Question

我有一个csv文件，其值如下：

Wt-Do-U-Do-Wit-The-Black,black
Yay-Its-Your-Birthday-Black,black
You-Are-My-Sunshine-Happy-Birthday-Red,red
You-Are-Special-Navy-Blue,navy-blue
You-Dont-Look-A-Day-Over-Fabulous-Green,green
You-My-Friend-Are-Ridiculously-Fabulous-Happy-Birthday-Pink,pink

我想在颜色名称出现之前拆分每个字符串。例如：

str1=Wt-Do-U-Do-Wit-The   
str1=Yay-Its-Your-Birthday 
str1=You-Are-My-Sunshine-Happy-Birthday
str1=You-Are-Special   
str1=You-Dont-Look-A-Day-Over-Fabulous
str1=You-My-Friend-Are-Fabulous-Happy-Birthday

搜索我正在使用的字符串： -

if  [ "$string" == *"Black"* ] && [ "$string" == *"White"* ] ; then
              echo "It's there!"
else
    echo "SOrry"
fi

搜索正常。但是我该怎么分裂字符串呢？

我使用的另一种方式：

colour_arr[0]='Red'
colour_arr[1]='Black'
colour_arr[2]='Navy-Blue'
colour_arr[3]='White'
inarray=$(echo ${colour_arr[@]} | grep -o "$string" | wc -w)
echo "$inarray"

但这不起作用。

Answer 1

你可以使用sed;灵感来自this answer

由于您已正确解析字符串，因此我将问题简化了一点;使用此输入文件：

This is red colour
Ball is  black colour
some more words before red and more after

表示字符串的第二部分;从颜色名称开始：

sed -n -e 's/^.*\(\(red\|black\).*\)/\1/p' test

给出：

red colour
black colour
red and more after

和

sed -n -e 's/\(^.*\)\(\(red\|black\).*\)/\1/p' test

给出：

This is 
Ball is  
some more words before

我不会解释所有选项;因为他们在我提到的答案中得到了很好的解释。您可以使用：

在bash变量上使用sed

leftpart=$(sed -n -e 's/\(^.*\)\(\(red\|black\).*\)/\1/p' <<< $INPUT_STRING)

OP改变输入格式后编辑： 我的回答仍然适用;用红色替换红色。其余的都适用。

Answer 2

为您的新输入

<强>输入

$ cat f2
Wt-Do-U-Do-Wit-The-Black,black
Yay-Its-Your-Birthday-Black,black
You-Are-My-Sunshine-Happy-Birthday-Red,red  S
You-Are-Special-Navy-Blue,navy-blue
You-Dont-Look-A-Day-Over-Fabulous-Green,green
You-My-Friend-Are-Ridiculously-Fabulous-Happy-Birthday-Pink,pink

输出（使用gawk）

$ awk  'BEGIN{IGNORECASE=1;FS="[ ,]";OFS=","}match($1,$2){print "str1="substr($1,1,RSTART-2)}' f2
str1=Wt-Do-U-Do-Wit-The
str1=Yay-Its-Your-Birthday
str1=You-Are-My-Sunshine-Happy-Birthday
str1=You-Are-Special
str1=You-Dont-Look-A-Day-Over-Fabulous
str1=You-My-Friend-Are-Ridiculously-Fabulous-Happy-Birthday

对于您的旧输入

<强>输入

$ cat f
"This is red colour",red
"Ball is  black colour",black
"Tshirt is white colour",white
"Shoes are blue colour",blue
"This is green colour",green

<强>输出

$ awk  'BEGIN{FS=OFS=","}{gsub(/"/,"");match($1,$2);print "str1="substr($1,1,RSTART-1),"str2=" substr($1,RSTART) }' f
str1=This is ,str2=red colour
str1=Ball is  ,str2=black colour
str1=Tshirt is ,str2=white colour
str1=Shoes are ,str2=blue colour
str1=This is ,str2=green colour

Answer 3

OneLiner使用awk（gnu用于IGNORECASE）

awk -F ',' '# sepeartor of field is coma
 # before first line
 BEGIN{
    # define case compair behaviour (ignoring  the case)
    IGNORECASE=1
    }
 # for each line
 {
    # substitute the pattern ( minus than field 2 content, so the color, at the end) in fields 1 by "" (remove)
    sub( "-" $NF "$", "", $1)
    # print the new content of filed 1 with str1= before
    print "str1="$1
    }' YourFile

自评论代码

SELECT SUBSTR(a.nm,1,10),
  SUBSTR(lead(a.sn) over(order by a.sn ),1,5)
FROM
  (SELECT 1 AS seq,'avrajit' nm,'Roy1' sn FROM dual
  UNION ALL
  SELECT 2 AS seq,'shubho' nm,'Roy2' sn FROM dual
  UNION ALL
  SELECT 3 AS seq,'papa' nm,'Roy3' sn FROM dual
  UNION ALL
  SELECT 3 AS seq,'romi' nm,'Roy4' sn FROM dual
  )a;

Answer 4

根据您的评论，您需要首先使用颜色＆＃34;虚线＆＃34;字段，而不是第二个字段的值（逗号分隔）。

如果第一个颜色是＃34;虚线＆＃34;字段始终是最后一个字符串（短划线），您可以简单地使用
a="You-Are-My-Sunshine-Happy-Birthday-Red" ; awk -F- '{print $NF}' <<<"$a"

PS：您可以使用cut或awk隔离整行的第一个字段： awk -f, '{print $1}' <<<"$fileline"或cut -d, -f1 <<<"$fileline"

您可以将上述两项结合起来，以达到您的需求。

Answer 5

保持简单：

$< input.txt
Wt-Do-U-Do-Wit-The-Black,black
Yay-Its-Your-Birthday-Black,black
You-Are-My-Sunshine-Happy-Birthday-Red,red
You-Are-Special-Navy-Blue,navy-blue
You-Dont-Look-A-Day-Over-Fabulous-Green,green
You-My-Friend-Are-Ridiculously-Fabulous-Happy-Birthday-Pink,pink

$sed  -E 's/(-[^-]+)(,.*)/\2/g' input.txt
Wt-Do-U-Do-Wit-The,black
Yay-Its-Your-Birthday,black
You-Are-My-Sunshine-Happy-Birthday,red
You-Are-Special-Navy,navy-blue
You-Dont-Look-A-Day-Over-Fabulous,green
You-My-Friend-Are-Ridiculously-Fabulous-Happy-Birthday,pink

（注意：在我的操作系统上，OSX，sed -E用于扩展正则表达式。）

匹配shell中的单词后拆分字符串

5 个答案: