python VS awk 在处理文本的不同之处

题目在这里 ，bbs有问题，我的几个回帖都被删了.

#!/usr/bin/env python

import subprocess

spec ={'node': ((53,102),(103,122),(223,376)),

        'renderG' : (('1-166','181-202'),('167-180','203-204'))
}

cmd ="qbhosts | awk '/node/||/renderG/'"

output = subprocess.Popen(cmd,shell=True,stdout=subprocess.PIPE).communicate()[0]

d ={}

for i in output.splitlines():

    items = i.split()
    if not items[0].endswith("xxx.com"):
        items[0]=items[0]+".xxx.com"
    d[items[0]] = items[3]

for k,v in spec.items() :

                for i in v:
                        # set default value for m1,m2.
                        m1=m2=None

                        if type(i[0]) == str or type(i[1]) == str :

                                begin = i[0].split("-")[0]
                                m1 = i[0].split("-")[1]
                                m2 = i[1].split("-")[0]
                                end = i[1].split("-")[1]
                                host = [ "%s-%03d.xxx.com" %(k,m ) for m in xrange(int(begin),int(end)+1) if m not in xrange(int(m1)+1,int(m2))]

                        else :
                                begin,end = i[0],i[1]
                                host = [ "%s-%03d.xxx.com" %(k,m) for m in xrange(begin,end+1) ]
                        #print "--------\n",host,"\n----------------"

                        temp = {}

                        for h in host :

                                if h in d.keys():
                                        temp[h] = d[h]

                                #else : print "%s doesn't exist."%h

                        act = len([k1 for k1,v1 in temp.items() if v1 =="active" ])
                        down = len([k1 for k1,v1 in temp.items() if v1 =="down" ])

                        if m1 and m2 :

                                print "\n%s[%03d-%03d,%03d-%03d].xxx.com have %d active node(s), %d down node(s)\n" % (k,int(begin),int(m1),int(m2),int(end),act,down)

                        else:
                                print "\n%s[%03d-%03d].xxx.com have %d active node(s), %d down node(s)\n" % (k,int(begin),int(end),act,down)

                        temp.clear()

相对而言，awk的简洁一些，但有些问题，

awk 'NR==FNR{

for(i=$2;i<=$3;i++){a[$1" "i]=$2;b[$1" "$2]=$3;}

if(NF>3)  {  b[$1" "$2]=$3" Exclude ["$4"-"$5"]";


for(i=$4;i<=$5;i++)


delete a[$1" "i]

}


}NR>FNR{



split($1,c,"[-.]")


n=c[1]" "(c[2]+0)


if(n in a)  d[c[1]" "a[n]][$4]++    # d[node" "53]["active"] ++ ； d[node" "103]["down"] ++


}END{for(i in d)for(j in d[i])print i"-"b[i],j,d[i][j]}'

gawk 4.0+ 可用。

题外话：要处理文件，截取固定的行，read/write都有3个，read/readline/readlines , read每次只读取若干字节，readline、readlines很相似，后者是一次性读取完，前者逐行（必须有while True之类的来保证继续），而且数据类型也不同，readlines是一个list.
他们都区别还在于是否有尾部的\n，要想往回读文件，要seek了。

例如：
with open("sample","r") as f:

        for i in f.readlines()[2:-4]:
                s =i.split()
                print "INSERT INTO tab VALUES (%s )"%",".join("'%s'"%i for i in s)

f.close()