sillyspec 3.20.2 → 3.20.4

package/test/knowledge-match.test.mjs

@@ -1,231 +1,231 @@
-/**
- * knowledge-match.test.mjs — knowledge 关键词匹配引擎测试
- */
-
-import { join } from 'path'
-import { existsSync, mkdirSync, writeFileSync, rmSync } from 'fs'
-import { fileURLToPath, pathToFileURL } from 'url'
-import { tmpdir } from 'os'
-
-const __filename = fileURLToPath(import.meta.url)
-const __dirname = join(__filename, '..') // join(__dirname, '..') to get parent
-const root = join(__dirname, '..')
-
-const { parseKnowledgeIndex, matchKnowledge } = await import(
-  pathToFileURL(join(root, 'src', 'knowledge-match.js')).href
-)
-
-let passed = 0, failed = 0
-
-function assert(cond, msg) {
-  if (cond) { console.log(`  ✅ PASS: ${msg}`); passed++ }
-  else { console.log(`  ❌ FAIL: ${msg}`); failed++ }
-}
-
-function setup(name) {
-  const dir = join(tmpdir(), `kh-test-${name}`)
-  mkdirSync(dir, { recursive: true })
-  return dir
-}
-function clean(...dirs) {
-  for (const d of dirs) try { rmSync(d, { recursive: true, force: true }) } catch {}
-}
-
-// ── 1: 有匹配条目 → matched=true, entries.length > 0 ──
-console.log('\n=== Test 1: 有匹配条目 → matched=true ===')
-{
-  const dir = setup('t1')
-  try {
-    writeFileSync(join(dir, 'INDEX.md'), [
-      '# Knowledge Index',
-      '',
-      '## Conventions',
-      '- ESM|module|import → [ESM Only](conventions.md#esm-only)',
-      '',
-      '## Patterns',
-      '- 阶段定义|stage|stages → [Stage Pattern](patterns.md#stage-step-pattern)',
-    ].join('\n'))
-
-    const result = matchKnowledge(dir, 'setup ESM module imports')
-    assert(result.matched === true, 'matched is true')
-    assert(result.entries.length === 1, 'one entry matched')
-    assert(result.entries[0].file === 'conventions.md', 'file is conventions.md')
-    assert(result.entries[0].anchor === 'esm-only', 'anchor is esm-only')
-    assert(result.entries[0].keywords.includes('ESM'), 'keywords include ESM')
-  } finally { clean(dir) }
-}
-
-// ── 2: 无匹配条目 → matched=false, report 包含 "no matches" ──
-console.log('\n=== Test 2: 无匹配条目 → matched=false ===')
-{
-  const dir = setup('t2')
-  try {
-    writeFileSync(join(dir, 'INDEX.md'), [
-      '# Knowledge Index',
-      '',
-      '## Patterns',
-      '- 阶段定义|stage → [Stage Pattern](patterns.md#stage-step-pattern)',
-    ].join('\n'))
-
-    const result = matchKnowledge(dir, 'implement authentication flow')
-    assert(result.matched === false, 'matched is false')
-    assert(result.entries.length === 0, 'no entries matched')
-    assert(result.report.includes('no matches'), 'report says no matches')
-  } finally { clean(dir) }
-}
-
-// ── 3: INDEX.md 不存在 → matched=false, report 包含 "not found" ──
-console.log('\n=== Test 3: INDEX.md 不存在 → matched=false ===')
-{
-  const dir = setup('t3')
-  try {
-    const result = matchKnowledge(dir, 'any context')
-    assert(result.matched === false, 'matched is false')
-    assert(result.report.includes('not found'), 'report says not found')
-    assert(result.json.matched === false, 'json.matched is false')
-    assert(result.json.entry_count === 0, 'json.entry_count is 0')
-  } finally { clean(dir) }
-}
-
-// ── 4: INDEX.md 有空锚点引用 → 正常处理（不过度校验）──
