将16行合并为一行 [英] Merge 16 lines into one line

问题描述

文件包含表示十六进制值的ASCII字符. (每行以，"结尾)

A file contains ASCII characters to represent hex value. (each line ends in ", ")

cat temp.txt
    0x6A, 
    0xF2, 
    0x44,
    .....
    0xF8, 
    0x1A,

我尝试将每16个字/行合并成这样的一行

I try to combines each 16 words/lines into one line like this

cat hex_result.txt 
0x6A, 0xF2, 0x44, 0xF8, 0x45, 0x41, 0x88, 0xD1,0x4E, 0x8B, 0xA3, 0xB1, 0x8C, 0xE0, 0x37, 0x2D, 
.... 
0xE2, 0x1C, 0x06, 0x8A, 0x75, 0x2B, 0xBC, 0x3C, 0xC5, 0x08, 0xB7, 0x4E, 0xB0, 0xE4, 0xF8, 0x1A,

是否有任何bash命令来完成它?

Is any bash commands to accomplish it ?

推荐答案

合并特定数量的行

基本，有很多命令:

merging specific number of lines

Basicaly, there are many commands:

pr -at16 <file

尝试:

pr -a -t -16 < <(seq 1 42)
1   2   3   4   5   6   7   8   9   10  11  12  13  14  15  16
17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32
33  34  35  36  37  38  39  40  41  42

xargs-从标准输入生成并执行命令行

xargs - build and execute command lines from standard input

...并执行命令(默认为/bin/echo)...

... and executes the command (default is /bin/echo) ...

xargs -n 16 <file

尝试:

xargs -n 16 < <(seq 1 42)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 34 35 36 37 38 39 40 41 42

paste-合并文件行

paste - merge lines of files

printf -v pasteargs %*s 16
paste -d\  ${pasteargs// /- } < <(seq 1 42)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 34 35 36 37 38 39 40 41 42

sed-用于过滤和转换文本的流编辑器

sed - stream editor for filtering and transforming text

printf -v sedstr 'N;s/\\n/ /;%.0s' {2..16};
sed -e "$sedcmd" < <(seq 1 42)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 34 35 36 37 38 39 40 41 42

awk-模式扫描和处理语言

awk - pattern scanning and processing language

awk 'NR%16{printf "%s ",$0;next;}1'  < <(seq 1 42)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 34 35 36 37 38 39 40 41 42 42

但是，您可以使用 pure bash :

group=()
while read -r line;do
    group+=("$line")
    ((${#group[@]}>15))&&{
        echo ${group[@]};
        group=()
    }
  done < <(seq 1 42) ; echo ${group[@]}
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 34 35 36 37 38 39 40 41 42

或作为功能:

lgrp () { 
    local group=() line
    while read -r line; do
        group+=("$line")
        ((${#group[@]}>=$1)) && { 
            echo ${group[@]}
            group=()
        }
    done
    [ "$group" ] && echo ${group[@]}
}

或

lgrp () { local g=() l;while read -r l;do g+=("$l");((${#g[@]}>=$1))&&{
          echo ${g[@]};g=();};done;[ "$g" ] && echo ${g[@]};}

然后

lgrp 16 < <(seq 1 42)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 34 35 36 37 38 39 40 41 42

好吧， 超过3条命令，让我们稍作练习吧！

我该怎么做:

Ok there are more than 3 commands, let do a little bench!

How I do it:

lgrp () { local g=() l;while read -r l;do g+=("$l");((${#g[@]}>=$1))&&{
          echo ${g[@]};g=();};done;[ "$g" ] && echo ${g[@]};}
export -f lgrp
printf -v sedcmd '%*s' 15
sedcmd=${sedcmd// /N;s/\\n/ /;}
export sedcmd
{ 
    printf "%s\n" cmd cnt$'\n'time{,,,,}

    for cmd in 'paste -d " " -{,,,}{,,,}' 'pr -at16' 'sed -e "$sedcmd"' \
               $'awk \47NR%16{printf "%s ",$0;next;}1;END{print}\47' \
               'lgrp 16' 'xargs -n 16'
    do
        echo ${cmd%% *}
        for length in 100 1000 10000 100000 1000000
        do
            echo "$(bash -c "TIMEFORMAT=%R;
                             time $cmd < <(seq 1 $length) | wc -l" 2>&1)";
        done
    done
} | pr -at11

在我的计算机上生成:

cmd   cnt   time  cnt   time  cnt   time  cnt   time  cnt   time
paste 7     0.002 63    0.003 625   0.002 6250  0.016 62500 0.052
pr    7     0.002 63    0.002 625   0.003 6250  0.017 62500 0.125
sed   7     0.002 63    0.010 625   0.006 6250  0.059 62500 0.457
awk   7     0.003 63    0.003 626   0.008 6251  0.049 62501 0.501
lgrp  7     0.004 63    0.027 625   0.256 6250  2.701 62500 26.84
xargs 7     0.012 63    0.049 625   0.387 6250  4.209 62500 41.75

我的树莓派上有一条长凳:

There is same bench on my raspberry pi:

cmd   cnt   time  cnt   time  cnt   time  cnt   time  cnt   time
paste 7     0.104 63    0.101 625   0.130 6250  0.297 62500 2.241
pr    7     0.107 63    0.112 625   0.166 6250  0.821 62500 7.076
sed   7     0.137 63    0.145 625   0.415 6250  2.868 62500 28.06
awk   7     0.198 63    0.179 626   0.620 6251  5.426 62501 51.20
lgrp  7     0.231 63    2.715 625   15.76 6250  150.3 62500 1544.
xargs 7     0.270 63    1.845 625   16.37 6250  165.3 62500 1648.

希望所有行数都相同，然后paste显然更快，然后是pr. 纯 bash 函数是不比xargs慢(我很惊讶！).

Hopefully all line count are same, then paste are clearly the quicker, followed by pr. Pure bash function is not slower than xargs (I'm surprised!).

这篇关于将16行合并为一行的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！