feat: 尝试改善hipy守护进程，难度较大，也许会出现不确定的bug

Author: Taois

docs/updateRecord.md

@@ -7,6 +7,7 @@
 1. 增加一堆 `hipy` 模板源 和相关APP模板
 2. hipy spider 增加 `setCache` `getCache` 函数
 3. map.txt 分隔符从 `@` 改为 `@@`
+4. 优化hipy守护进程执行逻辑，改善性能、稳定性、内存管理
 
 ### 20250821
 

libs/hipy.js

@@ -17,6 +17,10 @@ const _config_path = path.join(__dirname, '../config');
 const _lib_path = path.join(__dirname, '../spider/py');
 const timeout = 30000; // 30秒超时
 
+function stringify(arg) {
+    return Array.isArray(arg) || typeof arg == "object" ? JSON.stringify(arg) : arg
+}
+
 const json2Object = function (json) {
     // console.log('json2Object:', json);
     if (!json) {
@@ -107,13 +111,13 @@ const loadEsmWithHash = async function (filePath, fileHash, env) {
             `--script-path "${filePath}"`,
             `--method-name "${methodName}"`,
             `--env '${JSON.stringify(env)}'`,
-            ...args.map(arg => `--arg '${JSON.stringify(arg)}'`)
+            ...args.map(arg => `--arg '${stringify(arg)}'`)
         ].join(' ');
         // console.log(command);
         const cmd_args = [];
         args.forEach(arg => {
             cmd_args.push(`--arg`);
-            cmd_args.push(`${JSON.stringify(arg)}`);
+            cmd_args.push(`${stringify(arg)}`);
         });
         const options = {
             mode: 'text',
@@ -165,7 +169,7 @@ const loadEsmWithHash = async function (filePath, fileHash, env) {
     // 定义Spider类的方法
     const spiderMethods = [
         'init', 'home', 'homeVod', 'homeContent', 'category',
-        'detail', 'search', 'play', 'proxy', 'action', 'initEnv'
+        'detail', 'search', 'play', 'proxy', 'action'
     ];
 
     // 为代理对象添加方法
@@ -175,11 +179,6 @@ const loadEsmWithHash = async function (filePath, fileHash, env) {
         };
     });
 
-    // 处理initEnv调用
-    if (typeof spiderProxy.initEnv === 'function' && env) {
-        await spiderProxy.initEnv(env);
-    }
-
     return spiderProxy;
 }
 
@@ -194,12 +193,12 @@ const init = async function (filePath, env = {}, refresh) {
         const fileHash = computeHash(fileContent);
         const moduleName = path.basename(filePath, '.js');
         let moduleExt = env.ext || '';
-        const default_init_cfg = {
-            stype: 4, //T3/T4 源类型
-            skey: `hipy_${moduleName}`,
-            sourceKey: `hipy_${moduleName}`,
-            ext: moduleExt,
-        };
+        // const default_init_cfg = { // T3才需要这种结构
+        //     stype: 4, //T3/T4 源类型
+        //     skey: `hipy_${moduleName}`,
+        //     sourceKey: `hipy_${moduleName}`,
+        //     ext: moduleExt,
+        // };
         let SitesMap = getSitesMap(_config_path);
         if (moduleExt && SitesMap[moduleName]) {
             try {
@@ -226,7 +225,8 @@ const init = async function (filePath, env = {}, refresh) {
         module = await loadEsmWithHash(filePath, fileHash, env);
         // console.log('module:', module);
         const rule = module;
-        const initValue = await rule.init(default_init_cfg) || {};
+        // const initValue = await rule.init(default_init_cfg) || {};
+        const initValue = await rule.init(moduleExt) || {};
         let t2 = getNowTime();
         const moduleObject = deepCopy(rule);
         moduleObject.cost = t2 - t1;

spider/py/base/requirements.txt

@@ -5,4 +5,5 @@ ujson
 pyquery
 jsonpath
 json5
-jpype1
+jpype1
+pympler

spider/py/base/spider.py

@@ -51,8 +51,8 @@ def __init__(self, query_params=None, t4_api=None):
         self.t4_api = t4_api or ''
         self.extend = ''
         self.ENV = _ENV
-        # self.log(f't4_api:{t4_api}')
         self._cache = {}
+        self.log(f'BaseSpider __init__ t4_api:{t4_api}')
 
     def __new__(cls, *args, **kwargs):
         if cls._instance:
@@ -106,19 +106,13 @@ def isVideoFormat(self, url):
     def manualVideoCheck(self):
         pass
 
-    @abstractmethod
+    # @abstractmethod
     def getName(self):
-        pass
+        return 'BaseSpider'
 
     def init_api_ext_file(self):
         pass
 
-    def initEnv(self, env=None):
-        if env is None:
-            env = {}
-        self._ENV = env
-        self.t4_api = env.get('proxyUrl')
-
     def getProxyUrl(self):
         """
         获取本地代理地址