首页 > 代码库 > split function of Perl,Python,Awk

split function of Perl,Python,Awk

使用中常用到Perl,Python,AWK,R, 虽然Java,C,C++,Vala也学过但是就是不喜欢,你说怎么办。

看来一辈子脚本的命。

Perl

@rray = split /PATTERN/, STRING, LIMIT

可以看出split由2部分(STRING,PATTERN)和可选的LIMIT部分构成,反正split么,万变不离其宗,都要有

你要split的String,split的界定,split的保存,其他的可以外加

我们举一个简单的例子:

> cat test.txt (为了对齐,黄色表示<tab>,绿色表示一个或者多个空格)

[我们要把其中的数字和单词提取出来]

14    yahoo    17:56 Ray---boring

> perl -e ‘$str="14    yahoo\t 17:56 Ray---boring";@num_word=split /[\s:\-]+/, $str;print "@num_word\n"‘

14 yahoo 17 56 Ray boring

复杂一点的例子:

 

[我们要提取浅蓝色的部分]

 

../Sample_Tgh_leaf_1_rRNA_removal_20140624/Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R1_001.fastq.gz

 

../Sample_Tgh_leaf_1_rRNA_removal_20140624/Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R2_001.fastq.gz

 

> perl -e ‘$str="../Sample_Tgh_leaf_1_rRNA_removal_20140624/Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R1_001.fastq.gz";$name = join "_", @{[split /_/, [split /\// , $str]->[2]]}[0..7];print "$name\n"‘

Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008

我们解析一下:

split /\// , $str 把字符串分为3段"..","Sample_Tgh_leaf_1_rRNA_removal_20140624","Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R1_001.fastq.gz"

[split /\// , $str] 将split的结果变成一个匿名数组

[split /\// , $str]->[2] 是通过引用取得这个匿名数组的第3个元素

[split /_/, [split /\// , $str]->[2]] 是又将split结果变成一个数组(为10个元素) "Tgh","leaf","1","rRNA","removal","20140624","ATTCCT","L008","R1","001.fastq.gz"

由于匿名数组识别数组切片([array]->[0..7]行不通),所以需要把这个数组显性の加上@{}数组标记

@{[split /_/, [split /\// , $str]->[2]]}[0..7] 得到了一个长度为8的数组切片,仍为数组 "Tgh","leaf","1","rRNA","removal","20140624","ATTCCT","L008"

join "_", @rray 把这些元素连接起来,变成一个字符串:Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008

 

[要注意,分隔符部分是PATTERN,即正则表达式,不是字符串]

[其实可以直接只用一个split,什么参数也不带,这样默认的PATTREN为空格,默认的STRNIING为$_种的字符串(这个是Perl的通行证),这个用在函数,for循环等默认参数是$_的很方便]

例如:

A BC DEF GHIJ

KLMN OPQ RS T

U VWX YZ

> perl -e ‘@rray=("A BC DEF GHIJ","KLMN OPQ RS T","U VWX YZ");@all=();for (@rray){@tmp=split;push @all, @tmp};print "@all\n"‘

A BC DEF GHIJ KLMN OPQ RS T U VWX YZ

 

Python

Python的很人性化,但是不强大,可是通过其他方式提供和更强大的re.split

由于Python的一切都是对象和类,所以么,

简单的例子:

14    yahoo    17:56 Ray---boring

eehhhh, str类的split无法完成上面的分割(也可以但是你要想一想,看Attachment 1 below),只能借助re.split来完成

py3 -c ‘import re;stri="14    yahoo    17:56 Ray---boring";print(re.split("[\s:\-]+",stri));‘

[‘14‘, ‘yahoo‘, ‘17‘, ‘56‘, ‘Ray‘, ‘boring‘]

简单的例子:

[我们要提取浅蓝色的部分]

../Sample_Tgh_leaf_1_rRNA_removal_20140624/Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R1_001.fastq.gz

../Sample_Tgh_leaf_1_rRNA_removal_20140624/Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R2_001.fastq.gz

> py3 -c ‘stri="../Sample_Tgh_leaf_1_rRNA_removal_20140624/Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R1_001.fastq.gz";print("_".join(stri.split("/")[2].split("_")[0:7]))‘

Tgh_leaf_1_rRNA_removal_20140624_ATTCCT

这个用python实现的话就很简单

[split也可以什么都不加,只加一对空括号,默认是以一个或者多个空格分割的]

 

AWK

split(string,array,sep)是个函数,那种很老的函数

[我们要提取浅蓝色的部分]

../Sample_Tgh_leaf_1_rRNA_removal_20140624/Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R1_001.fastq.gz

../Sample_Tgh_leaf_1_rRNA_removal_20140624/Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R2_001.fastq.gz

echo ../Sample_Tgh_leaf_1_rRNA_removal_20140624/Tgh_leaf_1_rRNA_removal_20140624_ATTCCT_L008_R1_001.fastq.gz | awk {split($1,aa,"/");split(aa[3],bb,"_");} END{for(i=1;i<=8;i++){if(i<8){printf("%s_",bb[i])}else{print bb[i]}}}

或者

awk -F‘/‘ ‘{split($3,bb,"_");}END{for(i=1;i<=8;i++){if(i<8){printf("%s_",bb[i])}else{print bb[i]}}}‘

Attachment:

1. py3 -c ‘import re;stri="14    yahoo    17:56 Ray---boring";orig=stri.split();part2=orig[2].split(":");part3=orig[3].split("---");print(orig[0:2]+part2+part3);‘

 

split function of Perl,Python,Awk