> in your event handler for incoming requests, do something like:
> if ($heap->{still_retrieving}) {
> push(@{$heap->{requestfifo}, [$postback_to_sender, @requestdata]);
> }
> in the retrieving routine when everything has arrived:
> foreach(@{$heap->{requestfifo}}) {
> either post($_ ...);
> or process($_);
> }
> that depends on wether it would be bad if a lot of requests block the
> rest of your program. requests wouldn't get out of order since the
> still_retrieving state change is atomic.
Actually, I've got the blocking handled... I use this idiom:
my $per = 10;
while ($per--) {
# process a queued value
last if $no_more;
}
$kernel->yield($current_state) unless $no_more;
This works great where I have, for example, 11,000 rows, and 100 at a time
is fast enough that the rest of the program doesn't halt to badly. Or 10,
or 50... $per is a tunable setting.
Question, ... never mind. I was going to ask, what's a good triggering
point to handle all of these queued requests... but that's as simple as:
if ($no_more) {
# clean state markers
$kernel->yield($handle_queued_reqs);
} else {
$kernel->yield($current_state);
}
Right?
:-)
L8r,
Rob
#!/usr/bin/perl -w
use Disclaimer qw/:standard/;
