diff --git a/get_rmrbtop10.py b/get_rmrbtop10.py
new file mode 100644
index 0000000..8ba93b0
--- /dev/null
+++ b/get_rmrbtop10.py
@@ -0,0 +1,71 @@
+# !/usr/bin/env python
+# -*- coding:utf-8 -*- 
+
+import datetime
+import lxml.html
+from lxml import etree
+import re
+import os
+from docx import Document
+from docx.enum.style import WD_STYLE_TYPE
+from docx.shared import Pt 
+
+try:
+    from urllib.request import urlopen, Request
+except ImportError:
+    from urllib2 import urlopen, Request
+
+def today():
+	today = datetime.datetime.today().date()
+	return str(today)
+	# return '2015-04-27'
+
+def get_rmrb_top10(save_dir):
+	url = r'http://paperpost.people.com.cn/rmrb-%s.html' %today()
+	text = urlopen(url).read()
+
+	pattern = re.compile(r'<a href=\"(.*?)\"')
+	top10_urls = re.findall(pattern,text)
+
+	for eachurl in top10_urls:
+		try:
+			html = lxml.html.parse(eachurl)
+		except:
+			continue
+
+		res = html.xpath('//div[@id=\"ozoom\"]/p')
+		title = html.xpath('//h1')[0].text
+		author = html.xpath('//h4')[0].text
+		newdoc = Document()
+
+
+		# obj_styles = newdoc.styles
+		# obj_charstyle = obj_styles.add_style('CommentsStyle', WD_STYLE_TYPE.PARAGRAPH)
+		# obj_font = obj_charstyle.font
+		# obj_font.size = Pt(16)
+		# obj_font.name = u'方正仿宋_GBK'
+
+
+		newdoc.add_heading(title,0)
+		newdoc.add_paragraph(author)
+
+
+		for node in res:
+			# newdoc.add_paragraph(node.text,style = 'CommentsStyle')
+			paragraph = newdoc.add_paragraph()
+			paragraph_format = paragraph.paragraph_format
+			paragraph_format.line_spacing = Pt(30)
+			run = paragraph.add_run(node.text)
+			font = run.font
+			font.name = u'方正仿宋_GBK'
+			font.size = Pt(16)
+
+		newdoc.save(save_dir+title+r'.docx')
+
+
+if __name__ == '__main__':
+	save_dir =  r'/root/rmrb_top10/%s/'%(today())
+	if not os.path.isdir(save_dir):
+		os.mkdir(save_dir)
+
+	get_rmrb_top10(save_dir)