"grep -B 1" 每个结果一行

Question

大家好，希望以下内容有意义。

我有一个文件，它包含很多垃圾以及我希望保留的一些数据，我希望提取的数据是包含“=>”的任何行加上前一行，我的问题是，我真的需要输出（包含=>的行和上一行）包含在同一行示例：

junk
input <= lala
thing1 => lululu
more junk
more junks
even more junks
input <= junk
extra junk
hello123 => lelele
input <= lololo
thing3 => kt

我目前使用的"grep -B 1 '=>' file1"是：

input <= lala
thing1 => lululu
--
extra junk
hello123 => lelele
--
input <= lololo
thing3 => kt

我真正想要的是或类似的东西。

input <= lala thing1 => lululu
--
extra junk hello123 => lelele
--
input <= lololo thing3 => kt

Answer 1

假设您使用的是 bash，您可以一行完成：

t="";for i in $(grep -B 1 '=>' tmp.txt) ; do if [[ "$i" != "--" ]] ;then t="$t $i"; else echo "$t";echo "--";t="";fi;done;echo "$t"

输出：

alfasi$ t="";for i in $(grep -B 1 '=>' tmp.txt) ; do if [[ "$i" != "--" ]] ;then t="$t $i"; else echo "$t";echo "--";t="";fi;done;echo "$t"
 input <= lala thing1 => lululu
--
 extra junk hello123 => lelele
--
 input <= lololo thing3 => kt
lgml-alfasi2:bin alfasi$

Answer 2

使用 here string 然后在 while 循环中使用 toggle 读取输入的过程并不太困难。有许多不同的方法可以做到这一点。刚想到这个。

#!/bin/bash

# read input filename as first argument
ifn=""  

# validate file is readable
[ -r "$ifn" ] || {
    printf "error: file not readable '%s'.\n" "$ifn"
    exit 1
}

declare -i a=0    # use 'a' as first/second line toggle

## for each line in results
while read -r line; do

    [ ${line:0:1} == '-' ] && { a=0; continue; }  # skip -- lines

    [ $a -eq 0 ] && sav="$line"   # if a=0, save line

    if [ $a -eq 1 ]; then         # if a=1, form output
        myline="${sav} ${line}"
        # printf -- "--\n"        # uncomment for separator
        printf "%s\n" "$myline"
    fi

    [ $a -eq 0 ] && a=1 || a=0    # toggle a=0/1

done <<< "$(grep -B 1 '=>' "$ifn")"

exit 0

输入

$ cat dat/twotoone.dat
junk
input <= lala
thing1 => lululu
more junk
more junks
even more junks
input <= junk
extra junk
hello123 => lelele
input <= lololo
thing3 => kt

输出

$ bash twointoone.sh dat/twotoone.dat
input <= lala thing1 => lululu
extra junk hello123 => lelele
input <= lololo thing3 => kt

注意： 如果你想要行与行之间的分隔符，那么在打印之前添加以下内容 $myline:

        printf -- "--\n"

带分隔符的输出

$ bash twointoone.sh dat/twotoone.dat
--
input <= lala thing1 => lululu
--
extra junk hello123 => lelele
--
input <= lololo thing3 => kt

Answer 3

这很简单 awk

awk '/=>/ {print a,[=10=]"\n--"} {a=[=10=]}' file
input <= lala thing1 => lululu
--
extra junk hello123 => lelele
--
input <= lololo thing3 => kt
--

如果您不喜欢打印最后一个--：

awk '/=>/ {print (f++?"--\n":"")a,[=11=]} {a=[=11=]}' file
input <= lala thing1 => lululu
--
extra junk hello123 => lelele
--
input <= lololo thing3 => kt

"grep -B 1" 每个结果一行

"grep -B 1" one line per result

formatting

grep