更新api尝试增加搜索编码的兼容性

controllers/cms.py

@@ -14,8 +14,8 @@ from utils.web import *
 from utils.system import getHost
 from utils.config import playerConfig
 from utils.log import logger
-from utils.encode import base64Encode,base64Decode,fetch,post,request,getCryptoJS,getPreJs,buildUrl,getHome
-from utils.encode import verifyCode,setDetail,join,urljoin2,parseText,requireCache,forceOrder
+from utils.encode import base64Encode,base64Decode,fetch,post,request,getCryptoJS,getPreJs,buildUrl,getHome,atob,btoa
+from utils.encode import verifyCode,setDetail,join,urljoin2,parseText,requireCache,forceOrder,base64ToImage,encodeStr, decodeStr
 from utils.encode import md5 as mmd5
 from utils.safePython import safePython
 from utils.parser import runPy,runJScode,JsObjectWrapper,PyJsObject,PyJsString
@@ -79,7 +79,8 @@ py_ctx = {
 'buildUrl':buildUrl,'getHome':getHome,'setDetail':setDetail,'join':join,'urljoin2':urljoin2,
 'PC_UA':PC_UA,'MOBILE_UA':MOBILE_UA,'UC_UA':UC_UA,'UA':UA,'IOS_UA':IOS_UA,
 'setItem':setItem,'getItem':getItem,'clearItem':clearItem,'stringify':stringify,'encodeUrl':encodeUrl,
-'requireObj':requireObj,'md5':md5
+'requireObj':requireObj,'md5':md5,'atob': atob, 'btoa':btoa,'base64ToImage': base64ToImage, 'encodeStr': encodeStr,
+    'decodeStr': decodeStr
 }
 # print(getCryptoJS())
 
@@ -166,6 +167,7 @@ class CMS:
             self.oheaders['cookie'] = cookie
         limit = rule.get('limit',6)
         encoding = rule.get('编码', 'utf-8')
+        search_encoding = rule.get('搜索编码', '')
         self.limit = min(limit,30)
         keys = headers.keys()
         for k in headers.keys():
@@ -211,6 +213,7 @@ class CMS:
         self.推荐 = rule.get('推荐','')
         self.图片来源 = rule.get('图片来源','')
         self.encoding = encoding
+        self.search_encoding = search_encoding
         self.timeout = round(int(timeout)/1000,2)
         self.filter = rule.get('filter',[])
         self.filter_def = rule.get('filter_def',{})
@@ -1250,8 +1253,12 @@ class CMS:
         return result
 
     def searchContent(self, key, fypage=1,show_name=False):
-        if self.encoding and str(self.encoding).startswith('gb'):
-            key = quote(key.encode('utf-8').decode('utf-8').encode(self.encoding,'ignore'))
+        if self.search_encoding:
+            if str(self.search_encoding).lower() != 'utf-8':
+                key = encodeStr(key,self.search_encoding)
+        elif self.encoding and str(self.encoding).startswith('gb'):
+            # key = quote(key.encode('utf-8').decode('utf-8').encode(self.encoding,'ignore'))
+            key = encodeStr(key,self.encoding)
             # print(key)
         pg = str(fypage)
         if not self.searchUrl:

doc/写源教程.md

@@ -13,6 +13,7 @@
 var rule = {
     title:'',//规则标题,没有实际作用,但是可以作为cms类名称依据
     编码:'',//不填就默认utf-8
+    搜索编码:'',//不填则不编码，默认都是按utf-8.可优先于全局编码属性.比如网页源码编码是gbk,这里可以指定utf-8搜索独立编码。多数情况这个属性不填或者填写gbk应对特殊的网站搜索
     host:'',//网页的域名根,包含http头如 https://www,baidu.com
     homeUrl:'/latest/',//网站的首页链接,可以是完整路径或者相对路径,用于分类获取和推荐获取 fyclass是分类标签 fypage是页数
     url:'/fyclass/fypage.html[/fyclass/]',//网站的分类页面链接

doc/更新日志.md

-###### 2023/02/15
+###### 2023/03/06
+- [X] 3.9.37beta1 drpy增加属性搜索编码,用于解决某些网站搜索关键字必须是gbk编码结果的问题 目前js0已测试通过并实现了encodeStr和decodeStr函数。js1预留了逻辑暂未实现，函数已增加
+
+###### 2023/03/03
 - [X] 3.9.36beta1 drpy2.js和drpy2.min.js依赖引入改为从壳子内部引入和相对路径引入模板
 
