask('hard'), p, 'reviewer'), 'claude-fable-5');
|
||
assert.equal(pickModel(fakeTask('hard'), p, 'executor'), 'claude-opus-4-8');
|
||
});
|
||
});
|
||
|
||
test('pickModel:env 为空白字符串视为未设置', () => {
|
||
withEnv({ MAESTRO_MODEL_MEDIUM: ' ' }, () => {
|
||
assert.equal(pickModel(fakeTask('medium'), fakeProject(), 'executor'), 'claude-opus-4-8');
|
||
});
|
||
});
|
||
|
||
test('isModelError:模型不可用类错误才触发回退', () => {
|
||
// 应触发
|
||
assert.ok(isModelError('model claude-fable-5 not_found'));
|
||
assert.ok(isModelError('Invalid model: claude-x'));
|
||
assert.ok(isModelError('permission denied for model claude-fable-5'));
|
||
assert.ok(isModelError('404 model does not exist'));
|
||
assert.ok(isModelError('The requested model is unavailable'));
|
||
assert.ok(isModelError('unknown model "claude-fable-5"'));
|
||
// Claude Code 的「未开通某档模型」报错(线上 fable-5 实测)
|
||
assert.ok(isModelError("There's an issue with the selected model (claude-fable-5)"));
|
||
// 不应触发
|
||
assert.ok(!isModelError('执行超时'));
|
||
assert.ok(!isModelError('network connection refused'));
|
||
assert.ok(!isModelError('not found: /tmp/some/file')); // 没提到 model
|
||
assert.ok(!isModelError('model produced empty output')); // 提到 model 但不是可用性错误
|
||
});
|
||
|
||
test('pickFallbackModel:回退链取下一档(循环)', () => {
|
||
// 链:fable → opus → sonnet → fable(循环,fable 置链首=优先最强档)
|
||
assert.deepEqual([...MODEL_FALLBACK_CHAIN], ['claude-fable-5', 'claude-opus-4-8', 'claude-sonnet-4-6']);
|
||
assert.equal(pickFallbackModel('claude-fable-5'), 'claude-opus-4-8');
|
||
assert.equal(pickFallbackModel('claude-opus-4-8'), 'claude-sonnet-4-6');
|
||
assert.equal(pickFallbackModel('claude-sonnet-4-6'), 'claude-fable-5');
|
||
});
|