Shell Programming and Scripting

Hi,

I had a script as below

Test_pipe.ksh

Code:

#!/usr/bin/ksh
echo `pwd`
id=`pwd | cut -d"/" -f2`
echo $id

result : /export/home
export

The above script executed sucessfully and I modified the script by placing a back slash "\" before cut command after pipe it is throwing an error as below

Code:

#!/usr/bin/ksh
echo `pwd`
id=`pwd | \ cut -d"/" -f2`
echo $id

/export/home/
Test_pipe.ksh[3]: cut: not found


My question is what is the use of back slash in pipe command.Where can I use "\" in piping.


Thanks,
Raju

Backslash is normally used to change the interpretation of the next character. In your case a space is loosing its separator capacity.
The error message means the command " cut" (with 4 characters) is not found which is correct.

One quite odd way to "fix" it would be to run:

Code:

ln -s $(whence cut) "$(dirname $(whence cut))/ cut"

One normal way of using backslash is for a continuation line, i.e. telling the shell to ignore the next new-line.

That is the same as, but much more inefficient than:

Code:

echo "$PWD"
id=${PWD#*/}
id=${id%%/*}
echo "$id"

Why did you put it there? What were you trying to accomplish?

The backslash has nothing to do with the pipe command (in your example or anywhere else).

Why do you want to use the backslash in piping? What are you trying to achieve?

If you don't know what it does, don't use it.

'\' is used for telling shell to remove the special features associated with a character.

For eg: %s/search/replace/g

Here '/' is used to separate the pattern to be searched and pattern to be replaced.
But if you have '/' in the pattern to be searched, you have to dereference it like below.

%s/\/export/\/import/g

This will replace /export with /import. So I used '\' to remove the special feature of '/'.

Hi all,

Thanks for your replies. My aim is to get the Time stamp portion from a file using grep and cut command.

My input file inputfile_jobstart_grep contains data like below in a single line

DBsetup DFModify Infilter Uvchange Uvdelete Uvfile Uvread Uvtlconvert Uvwrite cat uvwrite basic block catalog cdict claccount clear.file cname conf.file Job start time=2009-12-05 11:00:00

I want to get Job start time from the input file like 2009-12-05 11:00:00

I tried using below command

echo `cat inputfile_jobstart_grep | \ grep "^[ ]*Jobstart time" | \ cut -f2 -d\=`

but the result is
bash: grep: command not found
bash: cut: command not found


In above command i used caret "^" for matching beging of the line and "*" to match zero or more preceding. As Tene mentioned I'm using back slah "\" to remove special features associated with character like cat and equal sign character in file. But I'm getting below error.

Can some body help me to understand why I'm getting above errors and how can I resolve the above error


Thanks,
Raju

Why are you using backslashes?

Remove them and all will be well.

You are removing the special feature from the space that follows the slash so that the shell sees the command as ' cut' not 'cut', and ' grep' not 'grep'.

Hi Johnson ,

I tried the command removing backslashes

`cat inputfile_jobstart_grep | grep "^[ ]*Jobstart time" | cut -f2 -d =`


but I'm not getting any output or result from the command on the screen.


Thanks,
Raju