-console.log('\n=== Test 4: INDEX.md 有空锚点引用 → 正常处理 ===')
-{
-  const dir = setup('t4')
-  try {
-    writeFileSync(join(dir, 'INDEX.md'), [
-      '# Knowledge Index',
-      '',
-      '## Conventions',
-      '- naming|camelCase → [Naming](conventions.md)',
-    ].join('\n'))
-
-    const result = matchKnowledge(dir, 'naming conventions')
-    assert(result.matched === true, 'matched is true with empty anchor')
-    assert(result.entries[0].anchor === '', 'anchor is empty string')
-    assert(result.entries[0].file === 'conventions.md', 'file is correct')
-  } finally { clean(dir) }
-}
-
-// ── 5: 大小写不敏感匹配 → "STAGE" 匹配 "stage" ──
-console.log('\n=== Test 5: 大小写不敏感匹配 ===')
-{
-  const dir = setup('t5')
-  try {
-    writeFileSync(join(dir, 'INDEX.md'), [
-      '# Knowledge Index',
-      '',
-      '## Patterns',
-      '- stage|stages → [Stage Pattern](patterns.md#stage-step-pattern)',
-    ].join('\n'))
-
-    const result = matchKnowledge(dir, 'implement STAGE definition')
-    assert(result.matched === true, 'STAGE matches stage')
-    assert(result.entries.length === 1, 'one entry matched')
-  } finally { clean(dir) }
-}
-
-// ── 6: report 格式包含 "Knowledge Context" header ──
-console.log('\n=== Test 6: report 格式包含 Knowledge Context header ===')
-{
-  const dir = setup('t6')
-  try {
-    writeFileSync(join(dir, 'INDEX.md'), [
-      '# Knowledge Index',
-      '',
-      '## Patterns',
-      '- 阶段定义|stage → [Stage Pattern](patterns.md#stage-step-pattern)',
-    ].join('\n'))
-
-    const result = matchKnowledge(dir, 'stage configuration')
-    assert(result.report.includes('Knowledge Context'), 'report has Knowledge Context header')
-    assert(result.report.includes('Status: matched'), 'report has Status: matched')
-    assert(result.report.includes('Entries: 1'), 'report has Entries: 1')
-    assert(result.report.includes('patterns.md#stage-step-pattern'), 'report has source')
-  } finally { clean(dir) }
-}
-
-// ── 7: json 结构正确 ──
-console.log('\n=== Test 7: json 结构正确 ===')
-{
-  const dir = setup('t7')
-  try {
-    writeFileSync(join(dir, 'INDEX.md'), [
-      '# Knowledge Index',
-      '',
-      '## Conventions',
-      '- ESM|module → [ESM Only](conventions.md#esm-only)',
-      '',
-      '## Patterns',
-      '- stage|stages → [Stage Pattern](patterns.md#stage-step-pattern)',
-    ].join('\n'))
-
-    const result = matchKnowledge(dir, 'stage module setup')
-    const j = result.json
-    assert(j.matched === true, 'json.matched is true')
-    assert(j.entry_count === 2, 'json.entry_count is 2')
-    assert(Array.isArray(j.entries), 'json.entries is array')
-    assert(j.entries.length === 2, 'json.entries length is 2')
-
-    const stageEntry = j.entries.find(e => e.anchor === 'stage-step-pattern')
-    assert(!!stageEntry, 'stage entry found in json')
-    assert(stageEntry.keywords.includes('stage'), 'stage keywords correct')
-    assert(stageEntry.category === 'Patterns', 'stage category correct')
-    assert(stageEntry.file === 'patterns.md', 'stage file correct')
-  } finally { clean(dir) }
-}
-
-// ── 8: 空目录（目录存在但无 INDEX.md）→ no matches ──
-console.log('\n=== Test 8: 空目录（无 INDEX.md）→ no matches ===')
-{
-  const dir = setup('t8')
-  try {
-    const result = matchKnowledge(dir, 'any task')
-    assert(result.matched === false, 'no INDEX.md → matched false')
-    assert(result.report.includes('not found'), 'report says not found')
-  } finally { clean(dir) }
-}
-
-// ── 9: parseKnowledgeIndex 返回正确分类 ──
-console.log('\n=== Test 9: parseKnowledgeIndex 分类正确 ===')
-{
-  const dir = setup('t9')
-  try {
-    writeFileSync(join(dir, 'INDEX.md'), [
-      '# Knowledge Index',
-      '',
-      '## Known Issues',
-      '- GLM|proxy → [GLM Proxy](known-issues.md#glm-proxy)',
-      '',
-      '## Conventions',
-      '- 命名|naming → [命名规范](conventions.md#naming)',
-    ].join('\n'))
-
-    const entries = parseKnowledgeIndex(dir)
-    assert(entries.length === 2, 'parsed 2 entries')
-    assert(entries[0].category === 'Known Issues', 'first entry category correct')
-    assert(entries[1].category === 'Conventions', 'second entry category correct')
-    assert(entries[0].display === 'GLM Proxy', 'display text correct')
-  } finally { clean(dir) }
-}
-
-// ── 10: taskContext 为空 → no matches ──
-console.log('\n=== Test 10: taskContext 为空 → no matches ===')
-{
-  const dir = setup('t10')
-  try {
-    writeFileSync(join(dir, 'INDEX.md'), [
-      '# Knowledge Index',
-      '',
-      '## Patterns',
-      '- stage → [Stage](patterns.md#stage)',
-    ].join('\n'))
-
-    const result = matchKnowledge(dir, '')
-    assert(result.matched === false, 'empty context → matched false')
-  } finally { clean(dir) }
-}
-
-// ── 汇总 ──
-console.log(`\n${'='.repeat(40)}`)
-console.log(`knowledge-match tests: ${passed} passed, ${failed} failed, ${passed + failed} total`)
-if (failed > 0) process.exit(1)
+/**
+ * knowledge-match.test.mjs — knowledge 关键词匹配引擎测试
+ */
+
+import { join } from 'path'
+import { existsSync, mkdirSync, writeFileSync, rmSync } from 'fs'
+import { fileURLToPath, pathToFileURL } from 'url'
+import { tmpdir } from 'os'
+
+const __filename = fileURLToPath(import.meta.url)
+const __dirname = join(__filename, '..') // join(__dirname, '..') to get parent
+const root = join(__dirname, '..')
+
+const { parseKnowledgeIndex, matchKnowledge } = await import(
+  pathToFileURL(join(root, 'src', 'knowledge-match.js')).href
+)
+
+let passed = 0, failed = 0
+
+function assert(cond, msg) {
+  if (cond) { console.log(`  ✅ PASS: ${msg}`); passed++ }
+  else { console.log(`  ❌ FAIL: ${msg}`); failed++ }
+}
+
+function setup(name) {
+  const dir = join(tmpdir(), `kh-test-${name}`)
+  mkdirSync(dir, { recursive: true })
+  return dir
+}
+function clean(...dirs) {
+  for (const d of dirs) try { rmSync(d, { recursive: true, force: true }) } catch {}
+}
+
+// ── 1: 有匹配条目 → matched=true, entries.length > 0 ──
+console.log('\n=== Test 1: 有匹配条目 → matched=true ===')
+{
+  const dir = setup('t1')
+  try {
+    writeFileSync(join(dir, 'INDEX.md'), [
+      '# Knowledge Index',
+      '',
+      '## Conventions',
+      '- ESM|module|import → [ESM Only](conventions.md#esm-only)',
+      '',
+      '## Patterns',
+      '- 阶段定义|stage|stages → [Stage Pattern](patterns.md#stage-step-pattern)',
+    ].join('\n'))
+
+    const result = matchKnowledge(dir, 'setup ESM module imports')
+    assert(result.matched === true, 'matched is true')
