Question

我正在Linux中编写脚本，脚本的结果是一个csv文件。我想格式化该文件。我的输出文件是：

"T_Hours","T_Count","T_Hours","P_Avg_5","Diffrence_Between_P_Avg_5_and_T_Count"
"00","446","00","533","87"
"01","159","01","224","65"
"02","95","02","140","45"
"03","84","03","117","33"
"04","92","04","113","21"

我想这样输出：

"T_Hours","T_Count","T_Hours","P_Avg_5","Diffrence_Between_P_Avg_5_and_T_Count"
"00"     ,"446"    ,"00"     ,"533"    ,"87"
"01"     ,"159"    ,"01"     ,"224"    ,"65"
"02"     ,"95"     ,"02"     ,"140"    ,"45"
"03"     ,"84"     ,"03"     ,"117"    ,"33"
"04"     ,"92"     ,"04"     ,"113"    ,"21"

我尝试的方法：我尝试在，之后和之前添加空格，但这不起作用，因为数字并不总是具有相同的宽度。例如，第二列中的数字可能为"55556"，因此添加特定数量的空格将不起作用。因此，我认为解决方案是将字符,移至所有行，并将其放在第一行的,下。有什么帮助吗？

Answer 1

这是一种可移植的方式：

sed 's/,/:,/g' output.csv |
column -t -s: |
sed 's/ ,/,/g'

说明：

column -t对齐列。可以使用-s指定输入分隔符，但是输出分隔符始终是一个空格（除非您有支持{{1的 GNU 或 util-linux 之类的版本}}标志，请参见this answer。

使用-o，输出将只是

column -s, -t output.csv

要在输出中保留"T_Hours" "T_Count" "T_Hours" ... "00" "446" "00" ... "01" "159" "01" ... ...，我们必须进行预处理和后处理：

使用,在每个逗号分隔符的前面插入一个附加的分隔符sed。您选择的字符必须是输入中没有的字符。
使用:指定该column -t字符作为分隔符。 :将对齐所有列，并用空格替换所有column。
使用:删除空格（最初是我们在第一个命令中插入的sed）。

更多信息，请参见"T_Hours","T_Count","T_Hours",... "00" ,"446" ,"00" ,... "01" ,"159" ,"01" ,... ...。

Answer 2

$ column -t -s, -o, file
"T_Hours","T_Count","T_Hours","P_Avg_5","Diffrence_Between_P_Avg_5_and_T_Count"
"00"     ,"446"    ,"00"     ,"533"    ,"87"
"01"     ,"159"    ,"01"     ,"224"    ,"65"
"02"     ,"95"     ,"02"     ,"140"    ,"45"
"03"     ,"84"     ,"03"     ,"117"    ,"33"
"04"     ,"92"     ,"04"     ,"113"    ,"21"

Answer 3

这可能对您有用（GNU sed）：

sed -r '1{p;s/[^,]/ /g;h;d};G;s/^/\n/;:a;ta;s/\n([^,])(.*\n)\s/\1\n\2/;ta;s/\n,(.*\n)([^,]*,)/\2\n\1/;ta;P;d' file

这使用第一行的标题设置列宽。每个数据列都写在一个由标题创建的空格分隔的模板上。

对齐逗号分隔文件中的列

3 个答案: