diff --git a/README.md b/README.md
index f06e4fd..cec0063 100644
--- a/README.md
+++ b/README.md
@@ -4,5 +4,6 @@ Apple Daily HK Crawler
 
 ## Usage
 ```bash
-python crawl.py  20150101 20150101.json
+python crawl.py 20150101 20150101.json
 ```
+
diff --git a/crawl.py b/crawl.py
index 939d79f..dc5e183 100644
--- a/crawl.py
+++ b/crawl.py
@@ -13,7 +13,7 @@ def fetch(item):
         root = etree.HTML(r.text)
         item['title'] = root.xpath("//table[@class=\"LinkTable\"]/tr/td/h1")[0].text.strip()
         item['image'] = root.xpath("//meta[@property=\"og:image\"]")[0].attrib['content'].strip()
-        item['text'] = ''.join([s.strip() for s in root.xpath("//div[@id=\"masterContent\"]")[0].itertext()]).strip()
+        item['text'] = re.sub('(please only add this icon at the end of the article)','',''.join([s.strip() for s in root.xpath("//div[@id=\"masterContent\"]")[0].itertext()]).strip())
     except Exception as e:
         print("cannot parse %s" % (item['link']))
         raise