print matching lines

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: print matching lines by davidrw (Prior) on Jul 10, 2005 at 19:20 UTC
You said each entity in the file is composed of 4 lines but the second one in your example only has 3 lines.. Do you have code/thoughts you've tried so far? one method would be to read 4 lines at a time. if the 4th line matches, print the first line. Another method would be to set `$/` to '# input' and then regex the resulting 'lines' (or grep all of them).	[reply] [d/l]
Re^2: print matching lines by Anonymous Monk on Jul 10, 2005 at 19:23 UTC
sorry yes, # input 2 xxx yyyy 1 so with $/ set to # input it would be possible to do a pattern search and print the match?	[reply]
Re^3: print matching lines by davidrw (Prior) on Jul 10, 2005 at 20:12 UTC
yes, that's why i suggested it .. I would (you ignored my questioned about providing code, so i'll continue just guiding with pesudo code until you have something to work from) set `$/`, then loop through the 'lines' (which will contain multiple `\n` newlines). During this loop, split on newline to get your array of 4 lines. Note the first one will be just a number (the remainder of the '# input 3' line) and the last one will be '# input ' (`$/`).	[reply] [d/l] [select]
Re^3: print matching lines by Anonymous Monk on Jul 10, 2005 at 19:24 UTC
`# input 2 xxx yyy 1` [download]	[reply] [d/l]
Re^4: print matching lines by Anonymous Monk on Jul 10, 2005 at 19:39 UTC
Re: print matching lines by Forsaken (Friar) on Jul 10, 2005 at 19:38 UTC
Sorry, but your question doesn't make sense. Should it print `# input 1` if any of the other inputs have the same string on line 3? Because in that case you might switch the whole thing around and just look for unique occurences of line 3 and print everything else. Please provide some more information on what you want exactly as well as some code you've already tried. update to prevent the question thread going 16 levels deep: how about something like: (untested code alert) `use strict; use warnings; my $FH = insert your filehandle stuff here; my $linecounter = 0; my $headerline; while(my $line = <$FH>) { chomp($line); #not sure if needed... unless(my $modulo = ($linecounter % 4)) { $headerline = $line; } elsif(($modulo == 3) && ($line eq 'ord')) { print "$headerline\n"; } $linecounter += 1; }` [download] Remember rule one...	[reply] [d/l] [select]
Re^2: print matching lines by Anonymous Monk on Jul 10, 2005 at 19:41 UTC
if the 4ith line matches ord it should print the # input line	[reply]
Re^3: print matching lines by Forsaken (Friar) on Jul 10, 2005 at 19:47 UTC
so are you looking for a match between # input 1 and another one? or between any 2 inputs? and, another important question, are all the inputs numerically ordered like that? Remember rule one...	[reply]
Re^4: print matching lines by Anonymous Monk on Jul 10, 2005 at 20:06 UTC
Re^5: print matching lines by ysth (Canon) on Jul 11, 2005 at 01:08 UTC
Re^5: print matching lines by CountZero (Bishop) on Jul 10, 2005 at 22:44 UTC
Re^4: print matching lines by Anonymous Monk on Jul 10, 2005 at 19:49 UTC
Re^5: print matching lines by Forsaken (Friar) on Jul 10, 2005 at 19:52 UTC
Re: print matching lines by Adrade (Pilgrim) on Jul 11, 2005 at 03:13 UTC
`while(<DATA>){$q[($.-1)%4]=$_;print $q[0] if (($.-1)%4==3)&&(m/ord/s)} __DATA__ # input 1 xxx yyyy ord # input 2 xxx yyy 1 # input 3 xxx yyy ord` [download] -- By a scallop's forelocks!	[reply] [d/l]
Re: print matching lines by anonymized user 468275 (Curate) on Jul 11, 2005 at 11:51 UTC
...or a simple alternative way... `my @block; while ( <$fh> ) { if ( /^\#/ ) { if ( $block[3] =~ /^ord\s$/ ) { print $block[0]; } $#block = -1; } push @block, $_; }` [download] One world, one people*	[reply] [d/l]