Log Parsing

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

Hi guys, Was wondering if I could get help on the code below. I would like to go through all the *.log files in a directory and perform the following code on them. Instead of just print $3 though, I would like to print this to a new file (a new file for each log file). So i need to create lots of new files (one for each log file) and then but the $3 into it.... Thanks in advance, Matt.

@files = glob "*.log";
foreach $file (@files) {
    next unless $file =~ /planetlab/;
    print "begin\n";
    local($\, $,) = ("\n", " "); 
    while(<>) {
        if(/([\d:.]+).*?\bReceived.*?\bBT_PIECE.*?\bdata.*?\bPeer: (\w
+):\s+(\d+\.\d+\.\d+\.\d+):\s*(\d+)/) {
            print $3;
        }
    }
    print "end";
    print "$file\n";
}
[download]

Comment on Log Parsing Download Code

Replies are listed 'Best First'.
Re: Log Parsing by Corion (Patriarch) on Sep 07, 2007 at 12:16 UTC
So you will need to open a new output file whenever you are processing a new input file. You will then need to change your print statement to output to the handle of the output file. That's it. Where exactly do you have problems?	[reply]
Re^2: Log Parsing by mattwortho (Acolyte) on Sep 07, 2007 at 13:02 UTC
Many thanks in advance... `@files = glob ".log"; $count = 1; foreach $file (@files) { next unless $file =~ /planetlab/; open FILE, >"/home/imperial_crossbt/results/logging/".$count.".log +"; print FILE "begin\n"; local($\, $,) = ("\n", " "); while(<>) { if(/([\d:.]+).?\bReceived.?\bBT_PIECE.?\bdata.?\bPeer: (\w +):\s+(\d+\.\d+\.\d+\.\d+):\s(\d+)/) { print FILE $3; } } print FILE "end"; print "$file\n"; close FILE; $count++; }` [download]	[reply] [d/l]
Re^2: Log Parsing by mattwortho (Acolyte) on Sep 07, 2007 at 12:28 UTC
Sorry, I'm new to file handling in perl and wasn't sure how to create files and write to them. Not sure either how to make the new files' filenames because they'd have to be say the newfile + a number that is incrementing for example (they can't all be the same). Thanks in advance for your help. Matt.	[reply]
Re^3: Log Parsing by mattwortho (Acolyte) on Sep 07, 2007 at 12:37 UTC
Could I for example: `open filehandle, >"/home/mj/newlogs"+a number+".log" print filehandle, $3 close filehandle` [download] Not sure how to get the number in there...	[reply] [d/l]
Re^4: Log Parsing by mattwortho (Acolyte) on Sep 07, 2007 at 12:40 UTC
Re^5: Log Parsing by Corion (Patriarch) on Sep 07, 2007 at 12:47 UTC
Re^2: Log Parsing by mattwortho (Acolyte) on Sep 07, 2007 at 12:42 UTC
`$count = 0; open filehandle, >"/home/mj/newlogs"+$count+".log" print filehandle, $3 close filehandle $count++;` [download] Something like that?? Thanks in advance.	[reply] [d/l]
Re^3: Log Parsing by moritz (Cardinal) on Sep 07, 2007 at 12:56 UTC
No. `open filehandle, ">", "/home/mj/newlogs$count.log"; # or open filehandle, '>', '/home/mj/newlogs/ . $count . '.log';` [download] Additionally you should check if open succeeded: `open filehandle, ">", "/home/mj/newlogs$count.log" or die "Can't write to file: $!";` (Update: fixed typo spotted by Corion++) Perl 6 in German	[reply] [d/l] [select]
Re^4: Log Parsing by mattwortho (Acolyte) on Sep 07, 2007 at 13:19 UTC
Re^3: Log Parsing by FunkyMonk (Bishop) on Sep 07, 2007 at 16:18 UTC
Just for completeness, this doesn't do what you think it does: `print filehandle, $3` [download] To print to a filehandle use `print FILEHANDLE VALUES`. Notice the lack of a comma between `FILEHANDLE` and `VALUES`. The correct version of your print would be: `print filehandle $3` [download]	[reply] [d/l] [select]