+    assert(result.entries.length === 1, 'one entry matched')
+    assert(result.entries[0].file === 'conventions.md', 'file is conventions.md')
+    assert(result.entries[0].anchor === 'esm-only', 'anchor is esm-only')
+    assert(result.entries[0].keywords.includes('ESM'), 'keywords include ESM')
+  } finally { clean(dir) }
+}
+
+// ── 2: 无匹配条目 → matched=false, report 包含 "no matches" ──
+console.log('\n=== Test 2: 无匹配条目 → matched=false ===')
+{
+  const dir = setup('t2')
+  try {
+    writeFileSync(join(dir, 'INDEX.md'), [
+      '# Knowledge Index',
+      '',
+      '## Patterns',
+      '- 阶段定义|stage → [Stage Pattern](patterns.md#stage-step-pattern)',
+    ].join('\n'))
+
+    const result = matchKnowledge(dir, 'implement authentication flow')
+    assert(result.matched === false, 'matched is false')
+    assert(result.entries.length === 0, 'no entries matched')
+    assert(result.report.includes('no matches'), 'report says no matches')
+  } finally { clean(dir) }
+}
+
+// ── 3: INDEX.md 不存在 → matched=false, report 包含 "not found" ──
+console.log('\n=== Test 3: INDEX.md 不存在 → matched=false ===')
+{
+  const dir = setup('t3')
+  try {
+    const result = matchKnowledge(dir, 'any context')
+    assert(result.matched === false, 'matched is false')
+    assert(result.report.includes('not found'), 'report says not found')
+    assert(result.json.matched === false, 'json.matched is false')
+    assert(result.json.entry_count === 0, 'json.entry_count is 0')
+  } finally { clean(dir) }
+}
+
+// ── 4: INDEX.md 有空锚点引用 → 正常处理（不过度校验）──
+console.log('\n=== Test 4: INDEX.md 有空锚点引用 → 正常处理 ===')
+{
+  const dir = setup('t4')
+  try {
+    writeFileSync(join(dir, 'INDEX.md'), [
+      '# Knowledge Index',
+      '',
+      '## Conventions',
+      '- naming|camelCase → [Naming](conventions.md)',
+    ].join('\n'))
+
+    const result = matchKnowledge(dir, 'naming conventions')
+    assert(result.matched === true, 'matched is true with empty anchor')
+    assert(result.entries[0].anchor === '', 'anchor is empty string')
+    assert(result.entries[0].file === 'conventions.md', 'file is correct')
+  } finally { clean(dir) }
+}
+
+// ── 5: 大小写不敏感匹配 → "STAGE" 匹配 "stage" ──
+console.log('\n=== Test 5: 大小写不敏感匹配 ===')
+{
+  const dir = setup('t5')
+  try {
+    writeFileSync(join(dir, 'INDEX.md'), [
+      '# Knowledge Index',
+      '',
+      '## Patterns',
+      '- stage|stages → [Stage Pattern](patterns.md#stage-step-pattern)',
+    ].join('\n'))
+
+    const result = matchKnowledge(dir, 'implement STAGE definition')
+    assert(result.matched === true, 'STAGE matches stage')
+    assert(result.entries.length === 1, 'one entry matched')
+  } finally { clean(dir) }
+}
+
+// ── 6: report 格式包含 "Knowledge Context" header ──
+console.log('\n=== Test 6: report 格式包含 Knowledge Context header ===')
+{
+  const dir = setup('t6')
+  try {
+    writeFileSync(join(dir, 'INDEX.md'), [
+      '# Knowledge Index',
+      '',
+      '## Patterns',
+      '- 阶段定义|stage → [Stage Pattern](patterns.md#stage-step-pattern)',
+    ].join('\n'))
+
+    const result = matchKnowledge(dir, 'stage configuration')
+    assert(result.report.includes('Knowledge Context'), 'report has Knowledge Context header')
