awk¶
判断1-100,如果数字为3的倍数, 输出Three, 数字为5的倍数,
输出Five,同时为3,5的倍数, 输出three and five
seq 100|awk '{if($0 % 3 == 0) print $0,"Three" ; \
if($0 % 5 == 0 )print $0,"Five"; \
if($0 % 3 == 0 && $0 % 5 == 0) print $0,"three and five"; \
if($0 % 3 != 0 && $0 % 5 != 0) print $0,"0";}'
输出单双引号¶
# 双引号
awk '{print "\""}'
# 单引号
echo 1 | awk '{print "\047" }'
# 或者
awk '{print "'\''"}'
awk 多个$用法¶
[yjj@centos ~]$ cat a
1 2 3 4 5
1 2 3 a 5
1 2 3 b 5
[yjj@centos ~]$ awk '{print $$4}' a
4
1 2 3 a 5
1 2 3 b 5
awk 字符集问题¶
export LC_ALL=C
echo {a..e} {A..E} {a..e}|xargs -n1|awk '/[A-Z]/'
A
B
C
D
E
awk FIELDWIDTHS¶
[root@centos ~]# echo "20161002"|awk -vFIELDWIDTHS="4 2 2" -vOFS="-" '{print $1,$2,$3}'
2016-10-02
[root@centos ~]# awk -v FIELDWIDTHS="4 2 2" -v OFS="-" '{print $1,$2,$3}' <<<20160202
2016-02-02
[root@centos ~]# awk -vFIELDWIDTHS="4 2 2" '{printf "%s-%s-%s\n",$1,$2,$3}' <<<20160202
2016-02-02
[root@centos ~]# sed -r 's@(.{4})(..)(..)@\1-\2-\3@' <<<"20161002"
2016-10-02
案例¶
成绩统计案例¶
[yjj@centos awk]$ cat chengji
yuwen shuxue yingyu
10 10 10
10 10 10
10 10 10
[yjj@centos awk]$ awk '{a=$1+$2+$3;print $0 "\t" (NR==1?"Total":a)}' chengji
yuwen shuxue yingyu Total
10 10 10 30
10 10 10 30
10 10 10 30
[yjj@centos awk]$ awk '{NR==1;$4="Total"}NR>1{$4=$1+$2+$3}1' OFS="\t" chengji
yuwen shuxue yingyu Total
10 10 10 30
10 10 10 30
10 10 10 30
用awk如何自动填充空数据的列为最近的不为空的数据¶
用awk如何自动填充空数据的列为最近的不为空的数据？
比如以下文本：
name1,1,21,address1
name2,0,,
name3,0,,
name4,1,30,address4
name5,0,24,address5
name6,1,,
name7,1,29,address7
其中name2、name3和name6的第三列和第四列都为空值，我想实现这些空值自动填充为它们上方的相应列不为空的数据，如下所示：
name1,1,21,address1
name2,0,21,address1
name3,0,21,address1
name4,1,30,address4
name5,0,24,address5
name6,1,24,address5
name7,1,29,address7
vi 1.txt
name1,1,21,address1
name2,0,,
name3,0,,
name4,1,30,address4
name5,0,24,address5
name6,1,,
name7,1,29,address7
# 1
awk -F',' 'BEGIN{OFS=","}{if($1!=""&&$2!=""&&$3!=""&&$4!=""){a=$1;b=$2;c=$3;d=$4}else{if($1==""){$1=a;}if($2==""){$2=b;}if($3==""){$3=c;}if($4==""){$4=d}}print;}' 1.txt
# 2
awk -F"," '{if($3){b=null;for(i=3;i<=NF;i++){b=b","$i}print $0}else{$0=$0b;gsub(/,+/,",",$0);print $0}}' 1.txt
# 3
awk -F"," '{if(FNR==1){tmp3=$3;tmp4=$4;}if($3==null)$3=tmp3;if($4==null)$4=tmp4;tmp3=$3;tmp4=$4;a[FNR]=$1","$2","$3","$4; print a[FNR]}' 1.txt
# 4
awk -F"," 'BEGIN{OFS=","}{if($3){th=$3;fo=$4;print $0}else{print $1,$2,th,fo}}' 1.txt
# 5
awk -F, '{
for(i = 1; i <= NF; i++) {
a[i] = $i != "" ? $i : a[i];
}
printf("%s %s %s %s\n", a[1], a[2], a[3], a[4])}' 1.txt
# 6
awk -F, 'BEGIN{OFS=","}
{
for(i=1;i<5;++i)
if(length($i)==0)
$i = rec[i];
split($0,rec);
print
}' 1.txt
复杂实现¶
#!/bin/bash
awk '{print}' aa.txt | while read line
do
a1=`echo $line | awk -F , '{print $1}'`
a2=`echo $line | awk -F , '{print $2}'`
a3=`echo $line | awk -F , '{print $3}'`
a4=`echo $line | awk -F , '{print $4}'`
if [[ ! -z $a1 && ! -z $a2 && ! -z $a3 && ! -z $a1 ]];then
echo "$a1,$a2,$a3,$a4" >> bb.txt
b1=$a1
b2=$a2
b3=$a3
b4=$a4
else
if [ -z $a1 ];then
a1=$b1
fi
if [ -z $a2 ];then
a2=$b2
fi
if [ -z $a3 ];then
a3=$b3
fi
if [ -z $a4 ];then
a4=$b4
fi
echo "$a1,$a2,$a3,$a4" >> bb.txt
fi
done