 ###### 2023/02/15

js/007影视.js

@@ -3,6 +3,7 @@ muban.首图2.二级.desc = '.data:eq(1)&&Text;;;.data:eq(6)&&Text;.data:eq(5)&&
 muban.首图2.二级.tabs = '.nav-tabs.dpplay&&li';
 var rule = {
     title:'007影视',
+    搜索编码:'utf-8',
     模板:'首图2',
     ali_token:'{{ali_token}}',
     bili_cookie:'{{bili_cookie}}',

js/version.txt

-3.9.36beta5
\ No newline at end of file
+3.9.37beta1
\ No newline at end of file

libs/drpy.js

@@ -54,7 +54,7 @@ function pre(){
 }
 
 let rule = {};
-const VERSION = 'drpy1 3.9.35 20230215';
+const VERSION = 'drpy1 3.9.37beta1 20230306';
 /** 已知问题记录
  * 1.影魔的jinjia2引擎不支持 {{fl}}对象直接渲染 (有能力解决的话尽量解决下，支持对象直接渲染字符串转义,如果加了|safe就不转义)[影魔牛逼，最新的文件发现这问题已经解决了]
  * Array.prototype.append = Array.prototype.push; 这种js执行后有毛病,for in 循环列表会把属性给打印出来 (这个大毛病需要重点排除一下)
@@ -394,6 +394,30 @@ function md5(text) {
     return CryptoJS.MD5(text).toString();
 }
 
+/**
+ * 字符串按指定编码
+ * @param input
+ * @param encoding
+ * @returns {*}
+ */
+function encodeStr(input,encoding){
+    encoding = encoding||'gbk';
+
+    return input
+}
+
+/**
+ * 字符串指定解码
+ * @param input
+ * @param encoding
+ * @returns {*}
+ */
+function decodeStr(input,encoding){
+    encoding = encoding||'gbk';
+
+    return input
+}
+
 function getCryptoJS(){
     // return request('https://gitcode.net/qq_32394351/dr_py/-/raw/master/libs/crypto-hiker.js');
     return 'console.log("CryptoJS已装载");'
@@ -2120,6 +2144,7 @@ function playParse(playObj){
 
         rule.timeout = rule.timeout||5000;
         rule.encoding = rule.编码||rule.encoding||'utf-8';
+        rule.search_encoding = rule.搜索编码||rule.search_encoding||'';
         rule.图片来源 = rule.图片来源||'';
         rule.play_json = rule.hasOwnProperty('play_json')?rule.play_json:[];
         rule.pagecount = rule.hasOwnProperty('pagecount')?rule.pagecount:{};
@@ -2265,6 +2290,15 @@ function play(flag, id, flags) {
  * @returns {string}
  */
 function search(wd, quick) {
+    if(rule.search_encoding){
+        if(rule.search_encoding.toLowerCase()!=='utf-8'){
+            // 按搜索编码进行编码
+            wd = encodeStr(wd,rule.search_encoding);
+        }
+    }else if(rule.encoding && rule.encoding.toLowerCase()!=='utf-8'){
+        // 按全局编码进行编码
+        wd = encodeStr(wd,rule.encoding);
+    }
     let searchObj = {
         searchUrl: rule.searchUrl,
         搜索: rule.搜索,

libs/drpy2.js

@@ -39,7 +39,7 @@ function pre(){
 
 let rule = {};
 let vercode = typeof(pdfl) ==='function'?'drpy2.1':'drpy2';
-const VERSION = vercode+' 3.9.36beta1 20230303';
+const VERSION = vercode+' 3.9.37beta1 20230306';
 /** 已知问题记录
  * 1.影魔的jinjia2引擎不支持 {{fl}}对象直接渲染 (有能力解决的话尽量解决下，支持对象直接渲染字符串转义,如果加了|safe就不转义)[影魔牛逼，最新的文件发现这问题已经解决了]
  * Array.prototype.append = Array.prototype.push; 这种js执行后有毛病,for in 循环列表会把属性给打印出来 (这个大毛病需要重点排除一下)
@@ -381,6 +381,30 @@ function md5(text) {
     return CryptoJS.MD5(text).toString();
 }
 
