增加drpy源，尝试修bug

app.py

@@ -35,6 +35,7 @@ def create_flask_app():
 
 app = create_flask_app()
 db.init_app(app)
+db.app = app
 db.create_all(app=app)
 migrate = Migrate(app, db)
 

controllers/cms.py

@@ -152,8 +152,8 @@ class CMS:
             headers['User-Agent'] = UA
         if not 'referer' in lower_keys:
             headers['Referer'] = host
-        # print(headers)
         self.headers = headers
+        # print(headers)
         self.host = host
         self.homeUrl = urljoin(host,homeUrl) if host and homeUrl else homeUrl or host
         if url.find('[') >-1 and url.find(']') > -1:
@@ -190,7 +190,6 @@ class CMS:
             'jsp':jsoup(self.url),
             'getParse':self.getParse,
             'saveParse':self.saveParse,
-            'headers':self.headers,
             'oheaders':self.oheaders,
             'encoding':self.encoding,
             'name':self.title,
@@ -506,7 +505,7 @@ class CMS:
                 'HOST': self.host,
                 'TYPE': 'home',  # 海阔js环境标志
                 'oheaders':self.d.oheaders,
-                'fetch_params': {'headers': self.d.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
+                'fetch_params': {'headers': self.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
                 'd': self.d,
                 'getParse': self.d.getParse,
                 'saveParse': self.d.saveParse,
@@ -684,7 +683,7 @@ class CMS:
                 'input': url,
                 'TYPE': 'cate',  # 海阔js环境标志
                 'oheaders': self.d.oheaders,
-                'fetch_params': {'headers': self.d.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
+                'fetch_params': {'headers': self.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
                 'd': self.d,
                 'cateID':fyclass, # 分类id
                 'MY_CATE':fyclass, # 分类id
@@ -804,7 +803,7 @@ class CMS:
                     'TYPE': 'detail',  # 海阔js环境标志
                     'cateID': fyclass,  # 当前分类
                     'oheaders': self.d.oheaders,
-                    'fetch_params': {'headers': self.d.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
+                    'fetch_params': {'headers': self.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
                     'd': self.d,
                     'getParse': self.d.getParse,
                     'saveParse': self.d.saveParse,
@@ -889,7 +888,7 @@ class CMS:
                         'TYPE': 'detail',  # 海阔js环境标志
                         'cateID': fyclass,  # 当前分类
                         'oheaders': self.d.oheaders,
-                        'fetch_params': {'headers': self.d.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
+                        'fetch_params': {'headers': self.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
                         'd': self.d,
                         'getParse': self.d.getParse,
                         'saveParse': self.d.saveParse,
@@ -995,7 +994,7 @@ class CMS:
         # print(result)
         return result
 
-    def searchContent(self, key, fypage=1):
+    def searchContent(self, key, fypage=1,show_name=False):
         pg = str(fypage)
         if not self.searchUrl:
             return self.blank()
@@ -1013,11 +1012,12 @@ class CMS:
             py_ctx.update({
                 'input': url,
                 'oheaders': self.d.oheaders,
-                'fetch_params': {'headers': self.d.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
+                'fetch_params': {'headers': self.headers, 'timeout': self.d.timeout, 'encoding': self.d.encoding},
                 'd': self.d,
                 'KEY': key,  # 搜索关键字
                 'TYPE': 'search',  # 海阔js环境标志
-                'detailUrl': self.detailUrl or '',  # 详情页链接
+                'detailUrl': self.detailUrl or '',
+                # 详情页链接
                 'getParse': self.d.getParse,
                 'saveParse': self.d.saveParse,
                 'jsp': jsp, 'setDetail': setDetail,
@@ -1100,6 +1100,11 @@ class CMS:
                 # print(videos)
             except Exception as e:
                 logger.info(f'搜索{self.getName()}发生错误:{e}')
+        if show_name and len(videos) > 0:
+            for video in videos:
+                video['vod_name'] = self.id + ' '+video['vod_name']
+                video['vod_rule'] = self.id
+                video['vod_id'] = video['vod_id'] +'#' + self.id
         result = {
             'list': videos
         }
@@ -1148,7 +1153,7 @@ class CMS:
                     py_ctx.update({
                         'input': play_url,
                         'oheaders': self.d.oheaders,
-                        'fetch_params':{'headers':self.d.headers,'timeout':self.d.timeout,'encoding':self.d.encoding},
+                        'fetch_params':{'headers':self.headers,'timeout':self.d.timeout,'encoding':self.d.encoding},
                         'd': self.d,
                         'jxs':jxs,
                         'getParse':self.d.getParse,

