view release on metacpan or  search on metacpan

README  view on Meta::CPAN

      Lvalue subroutine that can be used to set whether the module "croak"
      in Carps on failure.

      Default value is 0.

    renew_on_timeout

          $dev->renew_on_timeout;
       

      Closes HIDAPI handle and opens a new one transparently at read_data
      timeout and retries reading. When read_data returns successfully the
      first time, renew_on_timeout is reset and timeout is set to undef,
      but can be manually adjusted.

      For reasons unknown to me, Valve's Steam controller needs a couple of
      hid_open calls before hid_read manages to read data. None of the
      prior hid_open calls fail, they just block indefinitely. For devices
      that ought to report periodically what they're up to, set the timeout
      in new to a sensible value and call renew_on_timeout on the handle.
      The following hid_read will then be retried till data can be read.

lib/Device/HID.pm  view on Meta::CPAN

sub autodie : lvalue {
    my $self = shift;
    return $self->{autodie};
}

=item renew_on_timeout
 
    $dev->renew_on_timeout;
 
Closes HIDAPI handle and opens a new one transparently at C<read_data> timeout and retries reading. When C<read_data> returns successfully the first time, C<renew_on_timeout> is reset and timeout is set to C<undef>, but can be manually adjusted.

For reasons unknown to me, Valve's Steam controller needs a couple of C<hid_open> calls before C<hid_read> manages to read data. None of the prior C<hid_open> calls fail, they just block indefinitely. For devices that ought to report periodically wha...

=cut

sub renew_on_timeout  {
    my $self = shift;
    $self->{renew} = 1;
    defined $self->{timeout} or croak "Thou shalt not call renew_on_timeout when timeout is not in sight.";