#!/usr/bin/python
import sys
import re
f=open(sys.argv[1],"r")  #传参
sum=0
list=[]
for line in f:
        if line.startswith('#'): #判断开头
                continue         #跳过    
        line=line.rstrip()       #相当于perl的chomp    
        exons=re.findall('[0-9]+-[0-9]+',line) #正则，返回列表
    	for exon in exons:
        	    list.append(exon)
st=set(list)                    #去重
for i in st:                    
	    exon_one=i.split('-')   
	    exon_start=exon_one[0]
	    exon_end=exon_one[1]
	    sum+=int(exon_end)-int(exon_start)
print sum ```

wget ftp://ftp.ncbi.nlm.nih.gov/pub/CCDS/current_human/CCDS.20160908.txt