Process multiline input with awk

greyarea · 11-29-2013, 06:11 AM

Hello, I've been trying this and I'm stuck. I've tried to search for awk multiline record processing but all the examples I've seen use a blank line as the field seperator.

Given the following input:

Time: 131127 13:41:50
Query_time: 12.249749
Time: 131127 13:41:59
Query_time: 5.156203
Time: 131127 13:42:46
Query_time: 32.596008
Time: 131127 13:44:30
Query_time: 99.876977
Time: 131127 13:45:20
...
...

I want to extract fields 2 & 3 from the first line and field 2 from the second line, and repeat until EOF.

Can somebody suggest an approach as I'm at a loss.

Regards,

grail · 11-29-2013, 07:13 AM

How about using, when /Time/ print 2 & 3 and when /Query/ print just 2

danielbmartin · 11-29-2013, 07:31 AM

With this InFile ...

Code:

Time: 131127 13:41:50
Query_time: 12.249749
Time: 131127 13:41:59
Query_time: 5.156203
Time: 131127 13:42:46
Query_time: 32.596008
Time: 131127 13:44:30
Query_time: 99.876977

... this awk ...

Code:

awk '{printf $2" "$3" "; getline; print $2}' $InFile >$OutFile

... produced this OutFile ...

Code:

131127 13:41:50 12.249749
131127 13:41:59 5.156203
131127 13:42:46 32.596008
131127 13:44:30 99.876977

Daniel B. Martin