controllers/vod.py

@@ -17,12 +17,72 @@ from controllers.cms import CMS
 from base.database import db
 from models.ruleclass import RuleClass
 from models.playparse import PlayParse
+from js.rules import getRules
+from concurrent.futures import ThreadPoolExecutor,as_completed,thread  # 引入线程池
 vod = Blueprint("vod", __name__)
 
+
 @vod.route('/vod')
 def vod_home():
+    def search_one(rule, wd, before: str = ''):
+        t1 = time()
+        if not before:
+            with open('js/模板.js', encoding='utf-8') as f:
+                before = f.read()
+        js_path = f'js/{rule}.js'
+        ctx, js_code = parser.runJs(js_path, before=before)
+        if not js_code:
+            return None
+        ruleDict = ctx.rule.to_dict()
+        ruleDict['id'] = rule  # 把路由请求的id装到字典里,后面播放嗅探才能用
+        logger.info(f'规则{rule}装载耗时:{get_interval(t1)}毫秒')
+        try:
+            cms = CMS(ruleDict, db, RuleClass, PlayParse, cfg)
+            data = cms.searchContent(wd,show_name=True)
+            return data
+        except Exception as e:
+            print(f'{rule}发生错误:{e}')
+            return None
+
+    def multi_search(wd):
+        rules = getRules('js')['list']
+        rules_exclude = ['drpy']
+        new_rules = list(filter(lambda x: x.get('searchable', 0) and x.get('name', '') not in rules_exclude, rules))
+        search_sites = [new_rule['name'] for new_rule in new_rules]
+        logger.info(f'开始聚搜{wd},共计{len(search_sites)}个规则')
+        timeout = 5
+        res = []
+        with open('js/模板.js', encoding='utf-8') as f:
+            before = f.read()
+        with ThreadPoolExecutor(max_workers=len(search_sites)) as executor:
+            to_do = []
+            for site in search_sites:
+                future = executor.submit(search_one, site, wd, before)
+                to_do.append(future)
+            try:
+                for future in as_completed(to_do, timeout=timeout):  # 并发执行
+                    ret = future.result()
+                    print(ret)
+                    if ret:
+                        res.extend(ret['list'])
+            except Exception as e:
+                print(f'发生错误:{e}')
+                import atexit
+                atexit.unregister(thread._python_exit)
+                executor.shutdown = lambda wait: None
+        return jsonify({
+            "list": res
+        })
+
     t0 = time()
     rule = getParmas('rule')
+    ac = getParmas('ac')
+    ids = getParmas('ids')
+    if ac and ids and ids.find('#') > -1:  # 聚搜的二级
+        id_list = ids.split(',')
+        rule = id_list[0].split('#')[1]
+        print(rule)
+
     ext = getParmas('ext')
     filters = getParmas('f')
     tp = getParmas('type')
@@ -58,15 +118,13 @@ def vod_home():
     # print(rule)
     cms = CMS(ruleDict,db,RuleClass,PlayParse,cfg)
     wd = getParmas('wd')
-    ac = getParmas('ac')
     quick = getParmas('quick')
     play = getParmas('play') # 类型为4的时候点击播放会带上来
     flag = getParmas('flag') # 类型为4的时候点击播放会带上来
-    filter = getParmas('filter')
+    # myfilter = getParmas('filter')
     t = getParmas('t')
     pg = getParmas('pg','1')
     pg = int(pg)
-    ids = getParmas('ids')
     q = getParmas('q')
     play_url = getParmas('play_url')
 
