Shell Programming and Scripting

Hello, I have a file including some lines as follows

person (1):
a
d
t

person (2):
f
h
j

person (3):
z
x
v
.
.
.


and I want to extract lines between person (n) and person (n+1) into seperate files ...

could anyone help me ?

Hi.

These kinds of tasks are what command csplit q.v. is designed for.

Why do you need awk? What have you tried in awk to solve this problem? ... cheers, drl

1- First of all I am not a student and this is not a homework
2- I run some programs that give outputs in the format as I mentioned above. The datas must be ordered to plot some graphics and figures.
3- I have tried awk commands because I have been trying to learn it and now I am familiar to awk.

4- I have tried following command types but they didnt work


awk '/^person/{next}/ ' myfile > myfile2

#awk '/^======/{next}/^(person|19)/{out=$1".txt";next}{print >out}' a > a1

If records that need to go to separate files are separated by empty line, then this should work:

Code:

awk -vRS= '{n=gensub("[():]","","g",$2);print $0 > $1"_"n}' file

---------- Post updated at 09:20 PM ---------- Previous update was at 08:44 PM ----------

bartus11 I tried to apply your command to my script but it didnt work. The exact form of my data as fllows. Which parameter must be changed in your previous command line ???

r-point 1 : 0.0000
band No. DT-energies P-energies

1 0.2420 -6.5052 -9.0927
2 0.1171 -5.4766 -7.8143
3 0.1263 -6.5876 -7.9554
4 0.1295 -6.9997 -7.8226
5 0.4497 4.8708 -7.3509

r-point 2 : 0.0833
band No. DT-energies P-energies

1 1.4278 -6.1046 -9.4933
2 0.9626 -5.5129 -8.5399
3 1.2271 -3.9610 -7.5784

r-point 3 : 0.1667
band No. DT-energies P-energies

1 6.0293 -3.4034 -11.2713
2 7.4866 -1.4132 -9.3144
3 5.4676 2.5181 -7.5542

thank you ...

Hi, rpf.

I'll let bartus11 explain his code.

I think you were close on your tries.

A very simple code:

Code:

awk '
/^person/ { i++ ; out = "person_" i txt"  }
          { print > out }
' file

this simply looks for string person and increases the counter of which the output file name will be made. It doesn't care about the string inside the ( ... ).

Another code that will extract that string and use it as part of the filename:

Code:

awk '
/^person/ { line = $0; sub(/^.*[(]/, "" ); sub(/[)].*$/, "")
            out = "person_" $0 txtxt" ; print line > out ; next
          }
          { print > out }
' file

This deletes the strings before and after the string inside the parentheses so that

Code:

person (something) xxx

will become just something, which will become part of the output filename.

I hope you will continue to learn about awk, it is a very useful tool.

However, if another command already exists to perform a task, then it is reinventing the wheel to write something specific in awk (or perl or c or ... )

Best wishes ... cheers, drl

Try this:

Code:

awk -vRS= '{f=$1;n=$2;x=$0;getline;printf "%s\n\n%s",x,$0 > f"_"n}' file