+/**
+ * 字符串按指定编码
+ * @param input
+ * @param encoding
+ * @returns {*}
+ */
+function encodeStr(input,encoding){
+    encoding = encoding||'gbk';
+
+    return input
+}
+
+/**
+ * 字符串指定解码
+ * @param input
+ * @param encoding
+ * @returns {*}
+ */
+function decodeStr(input,encoding){
+    encoding = encoding||'gbk';
+
+    return input
+}
+
 function getCryptoJS(){
     // return request('https://gitcode.net/qq_32394351/dr_py/-/raw/master/libs/crypto-hiker.js');
     return 'console.log("CryptoJS已装载");'
@@ -1983,6 +2007,7 @@ function init(ext) {
 
         rule.timeout = rule.timeout||5000;
         rule.encoding = rule.编码||rule.encoding||'utf-8';
+        rule.search_encoding = rule.搜索编码||rule.search_encoding||'';
         rule.图片来源 = rule.图片来源||'';
         rule.play_json = rule.hasOwnProperty('play_json')?rule.play_json:[];
         rule.pagecount = rule.hasOwnProperty('pagecount')?rule.pagecount:{};
@@ -2128,6 +2153,15 @@ function play(flag, id, flags) {
  * @returns {string}
  */
 function search(wd, quick) {
+    if(rule.search_encoding){
+        if(rule.search_encoding.toLowerCase()!=='utf-8'){
+            // 按搜索编码进行编码
+            wd = encodeStr(wd,rule.search_encoding);
+        }
+    }else if(rule.encoding && rule.encoding.toLowerCase()!=='utf-8'){
+        // 按全局编码进行编码
+        wd = encodeStr(wd,rule.encoding);
+    }
     let searchObj = {
         searchUrl: rule.searchUrl,
         搜索: rule.搜索,

test/测试.py

@@ -5,10 +5,29 @@
 # Date  : 2022/8/28
 
 import re
+from urllib.parse import quote,unquote
 txt = 'var player_aaaa={"flag":"play","encrypt":3,"trysee":0,"points":0,"link":"\/vodplay\/44640-1-1.html","link_next":"","link_pre":"","url":"Zd2fZg56c6y10828ZDRiNzZjNzk1Y2E3OWQzNmQzYWEyM2IwODM0ZjM3MgO0O0OO0O0O","url_next":"d","from":"vip","server":"no","note":"","id":"44640","sid":1,"nid":1}'
 ret = re.search('var player_(.*?)=(.*?)<',txt,re.M|re.I)
 print(ret)
 
+def encodeStr(input, encoding='GBK'):
+    """
+    指定字符串编码
+    :param input:
+    :param encoding:
+    :return:
+    """
+    return quote(input.encode(encoding, 'ignore'))
 
 def lazyParse(input,jsp,getParse,saveParse,headers,encoding):
-    pass
\ No newline at end of file
+    pass
+str1 = '星'
+key = str1.encode('gb2312','ignore')
+print(quote(key))
+print(quote(str1))
+# print(str1.decode('utf-8',))
+# str_gbk = str1.encode("gbk",ignore=True).strip()
+# print("转码结果："+repr(str_gbk))
+# print( unquote('%D0%C7','GBK'))
+print(encodeStr('星','utf-8'))
+print(encodeStr('斗罗大陆','gbk')) # %B6%B7%C2%DE%B4%F3%C2%BD
\ No newline at end of file

utils/encode.py

@@ -176,6 +176,33 @@ def base64Decode(text):
     # print(text)
     return base64.b64decode(text).decode("utf-8") #base64解码
 
+def encodeStr(input, encoding='GBK'):
+    """
+    指定字符串编码
+    :param input:
+    :param encoding:
+    :return:
+    """
+    if isinstance(input,PyJsString):
+        input = parseText(str(input))
+    if isinstance(encoding,PyJsString):
+        encoding = parseText(str(input))
+    return quote(input.encode(encoding, 'ignore'))
+
+def decodeStr(input, encoding='GBK'):
+    """
+    指定字符串解码
+    :param input:
+    :param encoding:
+    :return:
+    """
+    if isinstance(input,PyJsString):
+        input = parseText(str(input))
+    if isinstance(encoding,PyJsString):
+        encoding = parseText(str(input))
+    return unquote(input,encoding)
+
+
 def parseText(text:str):
     text = text.replace('false','False').replace('true','True').replace('null','None')
     # print(text)

