Read a variable with multiple occurrences from a file using AWK - bash

Question

I want to get the value (usually a string) of var1 in a file my_file.dat and save this value to x.
I managed to do this using the following command:
x = `awk '$1 == "var1" {print $2}' my_file.dat`
It now turns out that there can be several occurrences of var1 in my_file.dat, e.g.:
Series1
var1 = temp/data/
Series2
var1 = lost/oldfiles/
My question is then how can I get only the value of the 'var1' which is located right after the line 'Series1', such that 'x' returns 'temp/data/'?

Answer 1

Given the sample you posted all you need is:
x=$(awk 'prev=="Series1" {print $NF} {prev=$0}' file)
but more robustly:
x=$(awk '
{ name=value=$0; sub(/[[:space:]]*=.*/,"",name); sub(/[^=]+=[[:space]]*/,"",value) }
(prev=="Series1") && (name=="var1") { print value }
{ prev=$0 }
' file)

Answer 2

What about a two state machine to solve the problem:
#!/bin/awk
BEGIN {
state = 0;
}
{
if( state == 0 )
{
if( index( $1, "Series" nseries ) )
{
state = 1
}
}
else
{
if( index( $1, "Series" ) > 0 )
{
exit
}
if( index( $1, "var1" ) > 0 )
{
idx = index( $0, "=" )
str = substr( $0, idx + 1 )
gsub(/^[ \t]+/, "", str )
print str
exit
}
}
}
# eof #
Test file:
Series1
var1 = temp/data/
Series2
var1 = lost/oldfiles/
Series3
var1 = foo/bar/
Series4
var1 = alpha/betta/
Series5
var1 = /foo/this=bad
Series6
var1 = /foo/Series/
Reading var1 from Series1:
x=$(awk -v nseries=1 -f solution.awk -- ./my_file.dat)
echo $x
temp/data/
Reading var1 from Series5:
x=$(awk -v nseries=5 -f solution.awk -- ./my_file.dat)
echo $x
/foo/this=bad
Reading var1 from Series6:
x=$(awk -v nseries=6 -f solution.awk -- ./my_file.dat)
echo $x
/foo/Series/
Hope it Helps!