UNIX for Dummies Questions & Answers

I am running a mailx command as follows in Linux:

Code:

mailx -s "Elapsed Time: " ora_dbas < $RUNDIR/sql_timings.out

I am trying to parse the file "sla_local_sql_timings.out" for the word Elapsed Time: and get the time from that file stored in a variable and display that variable in the subject line of my email. Is this possible at all?


Contents of sql_timing.out looks as below:

Code:

 
46 rows selected.
Elapsed: 00:10:00.10
SQL> 
SQL> spool off
 
46 rows selected.
Elapsed: 00:15:00.10
SQL> 
SQL> spool off

All i am trying to do is take the values "00:10:00.10" and "00:15:00.10" and add it to the subject line.

So the Subject line of my mail should be: Elapsed Time: 00:10:00.10 00:15:00.10


Can someone guide me on how to get this done please? thank you

Moderator's Comments:

Please use CODE tags to mark sample code and input and output samples. I guessed at where the CODE tags should go on your sample input and probably guessed incorrectly about where empty lines are supposed to appear.

You could try something like:

Code:

#!/bin/ksh
InputFile="$RUNDIR/sql_timings.out"
Subject="Elapsed Time:$(awk '
/Elapsed:/ {
        c++
        printf(" %s", $2)
}
END {   printf("%s\n", c ? "" : " None");
}' "$InputFile")"
mailx -s "$Subject" ora_dbas < "$InputFile"

I tested it using the Korn shell, but it will also work with any other POSIX conforming shell (such as bash).

If you want to try this on a Solaris/SunOS system, change awk to /usr/xpg4/bin/awk, /usr/xpg6/bin/awk, or nawk.

Thank you for giving me your time on this. I have worked on Regex commands before and have some knowledge. I am now trying to educate myself on your code works so I can further refine it.

thanks again!

This really is a pretty simple script, but if you aren't used to seeing all of the constiuent pieces combined together, it can be confusing. So, here is my script with line numbers added for discussion purposes. (This script will not work if the line numbers are included in the script.)

Code:

 1 #!/bin/ksh
 2 InputFile="$RUNDIR/sql_timings.out"
 3 Subject="Elapsed Time:$(awk '
 4 /Elapsed:/ {
 5        c++
 6        printf(" %s", $2)
 7 }
 8 END {  printf("%s\n", c ? "" : " None");
 9 }' "$InputFile")"
10 mailx -s "$Subject" ora_dbas < "$InputFile"

If you make this an executable script using the command:

Code:

chmod +x script_name

(where script_name is the name you chose to contain the original script), line 1 tells the system the name of the program to be used to interpret the remaining lines in the file. So, no matter what shell you are using when you invoke this script, /bin/ksh will be used to run this script.

Line 2 sets a variable containing the name of your input file. (It assumes that RUNDIR has been set as an exported shell variable before you call this script. I made that asuumption because your 1st message in this thread used this variable without showing how it was set.)

Lines 3-9 set the variable Subject to the string "Elapsed Time:" followed by the output of a command substitution. Command substitutions are of the form $(comand). In this case the command to be run is:

Code:

awk '
/Elapsed:/ {
       c++
       printf(" %s", $2)
}
END {  printf("%s\n", c ? "" : " None");
}' "$InputFile"

The lines in blue in this script look for lines that contain the string "Elapsed:". For every line that is found meeting that criteria, the variable c is incremented (all variables that haven't been set otherwise, have an initial value of 0 or an empty string depending on context) and then writes a space character followed by the contents of the 2nd field on that line (which is the elapsed time value you want to include in the subject line).

After every line in the file has been procesed, the commands in green associated with the special pattern END will be executed. In this case, it will print a string followed by a newline character. The string will be an empty string if c is not 0; otherwise, it will be the string " None".

And, the $InputFile on line 9 at the end of the awk script is the name of the file to be processed.

The trailing newline character written by this awk script will be removed as a side effect of command substitution.

Then line 10 sends mail to ora_dbas with the subject set to the string contained in the variable Subject and the body of the message being the contents of the file named by the variable InputFile.

This makes sense to me now. Thank you very much. You have been very helpful.