Need a shell script (ksh) that uses an Awk array to process an input file

Hello Experts,

I am looking for a shell script (preferabally KSH) which processes a large input file (a log file) and provides an output file in a certain format. I want to use an Awk array to process each line of the input file and print the required output feilds to a file.

I need an array, because there are various WebContainer Types and they have different Broker transactions within them.

Thanks,

-Katewadi

####################################
Sample Input file data:

2012-03-15 05:50:09,916 [WebContainer : 2] INFO  BPServlet (AddUpdateProgramGroup) - ------------------- basicPerform starting command <AddUpdateProgramGroup> -------------------
 05:50:10,043 [WebContainer : 2] INFO  IDMODTransaction (AddUpdateProgramGroup) - ####### Broker Transaction started
2012-03-15 05:50:10,215 [WebContainer : 2] INFO  IDMODTransaction (AddUpdateProgramGroup) - ####### Broker Transaction completed in 165 milliseconds : EXSIDMOD/MFPR/NOTIFY
2012-03-15 05:50:10,219 [WebContainer : 2] INFO  IDMODTransaction (AddUpdateProgramGroup) - ####### Broker Transaction started
2012-03-15 05:50:10,390 [WebContainer : 2] INFO  IDMODTransaction (AddUpdateProgramGroup) - ####### Broker Transaction completed in 168 milliseconds : EXSIDMOD/MFPR/NOTIFY
2012-03-15 05:50:10,392 [WebContainer : 2] INFO  BPServlet (AddUpdateProgramGroup) - ------------------- basicPerform command <AddUpdateProgramGroup> done in 476 milliseconds -------------------

##################################
Format of the Output File:

Date|Start-Time|WebContainer-Number|Type|End-Time|Broker-Time|Total-Transaction-Time
2012-03-15|05:50:09,916|2|AddUpdateProgramGroup|05:50:10,392|333|476
KatewadiAsked:
Who is Participating?
 
ahoffmannConnect With a Mentor Commented:
# traditional awk:
awk '/basicPerform starting command/{d=$1;t=$2;w=substr($5,1,1);y=substr($8,2,length($8)-2);b=0;next}/Broker Transaction completed in/{b+=$(NF-3);next}/basicPerform command.* done in/{e=$2;m=$(NF-2);printf "%s|%s|%s|%s|%s|%s|%s\n",d,t,w,y,e,b,m;next}'  your-file
0
 
gheistConnect With a Mentor Commented:
GNU awk would allow to customize row and field separators so that you can have multiline records parsed. Default BWK awk in FreeBSD does not do what you like, you need to install GNU AWK via # pkg_add -r gawk
https://www.gnu.org/software/gawk/manual/html_node/Multiple-Line.html
0
 
KatewadiAuthor Commented:
Ahoffmann,

Thanks for your input...

BTW, I tried the awk command (as-is), and apparently it did not work for me.

It gave me an output like this:

ontainer|:|C|-----------------|06:15:13,220|622|in

Am I doing anything wrong? Most of the output feilds do not look as expected -- including the first and last feilds.

Could you help any furhter?

-Katewadi
0
 
ahoffmannConnect With a Mentor Commented:
my script works with the payload you posted in your question, I simply c&p this data into a file and processed it with that awk script
if this does not work with the data from your file, it's most likely formated in another way

see my script, it expects the data you want in the proper fields $1 .. $8 which must be separated by white spaces

when I see
   ontainer|:|
I assume that your data contains more fields, or are arranged in other sequence
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.