autosnippet
Version:
Extract code patterns into a knowledge base for AI coding assistants
287 lines (286 loc) • 12.5 kB
JavaScript
/**
* GuardCrossFileChecks - Guard 跨文件检查
*
* 从 GuardCheckEngine._runCrossFileChecks 拆分
* 包含: 跨文件规则检查 + 路径归一化工具
*/
/**
* 解析相对 import 路径为归一化路径（去掉扩展名）
* @param fromDir 当前文件目录
* @param importPath 相对路径如 './foo' 或 '../bar/baz'
*/
export function resolveImportPath(fromDir, importPath) {
try {
const parts = `${fromDir}/${importPath}`.split('/');
const resolved = [];
for (const p of parts) {
if (p === '.' || p === '') {
continue;
}
if (p === '..') {
resolved.pop();
continue;
}
resolved.push(p);
}
// 去掉扩展名归一化
let result = resolved.join('/');
result = result.replace(/\.(js|ts|jsx|tsx|mjs|mts)$/, '');
// 移除 /index 后缀（index barrel 导入）
result = result.replace(/\/index$/, '');
return result;
}
catch {
return null;
}
}
/** 归一化文件路径（去扩展名，用于 import 比较） */
export function normalizeFilePath(filePath) {
return filePath.replace(/\.(js|ts|jsx|tsx|mjs|mts)$/, '').replace(/\/index$/, '');
}
export function runCrossFileChecks(files, options = {}) {
const violations = [];
const disabledSet = new Set(options.disabledRules || []);
const isDisabled = (ruleId) => disabledSet.has(ruleId);
// 过滤掉 content 为空的条目，防止下游 split 崩溃
files = files.filter((f) => typeof f.content === 'string');
// ── ObjC Category 跨文件重名检查 ──
if (!isDisabled('objc-cross-file-duplicate-category')) {
const categoryMap = new Map();
const categoryRegex = /@interface\s+(\w+)\s*\(\s*(\w+)\s*\)/g;
for (const { path: filePath, content } of files) {
const ext = filePath.split('.').pop()?.toLowerCase();
if (ext !== 'm' && ext !== 'mm' && ext !== 'h') {
continue;
}
const lines = content.split(/\r?\n/);
for (let i = 0; i < lines.length; i++) {
categoryRegex.lastIndex = 0;
let m;
while ((m = categoryRegex.exec(lines[i])) !== null) {
const key = `${m[1]}(${m[2]})`;
if (!categoryMap.has(key)) {
categoryMap.set(key, []);
}
categoryMap.get(key).push({
filePath,
line: i + 1,
snippet: lines[i].trim().slice(0, 120),
});
}
}
}
for (const [key, locations] of categoryMap) {
if (locations.length <= 1) {
continue;
}
const hFiles = locations.filter((l) => l.filePath.endsWith('.h'));
const mFiles = locations.filter((l) => !l.filePath.endsWith('.h'));
const hasDuplicateH = hFiles.length > 1;
const hasDuplicateM = mFiles.length > 1;
const tooMany = locations.length > 2;
if (hasDuplicateH || hasDuplicateM || tooMany) {
const conflictLocations = tooMany
? locations
: hasDuplicateH && hasDuplicateM
? locations
: hasDuplicateH
? hFiles
: mFiles;
violations.push({
ruleId: 'objc-cross-file-duplicate-category',
message: `Category ${key} 在 ${conflictLocations.length} 个文件中重复声明，可能导致方法覆盖或未定义行为`,
severity: 'warning',
locations: conflictLocations,
});
}
}
} // end isDisabled('objc-cross-file-duplicate-category')
// ── JS/TS 循环依赖检查 ──
// 检测 A imports B 且 B imports A 的直接循环
if (!isDisabled('js-circular-import')) {
const jsImportMap = new Map(); // filePath → Set<importedPath>
const jsExts = new Set(['js', 'ts', 'jsx', 'tsx', 'mjs', 'mts']);
const importRegex = /(?:import\s+.+?\s+from\s+['"](.+?)['"]|require\s*\(\s*['"](.+?)['"]\s*\))/g;
for (const { path: filePath, content } of files) {
const ext = filePath.split('.').pop()?.toLowerCase() ?? '';
if (!jsExts.has(ext)) {
continue;
}
const imports = new Set();
const lines = content.split(/\r?\n/);
for (const line of lines) {
importRegex.lastIndex = 0;
let m;
while ((m = importRegex.exec(line)) !== null) {
const importPath = m[1] || m[2];
if (importPath.startsWith('.')) {
// 解析相对路径为归一化 key
const dir = filePath.substring(0, filePath.lastIndexOf('/'));
const resolved = resolveImportPath(dir, importPath);
if (resolved) {
imports.add(resolved);
}
}
}
}
if (imports.size > 0) {
jsImportMap.set(normalizeFilePath(filePath), imports);
}
}
// 检测直接双向循环: A→B 且 B→A
const reportedCycles = new Set();
for (const [fileA, importsA] of jsImportMap) {
for (const depB of importsA) {
const importsB = jsImportMap.get(depB);
