Re: sampling with perl script

Please read How do I post a question effectively? It sounds like you have the following:

You have a 'Bayes' program running, writing output to a file
After every N lines of output, you want to perform some processing
This processing is to be done via a perl program

Is this right? If so, the easiest thing to do would be something like

open BAYES, "tail -f bayes.out |" || die "could not open: $!";
while (<BAYES>) {
# read until you have enough to process
}
[download]

If this is not right, you will need to provide a better explanation...

Comment on Re: sampling with perl script Download Code

Replies are listed 'Best First'.
Re^2: sampling with perl script by Fletch (Bishop) on Oct 16, 2008 at 20:09 UTC
Of course `open BAYES, "tail -f bayes.out \|" or die "could not open: $!";` would work even better as it lacks the subtle precedence problem that makes the error checking get optimized away . . . (Or one could be fashionable and go for a 3-argument, lexical filehandle form `open my $bayes, "-\|", qw( tail -f bayes.out ) or die "Couldn't open pipe: $!";` if you're into that kind of thing and have a recent enough perl.) The cake is a lie. The cake is a lie. The cake is a lie.	[reply] [d/l] [select]

Replies are listed 'Best First'.

Re^2: sampling with perl script
by Fletch (Bishop) on Oct 16, 2008 at 20:09 UTC

Of course open BAYES, "tail -f bayes.out |" or die "could not open: $!"; would work even better as it lacks the subtle precedence problem that makes the error checking get optimized away . . .

(Or one could be fashionable and go for a 3-argument, lexical filehandle form open my $bayes, "-|", qw( tail -f bayes.out ) or die "Couldn't open pipe: $!"; if you're into that kind of thing and have a recent enough perl.)

The cake is a lie.
The cake is a lie.
The cake is a lie.

[reply]
[d/l]
[select]