utils/pyctx.py

@@ -9,66 +9,76 @@ import json
 import requests
 from utils.web import *
 from utils.log import logger
-from utils.encode import base64Encode,base64Decode,fetch,post,request,getCryptoJS,getPreJs,buildUrl,getHome,parseText,atob
-from utils.encode import setDetail,join,urljoin2,parseText,requireCache,base64ToImage
+from utils.encode import base64Encode, base64Decode, fetch, post, request, getCryptoJS, getPreJs, buildUrl, getHome, \
+    parseText, atob,btoa
+from utils.encode import setDetail, join, urljoin2, parseText, requireCache, base64ToImage, encodeStr, decodeStr
 from utils.encode import md5 as mmd5
-from utils.parser import runPy,runJScode,JsObjectWrapper,PyJsObject,PyJsString
+from utils.parser import runPy, runJScode, JsObjectWrapper, PyJsObject, PyJsString
 from urllib.parse import quote
 from utils.htmlParser import jsoup
 from controllers.service import storage_service
 
-def setItem(key,value):
+
+def setItem(key, value):
     lsg = storage_service()
-    if isinstance(key,PyJsString):
+    if isinstance(key, PyJsString):
         key = parseText(str(key))
-    if isinstance(value,PyJsString):
+    if isinstance(value, PyJsString):
         value = parseText(str(value))
-    return lsg.setItem(key,value)
+    return lsg.setItem(key, value)
+
 
-def getItem(key,value=''):
+def getItem(key, value=''):
     lsg = storage_service()
-    if isinstance(key,PyJsString):
+    if isinstance(key, PyJsString):
         key = parseText(str(key))
-    if isinstance(value,PyJsString):
+    if isinstance(value, PyJsString):
         value = parseText(str(value))
-    return lsg.getItem(key,value)
+    return lsg.getItem(key, value)
+
 
 def clearItem(key):
     lsg = storage_service()
-    if isinstance(key,PyJsString):
+    if isinstance(key, PyJsString):
         key = parseText(str(key))
     return lsg.clearItem(key)
 
+
 def encodeUrl(url):
     # return base64Encode(quote(url))
     # return base64Encode(url)
     # print(type(url))
-    if isinstance(url,PyJsString):
+    if isinstance(url, PyJsString):
         # obj = obj.to_dict()
         url = parseText(str(url))
     return quote(url)
 
+
 def stringify(obj):
-    if isinstance(obj,PyJsObject):
+    if isinstance(obj, PyJsObject):
         # obj = obj.to_dict()
         obj = parseText(str(obj))
     return json.dumps(obj, separators=(',', ':'), ensure_ascii=False)
 
+
 def requireObj(url):
-    if isinstance(url,PyJsString):
+    if isinstance(url, PyJsString):
         url = parseText(str(url))
     return requireCache(url)
 
+
 def md5(text):
-    if isinstance(text,PyJsString):
+    if isinstance(text, PyJsString):
         text = parseText(str(text))
     return mmd5(text)
 
+
 py_ctx = {
-'requests':requests,'print':print,'base64Encode':base64Encode,'base64Decode':base64Decode,
-'log':logger.info,'fetch':fetch,'post':post,'request':request,'getCryptoJS':getCryptoJS,
-'buildUrl':buildUrl,'getHome':getHome,'setDetail':setDetail,'join':join,'urljoin2':urljoin2,
-'PC_UA':PC_UA,'MOBILE_UA':MOBILE_UA,'UC_UA':UC_UA,'IOS_UA':IOS_UA,
-'setItem':setItem,'getItem':getItem,'clearItem':clearItem,'stringify':stringify,'encodeUrl':encodeUrl,
-'requireObj':requireObj,'md5':md5,'atob':atob,'base64ToImage':base64ToImage
-}
\ No newline at end of file
+    'requests': requests, 'print': print, 'base64Encode': base64Encode, 'base64Decode': base64Decode,
+    'log': logger.info, 'fetch': fetch, 'post': post, 'request': request, 'getCryptoJS': getCryptoJS,
+    'buildUrl': buildUrl, 'getHome': getHome, 'setDetail': setDetail, 'join': join, 'urljoin2': urljoin2,
+    'PC_UA': PC_UA, 'MOBILE_UA': MOBILE_UA, 'UC_UA': UC_UA, 'IOS_UA': IOS_UA,
+    'setItem': setItem, 'getItem': getItem, 'clearItem': clearItem, 'stringify': stringify, 'encodeUrl': encodeUrl,
+    'requireObj': requireObj, 'md5': md5, 'atob': atob, 'btoa':btoa,'base64ToImage': base64ToImage, 'encodeStr': encodeStr,
+    'decodeStr': decodeStr
+}