Perl collecting xml snippets from log with specific contents - perl

Question

I have a script copied from another stackoverflow question, but it seems to replace the content of the variable, could someone point me to the error? If i remove the if check for the ">OK<" it prints the whole xml to a file, if i put the if back it only prints the line containing the ">OK<", why is the $xml variable modified by the =~?
# Example usage:
# perl script.pl data.xml RootTag > RootTag.xml
use strict;
use warnings;
my $tag = pop;
while (<>){
if ( s/.*(<$tag>)/$1/ .. s/(<(\/)$tag>).*/$1/ ) {
my $xml = $_;
if ($xml =~ m/>OK</) {
print "$xml";
}
}
}
An example of a input file could be
reioirioree
brebreberbre
rbebrbebre
<test>
<id>1</id>
<status>OK</status>
</test>
bbrtbtrbt
rtbtrb
<test>
<id>2</id>
<status>KO</status>
</test>
brtoibjtrbi
bebbetreb
<test>
<id>3</id>
<status>OK</status>
</test>
dfbreberbreb
berbrebre
In this case if we user "test" as parameter, i would like following output
<test>
<id>1</id>
<status>OK</status>
</test>
<test>
<id>3</id>
<status>OK</status>
</test>

Answer 1

The objective is to capture the whole tag when it contains a specific pattern (>OK<).
Here is a step-by-step way which spells out details. I keep your program interface.
use strict;
use warnings;
my $tag = pop;
my ($inside_tag, $found, #buff);
while (<>)
{
if (s/.*(<$tag>)/$1/) {
$inside_tag = 1;
}
elsif (s|(</$tag>).*|$1|) { #/
$inside_tag = 0;
if ($found) {
print #buff, $_;
$found = 0;
}
#buff = ();
}
next unless $inside_tag;
push #buff, $_;
$found = 1 if />OK</;
}
On the opening tag we set the flag that we are inside the tag. On the closing tag we unset it, and if the marker has been $found we print the buffer (and unset marker's flag). We clear the buffer here.
Then we skip the iteration if outside of the tag. Otherwise, add the line to the buffer and test for the marker on that line.
A glitch with using the range in this problem is that we must know when we are on the closing-tag line, and would like to know the opening line as well. Then we need further tests and flip-flop isn't so clean any more. We can use the sequence number that the .. operator returns
The value returned is either the empty string for false, or a sequence number (beginning with 1) for true. The sequence number is reset for each range encountered. The final sequence number in a range has the string "E0" appended to it, which doesn't affect its numeric value, but gives you something to search for if you want to exclude the endpoint. You can exclude the beginning point by waiting for the sequence number to be greater than 1.
It would go something like
if (my $seq = /BEG/ .. /END/)
{
if ($seq == 1) { # first line of range
# ...
}
elsif ($seq =~ /EO$/) { # last line of range
# ...
}
else { ... } # inside
and I don't see that this is clearer or better than keeping the state manually.