if (importsB?.has(fileA)) {
const cycleKey = [fileA, depB].sort().join(' <-> ');
if (!reportedCycles.has(cycleKey)) {
reportedCycles.add(cycleKey);
violations.push({
ruleId: 'js-circular-import',
message: `检测到循环依赖，两个模块互相导入可能导致运行时 undefined`,
severity: 'warning',
locations: [
{ filePath: fileA, line: 1, snippet: `imports ${depB.split('/').pop()}` },
{ filePath: depB, line: 1, snippet: `imports ${fileA.split('/').pop()}` },
],
});
}
}
}
}
} // end isDisabled('js-circular-import')
// ── Java/Kotlin 同名类跨文件检查 ──
if (!isDisabled('java-duplicate-class-name')) {
const classMap = new Map(); // className → [{filePath, line, snippet}]
const javaClassRegex = /(?:public\s+)?(?:abstract\s+)?(?:final\s+)?class\s+(\w+)/;
const jkExts = new Set(['java', 'kt']);
for (const { path: filePath, content } of files) {
const ext = filePath.split('.').pop()?.toLowerCase() ?? '';
if (!jkExts.has(ext)) {
continue;
}
const lines = content.split(/\r?\n/);
for (let i = 0; i < lines.length; i++) {
const m = javaClassRegex.exec(lines[i]);
if (m) {
const className = m[1];
if (!classMap.has(className)) {
classMap.set(className, []);
}
classMap.get(className).push({
filePath,
line: i + 1,
snippet: lines[i].trim().slice(0, 120),
});
}
}
}
for (const [className, locations] of classMap) {
if (locations.length > 1) {
violations.push({
ruleId: 'java-duplicate-class-name',
message: `类名 "${className}" 在 ${locations.length} 个文件中定义，可能导致导入歧义`,
severity: 'info',
locations,
});
}
}
} // end isDisabled('java-duplicate-class-name')
// ── Go 多文件 init() 函数检查 ──
// 同一 package 下多个文件都有 init()，执行顺序依赖文件名排序，容易出错
if (!isDisabled('go-multiple-init')) {
const goInitMap = new Map(); // dirPath → [{filePath, line}]
for (const { path: filePath, content } of files) {
if (!filePath.endsWith('.go')) {
continue;
}
const lines = content.split(/\r?\n/);
for (let i = 0; i < lines.length; i++) {
if (/^func\s+init\s*\(\s*\)/.test(lines[i].trim())) {
const dir = filePath.substring(0, filePath.lastIndexOf('/'));
if (!goInitMap.has(dir)) {
goInitMap.set(dir, []);
}
goInitMap.get(dir).push({
filePath,
line: i + 1,
snippet: lines[i].trim().slice(0, 120),
});
break; // 每个文件只记录一次
}
}
}
for (const [dir, locations] of goInitMap) {
if (locations.length > 2) {
violations.push({
ruleId: 'go-multiple-init',
message: `同一 package (${dir.split('/').pop()}) 中 ${locations.length} 个文件都定义了 init()，执行顺序依赖文件名排序`,
severity: 'info',
locations,
});
}
}
} // end isDisabled('go-multiple-init')
// ── Swift Extension 方法跨文件冲突检查 ──
if (!isDisabled('swift-cross-file-extension-conflict')) {
const swiftExtMethodMap = new Map(); // "TypeName.methodName" → [{filePath, line}]
const swiftExtRegex = /extension\s+(\w+)/;
const swiftFuncRegex = /func\s+(\w+)\s*\(/;
for (const { path: filePath, content } of files) {
if (!filePath.endsWith('.swift')) {
continue;
}
const lines = content.split(/\r?\n/);
let currentExt = null;
let braceDepth = 0;
for (let i = 0; i < lines.length; i++) {
const extMatch = swiftExtRegex.exec(lines[i]);
if (extMatch && !currentExt) {
currentExt = extMatch[1];
braceDepth = 0;
}
if (currentExt) {
for (const ch of lines[i]) {
if (ch === '{') {
braceDepth++;
}
else if (ch === '}') {
braceDepth--;
}
}
const funcMatch = swiftFuncRegex.exec(lines[i]);
if (funcMatch && braceDepth >= 1) {
const key = `${currentExt}.${funcMatch[1]}`;
if (!swiftExtMethodMap.has(key)) {
swiftExtMethodMap.set(key, []);
}
swiftExtMethodMap.get(key).push({
filePath,
line: i + 1,
snippet: lines[i].trim().slice(0, 120),
});
}
if (braceDepth <= 0) {
currentExt = null;
}
}
}
}
for (const [key, locations] of swiftExtMethodMap) {
if (locations.length > 1) {
const uniqueFiles = new Set(locations.map((l) => l.filePath));
if (uniqueFiles.size > 1) {
violations.push({
ruleId: 'swift-cross-file-extension-conflict',
message: `Extension 方法 ${key} 在 ${uniqueFiles.size} 个文件中定义，可能导致方法冲突`,
severity: 'warning',
locations,
});
}
}
}
} // end isDisabled('swift-cross-file-extension-conflict')
return violations;
}