In Supply §

See primary documentation in context for method throttle

Raku highlighting

multi method throttle(Supply:D:
      Int()  $elems,
      Real() $seconds,
      Real() $delay  = 0,
      :$scheduler    = $*SCHEDULER,
      :$control,
      :$status,
      :$bleed,
      :$vent-at,
    )

Raku highlighting

multi method throttle(Supply:D:
      Int()  $elems,
      Callable:D $process,
      Real() $delay = 0,
      :$scheduler   = $*SCHEDULER,
      :$control,
      :$status,
      :$bleed,
      :$vent-at,
    )

Arguments to .throttle are defined as follows:

Argument	Meaning
$limit,	values / time or simultaneous processing
$seconds or $process	time-unit / code to process simultaneously
$delay = 0,	initial delay before starting, in seconds
:$control,	supply to emit control messages on (optional)
:$status,	supply to tap status messages from (optional)
:$bleed,	supply to bleed messages to (optional)
:$vent-at,	bleed when so many buffered (optional)
:$scheduler,	scheduler to use, default $*SCHEDULER

This method produces a Supply from a given one, but makes sure the number of messages passed through is limited.

It has two modes of operation: per time-unit or by maximum number of executions of a block of code: this is determined by the type of the second positional parameter.

The first positional parameter specifies the limit that should be applied.

If the second positional parameter is a Callable, then the limit indicates the maximum number of parallel processes executing the Callable, which is given the value that was received. The emitted values in this case will be the Promises that were obtained from starting the Callable.

If the second positional parameter is a real number, it is interpreted as the time-unit (in seconds). If you specify .1 as the value, then it makes sure you don't exceed the limit for every tenth of a second.

If the limit is exceeded, then incoming messages are buffered until there is room to pass on / execute the Callable again.

The third positional parameter is optional: it indicates the number of seconds the throttle will wait before passing on any values.

The :control named parameter optionally specifies a Supply that you can use to control the throttle while it is in operation. Messages that can be sent, are strings in the form of "key:value". Please see below for the types of messages that you can send to control the throttle.

The :status named parameter optionally specifies a Supply that will receive any status messages. If specified, it will at least send one status message after the original Supply is exhausted. See status message below.

The :bleed named parameter optionally specifies a Supply that will receive any values that were either explicitly bled (with the bleed control message), or automatically bled (if there's a vent-at active).

The :vent-at named parameter indicates the number of values that may be buffered before any additional value will be routed to the :bleed Supply. Defaults to 0 if not specified (causing no automatic bleeding to happen). Only makes sense if a :bleed Supply has also been specified.

The :scheduler named parameter indicates the scheduler to be used. Defaults to $*SCHEDULER.

In Supply§

In Supply §