From 3e628010525e2a6ac1283175636e64c4af8863dd Mon Sep 17 00:00:00 2001
From: llluiop <290522165@qq.com>
Date: Mon, 25 May 2015 10:08:03 +0800
Subject: [PATCH] add 0009

---
 llluiop/0009/FindLinks.py | 37 +++++++++++++++++++++++++++++++++++++
 1 file changed, 37 insertions(+)
 create mode 100644 llluiop/0009/FindLinks.py

diff --git a/llluiop/0009/FindLinks.py b/llluiop/0009/FindLinks.py
new file mode 100644
index 00000000..6cd2f326
--- /dev/null
+++ b/llluiop/0009/FindLinks.py
@@ -0,0 +1,37 @@
+#!/usr/bin.env python
+
+
+from HTMLParser import HTMLParser
+from re import sub
+import urllib2
+import sys
+
+
+class HtmlParserMainText(HTMLParser):
+    def __init__(self):
+        HTMLParser.__init__(self)
+        self.text = []
+
+    def handle_starttag(self, tag, attrs):
+        for key, value in attrs:
+            if value and 'http' in value:
+                self.text.append(''.join(value) + '\n')
+
+
+
+
+def GetLinks():
+    url = "http://www.cnbeta.com/"
+    html = urllib2.urlopen(url).read()
+
+    parser = HtmlParserMainText()
+    parser.feed(html)
+    parser.close()
+
+    return ''.join(parser.text).strip()
+
+
+if __name__ == "__main__":
+    reload(sys)
+    sys.setdefaultencoding('utf-8')
+    print GetLinks()