Jim Gibson <jimsgib...@gmail.com> writes: linux expert wrote: >>> Try adding the following line after your 'use warnings' line: >>> $|++; >>> >>> That will disable output buffering on the currently-selected >>> filehandle (STDOUT be default).
Harry replied: >>Doesn't appear to make any difference whatever. Jim G. added: > You need to select the logger file before executing $|++: > > select LOGNAME; > $|++; > select STDOUT; > > See the advice given by 'perldoc -q flush' "How do I flush/unbuffer an > output filehandle? Thanks, that works fine. The page offered by perldoc -q flush mentions a write up about buffering ( http://perl.plover.com/FAQs/Buffering.html ), that I also went through. I'm sorry to say that it was mostly like trying to read ancient sanskrit, since its written in OOPese. Something I have yet to be able to look at and have any clue at all about what it means or is doing. But I saw mention of: If you're using the FileHandle or IO modules, there's a simpler interface: $filehandle->flush(); # Flush the buffer It does just the same as the code above. So looked at IO::Handle for quite a while and finally noticed near the bottom some thing I could almost understand. I experimented quite a bit and ended up with either your suggestion or or using IO::Handle like so: open my( $io_fh ), ">", "$logdir/$logname" or die "Can't open $logdir/$logname: $!"; $io_fh->autoflush(1); I wondered if there would be any difference in performance between using your suggestion: > select LOGNAME; > $|++; > select STDOUT; or calling in the module,IO::Handle like in the full script below? In the course of a days run, there would be a whole lot of calls to print immediately instead of buffer what appears to be 4096 bytes before writing to log. If I left any flushing out altogether, I guess it would mean there would be (($TIMERUN * $lines_read_per_second ) / 4096) less calls to write to log. So considering I'm asking perl to work a lot harder; would either of the techniques be better performance wise, ? I can't really think of a way to quickly tell. ------- 8< snip ---------- 8< snip ---------- 8<snip ------- #!/usr/local/bin/perl use strict; use warnings; use IO::Handle; my $logdir ='/var/adm/log'; my $logname = 'pipe1.log'; my $ucsb = qr/ucsb/; my $temp_thresh =qr/Temperature above Threshold/; my $hddtemp = qr/hddtemp/; open my( $io_fh ), ">", "$logdir/$logname" or die "Can't open $logdir/$logname: $!"; $io_fh->autoflush(1); while(<>){ if (/$ucsb|$temp_thresh|$hddtemp/){ print $io_fh $_; print $_; } } -- To unsubscribe, e-mail: beginners-unsubscr...@perl.org For additional commands, e-mail: beginners-h...@perl.org http://learn.perl.org/