@@ -105,15 +163,21 @@ def vod_home():
         return jsonify(data)
     if ac and ids: # 二级
         id_list = ids.split(',')
+        if ids.find('#') > -1:
+            id_list = list(map(lambda x:x.split('#')[0],id_list))
         # print('app:377',len(id_list))
         # print(id_list)
         data = cms.detailContent(pg,id_list)
         # print(data)
         return jsonify(data)
     if wd: # 搜索
-        data = cms.searchContent(wd)
-        # print(data)
-        return jsonify(data)
+        if rule == 'drpy':
+            # print(f'准备单独处理聚合搜索:{wd}')
+            return multi_search(wd)
+        else:
+            data = cms.searchContent(wd)
+            # print(data)
+            return jsonify(data)
     # return jsonify({'rule':rule,'js_code':js_code})
     home_data = cms.homeContent(pg)
     return jsonify(home_data)
\ No newline at end of file

js/菜狗.js

@@ -10,7 +10,7 @@ var rule = {
     url:'/napi/video/classlist?abtest=0&iploc=CN1304&spver=&listTab=fyclass&filter=&start=((fypage-1)*15)&len=15&fr=filter',
     filter_url:'style={{fl.style}}&zone={{fl.zone}}&year={{fl.year}}&fee={{fl.fee}}&order={{fl.order}}',
     headers:{
-        'User-Agent':'MOBILE_UA'
+        'User-Agent':'PC_UA'
     },
     timeout:5000,
     class_name:'电视剧&电影&动漫&综艺&纪录片',
@@ -26,5 +26,5 @@ var rule = {
     二级:'js:var vod={vod_id:input};let html=request(input);function adhead(url){let hd="https://v.sogou.com";if(!url.startsWith(hd)){url=hd+url}return urlencode(url)}try{let json=JSON.parse(html.match(/INITIAL_STATE.*?({.*});/)[1]).detail.itemData;let key=json.dockey;let name=json.name;let zone=json.zone;let score=json.score?json.score:"暂无";let style=json.style;let emcee=json.emcee?"主持："+json.emcee:json.name;let director=json.director?"导演："+json.director:name;director=director.replace(/;/g,"\\t");let starring=json.starring?"演员："+json.starring:"声优："+json.shengyou;starring=starring.replace(/.*undefined/,"").replace(/;/g,"\\t");let update=json.update_wordstr?json.update_wordstr:"";let tv_station=json.tv_station?json.tv_station:zone;let introduction=json.introduction;let shengyou=json.shengyou;let shows=json.play_from_open_index;let plays=json.play.item_list;if(shows){vod.vod_name=name;vod.vod_area=emcee+","+tv_station;vod.vod_director=director;vod.vod_actor=starring;vod.vod_pic=jsp.pd(html,"#thumb_img&&img&&src");vod.vod_remarks=style+" 评分:"+score+","+update;vod.vod_content=introduction}else{vod.vod_name=name;vod.vod_director=director;vod.vod_actor=starring;vod.vod_pic=jsp.pd(html,"#thumb_img&&img&&src");vod.vod_content=introduction}let tp="&type=json";try{let tabs=[];let lists=[];plays.forEach(function(it){lists.push(it.info);let tbn=it.sitename[0]||it.site.replace(".com","");tbn=tbn.split("").join(" ");tabs.push(tbn)});vod.vod_play_from=tabs.join("$$$");vod_lists=[];play_url=play_url.replace("&play_url=","&type=json&play_url=");lists.forEach(function(item,idex){if(item||shows){if(item&&Array.isArray(item)&&item.length>1){let tmp=item.slice(1).map(function(its){return its.index+"$"+play_url+base64Encode(adhead(its.url))});vod_lists.push(tmp.join("#"))}if(shows){let arr=[];let tmp=[];let zy=shows.item_list[idex];zy.date.forEach(function(date){let day=date.day;for(let j=0;j<day.length;j++){let dayy=day[j][0]>=10?day[j][0]:"0"+day[j][0];let Tdate=date.year+date.month+dayy;arr.push(Tdate)}});for(let k=0;k<arr.length;k++){let url="https://v.sogou.com/vc/eplay?query="+arr[k]+"&date="+arr[k]+"&key="+key+"&st=5&tvsite="+plays[idex].site;tmp.push("第"+arr[k]+"期"+"$"+play_url+base64Encode(adhead(url)))}vod_lists.push(tmp.join("#"))}}else if(plays[idex].site){let tmp=[];if(!plays[idex].flag_list.includes("trailer")){tmp.push(plays[idex].sitename[0]+"$"+play_url+base64Encode(adhead(plays[idex].url)))}else{tmp.push(plays[idex].sitename[0]+"—预告"+"$"+play_url+base64Encode(adhead(plays[idex].url)))}vod_lists.push(tmp.join("#"))}});vod.vod_play_url=vod_lists.join("$$$")}catch(e){let img=json.photo.item_list;vod.vod_name="本片无选集";vod.vod_pic=img.length>0?img[0]:""}}catch(e){}',
