From 104faac5e1bd87a9dd0d5a5733fe94db467495ef Mon Sep 17 00:00:00 2001
From: wwqgtxx <wwqgtxx@gmail.com>
Date: Mon, 13 Jun 2016 13:17:18 +0800
Subject: [PATCH] make urlcache be clean when len(urlcache)>100

Signed-off-by: wwqgtxx <wwqgtxx@gmail.com>
---
 wwqLyParse/common.py | 19 ++++++++++++++++---
 wwqLyParse/main.py   |  2 +-
 2 files changed, 17 insertions(+), 4 deletions(-)

diff --git a/wwqLyParse/common.py b/wwqLyParse/common.py
index ca8afba..6945435 100644
--- a/wwqLyParse/common.py
+++ b/wwqLyParse/common.py
@@ -1,15 +1,19 @@
 #!/usr/bin/env python3.5
 # -*- coding: utf-8 -*-
 # author wwqgtxx <wwqgtxx@gmail.com>
-import urllib.request,io,os,sys,json,re,gzip
+import urllib.request,io,os,sys,json,re,gzip,time
 
 urlcache = {}
 def getUrl(oUrl, encoding = 'utf-8' , headers = {}, data = None, method = None) :
     url_json = {"oUrl":oUrl,"encoding":encoding,"headers":headers,"data":data,"method":method}
     url_json = json.dumps(url_json,sort_keys=True, ensure_ascii=False)
     if url_json in urlcache:
-        html_text = urlcache[url_json]
+        item = urlcache[url_json]
+        html_text = item["html_text"]
+        item["lasttimestap"] = int(time.time())
         print("cache get:"+url_json)
+        if (len(urlcache)>100):
+            cleanUrlcache()
         return html_text
     print("normal get:"+url_json)
     # url 包含中文时 parse.quote_from_bytes(oUrl.encode('utf-8'), ':/&%?=+')
@@ -26,9 +30,18 @@ def getUrl(oUrl, encoding = 'utf-8' , headers = {}, data = None, method = None)
             html_text = data.decode(encoding,'ignore')
         else:
             html_text = blob.decode(encoding,'ignore')
-        urlcache[url_json] = html_text
+        urlcache[url_json] = {"html_text":html_text,"lasttimestap":int(time.time())}
         return html_text
 
+def cleanUrlcache():
+    sortedDict = sorted(urlcache.items(), key=lambda d: d[1]["lasttimestap"], reverse=True)
+    newDict = sortedDict[:100] # 从数组中取索引start开始到end-1的记录
+    urlcache = newDict
+    print("urlcache has been cleaned")
+        
+def debug(input):
+    print (((str(input))).encode('gbk', 'ignore').decode('gbk') )
+        
 class Parser(object):
     filters = []
     def Parse(self,url,types=None):
diff --git a/wwqLyParse/main.py b/wwqLyParse/main.py
index 0a9e06f..dc836fe 100644
--- a/wwqLyParse/main.py
+++ b/wwqLyParse/main.py
@@ -37,7 +37,7 @@
 version = {
     'port_version' : "0.5.0", 
     'type' : 'parse', 
-    'version' : '0.2.2', 
+    'version' : '0.2.3', 
     'uuid' : '{C35B9DFC-559F-49E2-B80B-79B66EC77471}',
     'filter' : [],
     'name' : 'WWQ猎影解析插件',