How to get a number from a String 1

[zytalyb]

Dear all, I need your help in the following problem:
When SQL script is executed, so I get the following data as an email message. I saved this message as a text and renamed it.

SQL*Plus: Release xx.x.x.x.x Production on Mon Jun 16 10:30:00 2011

Copyright (c) 1982, 2010, Oracle. All rights reserved.
SQL> Connected.
SQL> SQL> SQL> SQL> FILE_ONE_007
SQL> 2 9876
SQL> FILE_TWO_007
SQL> 2 12345.5
SQL> FILE_THREE_007
SQL> 2 10.5
SQL> SQL> FILE_4
SQL> 2 -999998

I need a script to get the following data with the format:
Mon Jun 16 10:30:00 2011, 007,98767, 12345.5, 109.5

I have tried the following:
gawk " / Production on / {print $6, $7, $8, $9, $10 }" test.txt, and got the first part:
Mon Jun 16 10:30:00 2011

but I think that there is other way to get it also instead of that i used, also I have no idea how to get the rest.
Any idea ? thanks a lot.
with regards

 

[CaKiwi]

This may get you started. I assumed 10.7 in the file and 109.7 in the required output were supposed to be the same number

/Production on/ {a= $6 " " $7 " " $8 " " $9 " " $10 ","}
/FILE_ONE/ {a=a substr($NF,10)}
$2==2 {b[++i]= ", " $3}
END {print a b[1] b[2] b[3]}

CaKiwi

 

[CaKiwi]

I meant 10.5 and 109.5, of course

CaKiwi

 

[zytalyb]

Thanks a lot CaKiwi for your fast answer, but I'm afraid it does not working well when I tried it. I do not no why?
I have also another question, assume we have an output as a result of my problem described above, (Mon Jun 16 10:30:00 2011, 007,98767, 12345.5, 109.5) is it possible to use awk to create any txt. file that save that result? Sorry but I'm new programmer and have no idea about awk at all.

kind regards

 

[CaKiwi]

What errors or results did you get?

You can put the program into a file, test.awk say, and run it with

gawk -f test.awk test.txt

Or you can put include the program on the command line

gawk '/Production on/ {a= $6 " " $7 " " $8 " " $9 " " $10 ","}/FILE_ONE/ {a=a substr($NF,10)} $2==2 {b[++i]= ", " $3} END {print a b[1] b[2] b[3]}' test.txt

To save the output into a file, add

> test.out

to the end of the command line

CaKiwi

 

[zytalyb]

Hi CaKiwi,
I have tried it and became the error message "unexpected newline or end of string".
The good thing is the >test.out works well, I have tried it with my first code and become only the date in the output file.
The other gawk -f test.awk test.txt gives a syntax error message.

thanks in advance

 

[CaKiwi]

I copied the line for the 2nd method I gave above and pasted it into a terminal. It worked without error. What error do you get with the 1st method using test.awk?

CaKiwi

 

[zytalyb]

Hi,
The second method you used gave me this error:
gawk: '/Production
gawk: ^ invalid char ''' in expression

I have tried with "/ Production.... instead of '/Production .... then I have:
gawk: cmd. line:1: /Production on/ {a= $6
gawk: cmd. line:1: ^ unexpected newline or end of string
errcount: 1

 

[CaKiwi]

What operating system are you using? I'm guessing windows. If so, put this code into a file test.awk

/Production on/ {a= $6 " " $7 " " $8 " " $9 " " $10 ","}
/FILE_ONE/ {a=a substr($NF,10)}
$2==2 {b[++i]= ", " $3}
END {print a b[1] b[2] b[3]}

Then run by using

gawk -f test.awk test.txt

CaKiwi

 

[zytalyb]

Hi CaKiwi,
I use the win 7 OS.
I have tried your last code and it seems to be working except that the Number in the string FILE_ONE_ (007) is not displayed.
Here is the output:
Mon Jun 16 10:30:00 2011,, 9876, 12345.5, 10.5
The number 007 should be displayed between the ' '.
thanks a lot for your replay

 

[zytalyb]

Thank you very much CaKiwi for your help, it is really great , I'm very happy now.

 

[CaKiwi]

You're welcome, glad it worked for you

CaKiwi