     // 二级:'js:var vod={vod_id:input};let html=request(input);function adhead(url){let hd="https://v.sogou.com";if(!url.startsWith(hd)){url=hd+url}return url}try{let json=JSON.parse(html.match(/INITIAL_STATE.*?({.*});/)[1]).detail.itemData;let key=json.dockey;let name=json.name;let zone=json.zone;let score=json.score?json.score:"暂无";let style=json.style;let emcee=json.emcee?"主持："+json.emcee:json.name;let director=json.director?"导演："+json.director:name;director=director.replace(/;/g,"\\t");let starring=json.starring?"演员："+json.starring:"声优："+json.shengyou;starring=starring.replace(/.*undefined/,"").replace(/;/g,"\\t");let update=json.update_wordstr?json.update_wordstr:"";let tv_station=json.tv_station?json.tv_station:zone;let introduction=json.introduction;let shengyou=json.shengyou;let shows=json.play_from_open_index;let plays=json.play.item_list;if(shows){vod.vod_name=name;vod.vod_area=emcee+","+tv_station;vod.vod_director=director;vod.vod_actor=starring;vod.vod_pic=jsp.pd(html,"#thumb_img&&img&&src");vod.vod_remarks=style+" 评分:"+score+","+update;vod.vod_content=introduction}else{vod.vod_name=name;vod.vod_director=director;vod.vod_actor=starring;vod.vod_pic=jsp.pd(html,"#thumb_img&&img&&src");vod.vod_content=introduction}let tp="&type=json";try{let tabs=[];let lists=[];plays.forEach(function(it){lists.push(it.info);let tbn=it.sitename[0]||it.site.replace(".com","");tbn=tbn.split("").join(" ");tabs.push(tbn)});vod.vod_play_from=tabs.join("$$$");vod_lists=[];play_url=play_url.replace("&play_url=","&type=json&play_url=");lists.forEach(function(item,idex){if(item||shows){if(item&&Array.isArray(item)&&item.length>1){let tmp=item.slice(1).map(function(its){return its.index+"$"+play_url+base64Encode(adhead(its.url))});vod_lists.push(tmp.join("#"))}if(shows){let arr=[];let tmp=[];let zy=shows.item_list[idex];zy.date.forEach(function(date){let day=date.day;for(let j=0;j<day.length;j++){let dayy=day[j][0]>=10?day[j][0]:"0"+day[j][0];let Tdate=date.year+date.month+dayy;arr.push(Tdate)}});for(let k=0;k<arr.length;k++){let url="https://v.sogou.com/vc/eplay?query="+arr[k]+"&date="+arr[k]+"&key="+key+"&st=5&tvsite="+plays[idex].site;tmp.push("第"+arr[k]+"期"+"$"+play_url+base64Encode(adhead(url)))}vod_lists.push(tmp.join("#"))}}else if(plays[idex].site){let tmp=[];if(!plays[idex].flag_list.includes("trailer")){tmp.push(plays[idex].sitename[0]+"$"+play_url+base64Encode(adhead(plays[idex].url)))}else{tmp.push(plays[idex].sitename[0]+"—预告"+"$"+play_url+base64Encode(adhead(plays[idex].url)))}vod_lists.push(tmp.join("#"))}});vod.vod_play_url=vod_lists.join("$$$")}catch(e){let img=json.photo.item_list;vod.vod_name="本片无选集";vod.vod_pic=img.length>0?img[0]:""}}catch(e){}',
     搜索:'',
-    搜索:'js:let d=[];let html=request(input);let jsonA=JSON.parse(html.match(/INITIAL_STATE.*?({.*});/)[1]).result.longVideo.results;jsonA.forEach(function(it){let name=it.name;let introduction=it.introduction;let pic=it.v_picurl;let url=it.tiny_url;let zone=it.zone;let score=it.score||"暂无";let style=it.style;if(it.play.item_list){let r={};r.title=name.replace(//,"").replace(//,"");r.url="https://v.sogou.com"+url;r.desc=it.list_category.join(",");r.content=introduction;r.pic_url=pic;d.push(r)}});setResult(d);',
+    搜索:'js:let d=[];let html=request(input);let jsonA=JSON.parse(html.match(/INITIAL_STATE.*?({.*});/)[1]);print(jsonA);jsonA=jsonA.result.longVideo.results;jsonA.forEach(function(it){let name=it.name;let introduction=it.introduction;let pic=it.v_picurl;let url=it.tiny_url;let zone=it.zone;let score=it.score||"暂无";let style=it.style;if(it.play.item_list){let r={};r.title=name.replace(//,"").replace(//,"");r.url="https://v.sogou.com"+url;r.desc=it.list_category.join(",");r.content=introduction;r.pic_url=pic;d.push(r)}});setResult(d);',
 }
\ No newline at end of file

libs/pre.js

@@ -227,7 +227,7 @@ function request(url,obj){
     }else{
         new_obj = obj||{}
     }
-    if(!obj||!obj.headers||(!obj.headers['User-Agent']&&!obj.headers['user-agent'])){
+    if(!new_obj||!new_obj.headers||(!new_obj.headers['User-Agent']&&!new_obj.headers['user-agent'])){
         new_obj.headers['User-Agent'] = MOBILE_UA;
     }
     // delete new_obj.headers['Referer'];

py/搜狗搜索.js

 js:
 let d=[];
 let html = request(input);
-let jsonA = JSON.parse(html.match(/INITIAL_STATE.*?({.*});/)[1]).result.longVideo.results;
+// print(html);
+let jsonA = JSON.parse(html.match(/INITIAL_STATE.*?({.*});/)[1]);
+print(jsonA);
+jsonA = jsonA.result.longVideo.results;
 jsonA.forEach(function (it){
     let name=it.name;
     let introduction=it.introduction;

utils/web.py

@@ -83,8 +83,8 @@ def get_interval(t):
 def getHeaders(url):
     headers = {}
     if url:
-        headers.setdefault("Referer", url)
-    headers.setdefault("User-Agent", PC_UA)
-    headers.setdefault("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9")
-    headers.setdefault("Accept-Language", "zh-CN,zh;q=0.8,zh-TW;q=0.7,zh-HK;q=0.5,en-US;q=0.3,en;q=0.2")
+        headers["Referer"] = url
+    headers["User-Agent"] = UA
+    # headers.setdefault("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9")
+    # headers.setdefault("Accept-Language", "zh-CN,zh;q=0.8,zh-TW;q=0.7,zh-HK;q=0.5,en-US;q=0.3,en;q=0.2")
     return headers
\ No newline at end of file