在KSH数组长度始终返回1，为什么数组不线 [英] array length in ksh always return 1 and why array is not lines

问题描述

我要呼应过程的信息在KSH一个UID：

I need to echo information of a process for a UID in ksh:

#!/bin/ksh
read userid
arr=$(ps -elf | nawk -v pattern=${userid} '{if ($3==pattern) print}')
arrlen=${#arr[@]}
echo $arrlen
for f in "${arr[@]}"; do
  echo $f
done

改编过程是为这个UID的数组。
arrlen总是等于1。
为什么？
我的第二个问题：
我尝试回声所有元素ARR和输出

arr is an array of process for this UID. arrlen always equal 1. Why? My second question: I try to echo all elements in arr and output is

0 
S  
s157759 
22594   
1   
0  
50 
20  
?  
2:06 
/usr/lib/firefox/firefox-bin

而不是

 0 S  s157759 22594   1   0  50 20  ?  62628  ? 11:14:06 ?  2:06 /usr/lib/firefox/firefox-bin

在同一行

我想创建一个行，而不是单词的数组。

I want to create an array with lines, not words.

推荐答案

您不能创建一个数组;要创建新行分隔值的字符串。替换

You aren't creating an array; you're creating a string with newline-separated values. Replace

arr=$(ps -elf | nawk -v pattern=${userid} '{if ($3==pattern) print}')

与

arr=( $(ps -elf | nawk -v pattern=${userid} '{if ($3==pattern) print}') )

然而，这仍然留下您与该阵列将对待每行的每个字段从 PS 作为一个单独的元件的问题。更好的方法是使用来自 PS 阅读目录读内置：

However, this still leaves you with the problem that the array will treat each field of each line from ps as a separate element. A better solution is to read directory from ps using the read built-in:

ps -elf | while read -a fields; do
    if [[ ${fields[2]} = $userid ]]; then
      continue
    fi
    echo "${fields[@]}"
done

这篇关于在KSH数组长度始终返回1，为什么数组不线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！