Question

If I perform this grep on my target file I get eg 275 as result.

But I want to learn awk so tried this in awk:

 awk 'BEGIN { count=0 } /my pattern/ { count++ } END { print count }' myfile.log

And this prints the 275 as expected.

So getting ambitious I created an awk script like this:

BEGIN { 
   print "Log File Analysis";
   message=0;
   events=0;
}

{
  /message/  {  messages++;  }
  /event/    { events++;  }
}
END { 
  print "messages:\t" messages;
  print "events:\t" events;
}

I get a syntax error,

$ awk -f test_learn.awk test_log.log
awk: test_learn.awk:16:   /message/  {  messages++;  }
awk: test_learn.awk:16:              ^ syntax error

What am I doing wrong?

I am using awk from MinGW shell on windows 7.

Answer 1

try

 awk 'BEGIN { count=0 }; /my pattern/{count++ }; END { print count }' myfile.log

OR

awk 'BEGIN { count=0};  { if ($0 ~ /my pattern/)  count++ }; END { print count };' myfile.log

Better yet, as variables are initialized as zero by default, you don’t need the BEGIN block, so

awk '/my pattern/{count++ }; END { print count };' myfile.log

You can either have a default loop applied to all lines in a file, as in 2d example with the if, or you can have multiple blocks, “filtered” by pattern, as above, and in your edited addition.

When doing one-liners have you have, some awks required the semi-colon to separate the BEGIN and END blocks from the main loop block.

Edit

Same Idea with your 2nd issue, and integrating Ed Morton’s improvments (thanks)

/message/  {  messages++  }
/event/    { events++  }

END { 
 print "Log File Analysis"
 print "messages:\t" messages
  print "events:\t" events
}

IHTH