Question

我正在寻找SED（首选）或其他实用程序的模式，可以处理regexp以从file.txt中删除不必要的-和,个字符。

我想删除所有看起来像这样的内容：

-----------,-------------------------------------,---------,-,---------,---------

不删除任何不接近减号的逗号，并且如果后面跟着数字或字母，则不删除任何缺点。

我正在寻找的模式是-,和,-以及--（我想是这样）

与模式不匹配的彗差和缩误应保持不变;）

输入如下：

xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxx,xxxxxxxxx,xxxx
------------------,------------------,----------------------------------------------------------------------------
-----------------------------------------------------,-,---,---------------------------,-----------------,---------
1999,2014,591025,xöxxxxxxx,~,~,xxxxöxxx,xxx,1103469,27-xxx-76,x,01-xxx-99,31-xxx-01,,273,321,321,2
1999,2014,591025,xöxxxxxxx,~,~,xxxx,xxxxx,1103759,13-xxx-61,x,01-xxx-99,30-xxx-01,,~,321,321,1
1999,2014,591025,xöxxxxxxx,~,~,xxxxxxx,xxxx,1109909,01-xxx-70,x,01-xxx-99,31-xxx-00,,~,321,321,1

所需的输出：

xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxx,xxxxxxxxx,xxxx
1999,2014,591025,xöxxxxxxx,~,~,xxxxöxxx,xxx,1103469,27-xxx-76,x,01-xxx-99,31-xxx-01,,273,321,321,2
1999,2014,591025,xöxxxxxxx,~,~,xxxx,xxxxx,1103759,13-xxx-61,x,01-xxx-99,30-xxx-01,,~,321,321,1
1999,2014,591025,xöxxxxxxx,~,~,xxxxxxx,xxxx,1109909,01-xxx-70,x,01-xxx-99,31-xxx-00,,~,321,321,1

Answer 1

您可以使用此sed：

sed -i.bak -r '/(-+,)+-+$/d' input
xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxx,xxxxxxxxx,xxxx
1999,2014,591025,xöxxxxxxx,~,~,xxxxöxxx,xxx,1103469,27-xxx-76,x,01-xxx-99,31-xxx-01,,273,321,321,2
1999,2014,591025,xöxxxxxxx,~,~,xxxx,xxxxx,1103759,13-xxx-61,x,01-xxx-99,30-xxx-01,,~,321,321,1
1999,2014,591025,xöxxxxxxx,~,~,xxxxxxx,xxxx,1109909,01-xxx-70,x,01-xxx-99,31-xxx-00,,~,321,321,1

你可以使用这个awk：

awk -F '(-+,)+-+$' '$NF' input
xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxx,xxxxxxxxx,xxxx
1999,2014,591025,xöxxxxxxx,~,~,xxxxöxxx,xxx,1103469,27-xxx-76,x,01-xxx-99,31-xxx-01,,273,321,321,2
1999,2014,591025,xöxxxxxxx,~,~,xxxx,xxxxx,1103759,13-xxx-61,x,01-xxx-99,30-xxx-01,,~,321,321,1
1999,2014,591025,xöxxxxxxx,~,~,xxxxxxx,xxxx,1109909,01-xxx-70,x,01-xxx-99,31-xxx-00,,~,321,321,1

甚至这个awk也可以：

awk -F '(-+,)+-+$' 'NF==1' input

Answer 2

怎么样

$ sed 's/--\+\|-,\+\|,-//g; /^$/d' input
xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxxxxxxxxxxxxx,xxxxxxxxxxx,xxxxxxxxx,xxxx
1999,2014,591025,xöxxxxxxx,~,~,xxxxöxxx,xxx,1103469,27-xxx-76,x,01-xxx-99,31-xxx-01,,273,321,321,2
1999,2014,591025,xöxxxxxxx,~,~,xxxx,xxxxx,1103759,13-xxx-61,x,01-xxx-99,30-xxx-01,,~,321,321,1
1999,2014,591025,xöxxxxxxx,~,~,xxxxxxx,xxxx,1109909,01-xxx-70,x,01-xxx-99,31-xxx-00,,~,321,321,1

使用-r选项

看起来很好看的sed

$ sed -r 's/--+|-,+|,-//g; /^$/d' input

Answer 3

另一个sed选项

sed '/^[-,]\+$/d'  file

另一个awk

awk '!/^[-,]+$/' file

这些将适用于您的输出

用于删除“减号”字符的模式，它们之间带有一些逗号

3 个答案: