Shell Programming and Scripting

Guys,
I have a script which hits the database and pulls the information that I need into files. Now I want to format these files to make them easy to read.
The sample format of the file will be like....

<Start_of_File>

Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

A1 as_p:xxxxxxxxx as_p:yyyyyyyyy as_p:zzzzzzzzz
A2 as_p:dskfjsjfsd as_p:sdfafjlaflkd


Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

B1 as_p:xxxxxxxxx as_p:yyyyyyyyy as_p:zzzzzzzzz
B2 as_p:dskfjsjfsd as_p:sdfafjlaflkd
B3 as_p:dsoiugtoi as_p:woiuiosgsd as_pwuewuu

Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

C1 as_p:xxxxxxxxx as_p:yyyyyyyyy as_p:zzzzzzzzz
C2 as_p:dskfjsjfsd as_p:sdfafjlaflkd
C3
C4
C5
.
.
.
<End_of_file>


The number of fields in the above records which have "as_p" in them are variable (actual content apart from Hearders).

I would like the file to be formated as....

<Start_of_File>

Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

A1
as_p:xxxxxxxxx
as_p:yyyyyyyyy
as_p:zzzzzzzzz
A2
as_p:dskfjsjfsd
as_p:sdfafjlaflkd


Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

B1
as_p:xxxxxxxxx
as_p:yyyyyyyyy
as_p:zzzzzzzzz
B2
as_p:dskfjsjfsd
as_p:sdfafjlaflkd
B3
as_p:dsoiugtoi
as_p:woiuiosgsd
as_pwuewuu

Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

C1
as_p:xxxxxxxxx
as_p:yyyyyyyyy
as_p:zzzzzzzzz
C2
as_p:dskfjsjfsd
as_p:sdfafjlaflkd
C3
C4
C5
.
.
.
<End_of_file>


Basically I want to search the records which have 'as_p' in them and transpose all the fields in that record in one column. The rest of the records which don't have 'as_p' in them should be left untouched.

Thanks in advance.

Maybe...
awk '/as_p/ { gsub(" ", "\n")}{print} '

It says syntax error.
I think with gsub you are trying to search and replace globally all matches of " " with new line. That too for records only with 'as_p'.

But since the records will be tab delimited....does it need to be like

Code:

gsub("\t", "\n")

So, it would be something like..

Code:

awk '/as_p/ {gsub("\t", "\n")}; print' FileName

Its stil throwing error.

And since the last field will be delimited by "\n", the gsub might not work for the last field.

If I use the following code....its stripping out the headers....and prints only the records with 'as_p'.

Code:

awk '/as_p/ {for (field=1; field<=NF; ++field) {printf ("\n" $field);}} ' FileName

Do you think the below logic would work?

awk ' { if 'as_p' is there record print
{for (field=1; field<=NF; ++field) {printf ("\n" $field);}}
else
print line
} ' FileName


Sorry I'm new to awk.

try using slashes eg gsub(/your pattern/, "" )

Works for me...

Code:

$ cat data
Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

A1 as_p:xxxxxxxxx as_p:yyyyyyyyy as_p:zzzzzzzzz
A2 as_p:dskfjsjfsd as_p:sdfafjlaflkd


Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

B1 as_p:xxxxxxxxx as_p:yyyyyyyyy as_p:zzzzzzzzz
B2 as_p:dskfjsjfsd as_p:sdfafjlaflkd
B3 as_p:dsoiugtoi as_p:woiuiosgsd as_pwuewuu

Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

C1 as_p:xxxxxxxxx as_p:yyyyyyyyy as_p:zzzzzzzzz
C2 as_p:dskfjsjfsd as_p:sdfafjlaflkd
C3
$
$
$
$
$ awk '/as_p/ { gsub(" ", "\n")}{print} ' data
Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

A1
as_p:xxxxxxxxx
as_p:yyyyyyyyy
as_p:zzzzzzzzz

A2
as_p:dskfjsjfsd
as_p:sdfafjlaflkd


Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

B1
as_p:xxxxxxxxx
as_p:yyyyyyyyy
as_p:zzzzzzzzz

B2
as_p:dskfjsjfsd
as_p:sdfafjlaflkd
B3
as_p:dsoiugtoi
as_p:woiuiosgsd
as_pwuewuu

Header1 .....xsdfsfa...adfa......
Header2 ....afefas .aefaefsdf...
-------------------------------------------------

C1
as_p:xxxxxxxxx
as_p:yyyyyyyyy
as_p:zzzzzzzzz

C2
as_p:dskfjsjfsd
as_p:sdfafjlaflkd
C3
$

The above code works but if the data is tab delimited then it needs to be changed so very slightly.

Code:

nawk '{gsub("as_p","\nas_p")}{print}' file

True enough, but still, the code I posted shouldn't result in a syntax error.