Brother haukex has already replied, but here's my answer, which more or less does the same thing:
#!perl
use strict;
use warnings;
{
# Disable the line ending magic, and slurp the entire string into
# a scalar.
undef $/;
my $data = <DATA>;
# If we see some text between 'Test:' and 'Test2:' while looking
# at a multi-line string, display the resulting capture.
if ( $data =~ /Test:(.+)Test2:/s ) {
print "Found |$1| between titles.\n"
}
}
__DATA__
Test:
Blah blah blah 1
Blah blah blah 2
Blah blah blah 5
Blah blah blah 9
Test2:
What is this for?
How is this happning?
Why am I here?
Hello3:
What
The
$/ variable is the one that tells Perl what the 'end of line' character is. When I
undef that variable, the whole file gets treated as a single line from an input point of view when I read the text into that variable.
The 's' option defines the regexp as a multi-line regexp, telling it to ignore the carriage return (\r) and line feed (\n) characters. When this script is run, I get
Found |
Blah blah blah 1
Blah blah blah 2
Blah blah blah 5
Blah blah blah 9
| between titles.
That's one way of parsing the file -- another way would be to
- Read lines until you see the start of the capture (Test:, in this case), then start capturing;
- Did we see the end of the capture (Test2:, in this case)? If not, capture the line and repeat; otherwise, stop.
You could also just capture each block of text into an array, store that array as a hash value, using the label (like Test) as the hash key.
Later, just go and get the entry for whichever key you want.
Alex / talexb / Toronto
Thanks PJ. We owe you so much. Groklaw -- RIP -- 2003 to 2013.