+    assert(result.report.includes('Status: matched'), 'report has Status: matched')
+    assert(result.report.includes('Entries: 1'), 'report has Entries: 1')
+    assert(result.report.includes('patterns.md#stage-step-pattern'), 'report has source')
+  } finally { clean(dir) }
+}
+
+// ── 7: json 结构正确 ──
+console.log('\n=== Test 7: json 结构正确 ===')
+{
+  const dir = setup('t7')
+  try {
+    writeFileSync(join(dir, 'INDEX.md'), [
+      '# Knowledge Index',
+      '',
+      '## Conventions',
+      '- ESM|module → [ESM Only](conventions.md#esm-only)',
+      '',
+      '## Patterns',
+      '- stage|stages → [Stage Pattern](patterns.md#stage-step-pattern)',
+    ].join('\n'))
+
+    const result = matchKnowledge(dir, 'stage module setup')
+    const j = result.json
+    assert(j.matched === true, 'json.matched is true')
+    assert(j.entry_count === 2, 'json.entry_count is 2')
+    assert(Array.isArray(j.entries), 'json.entries is array')
+    assert(j.entries.length === 2, 'json.entries length is 2')
+
+    const stageEntry = j.entries.find(e => e.anchor === 'stage-step-pattern')
+    assert(!!stageEntry, 'stage entry found in json')
+    assert(stageEntry.keywords.includes('stage'), 'stage keywords correct')
+    assert(stageEntry.category === 'Patterns', 'stage category correct')
+    assert(stageEntry.file === 'patterns.md', 'stage file correct')
+  } finally { clean(dir) }
+}
+
+// ── 8: 空目录（目录存在但无 INDEX.md）→ no matches ──
+console.log('\n=== Test 8: 空目录（无 INDEX.md）→ no matches ===')
+{
+  const dir = setup('t8')
+  try {
+    const result = matchKnowledge(dir, 'any task')
+    assert(result.matched === false, 'no INDEX.md → matched false')
+    assert(result.report.includes('not found'), 'report says not found')
+  } finally { clean(dir) }
+}
+
+// ── 9: parseKnowledgeIndex 返回正确分类 ──
+console.log('\n=== Test 9: parseKnowledgeIndex 分类正确 ===')
+{
+  const dir = setup('t9')
+  try {
+    writeFileSync(join(dir, 'INDEX.md'), [
+      '# Knowledge Index',
+      '',
+      '## Known Issues',
+      '- GLM|proxy → [GLM Proxy](known-issues.md#glm-proxy)',
+      '',
+      '## Conventions',
+      '- 命名|naming → [命名规范](conventions.md#naming)',
+    ].join('\n'))
+
+    const entries = parseKnowledgeIndex(dir)
+    assert(entries.length === 2, 'parsed 2 entries')
+    assert(entries[0].category === 'Known Issues', 'first entry category correct')
+    assert(entries[1].category === 'Conventions', 'second entry category correct')
+    assert(entries[0].display === 'GLM Proxy', 'display text correct')
+  } finally { clean(dir) }
+}
+
+// ── 10: taskContext 为空 → no matches ──
+console.log('\n=== Test 10: taskContext 为空 → no matches ===')
+{
+  const dir = setup('t10')
+  try {
+    writeFileSync(join(dir, 'INDEX.md'), [
+      '# Knowledge Index',
+      '',
+      '## Patterns',
+      '- stage → [Stage](patterns.md#stage)',
+    ].join('\n'))
+
+    const result = matchKnowledge(dir, '')
+    assert(result.matched === false, 'empty context → matched false')
+  } finally { clean(dir) }
+}
+
+// ── 汇总 ──
+console.log(`\n${'='.repeat(40)}`)
+console.log(`knowledge-match tests: ${passed} passed, ${failed} failed, ${passed + failed} total`)
+if (failed > 0) process.exit(1)