nanonext (version 0.1.0)

recv_aio: Receive Async

Description

Receive serialised data asynchronously over a Socket (with ability to set a timeout). For sending and receiving within R.

Usage

recv_aio(socket, n = 1L, timeout, keep.raw = TRUE)

Arguments

socket

a Socket.

n

[default 1L] number of messages to receive asynchronously.

timeout

in ms. If unspecified, a socket-specific default timeout will be used.

keep.raw

[default TRUE] TRUE to keep the received raw vector (useful for verification e.g. via hashing). If FALSE, will return the unserialised object only.

Value

Named list of 2 elements: 'raw' containing a list of received raw vectors and 'data' containing a list of unserialised R objects, or else a list of unserialised R objects if keep.raw is set to FALSE.

Note: a list of lists is always returned even when n = 1. To access the first raw element, for example, use $raw[[1]] and the first data element use $data[[1]].

Details

Async recv will block while awaiting all 'n' messages to arrive. Set a timeout to ensure that the function returns under all conditions.

In case of an error in unserialisation (e.g. the data was not sent serialised), the function will still return a list of received raw vectors to allow the data to be recovered.

Examples

# NOT RUN {
s1 <- socket("pair", listen = "inproc://nanonext")
s2 <- socket("pair", dial = "inproc://nanonext")

send_aio(s1, data.frame(a = 1, b = 2), data.frame(c = 3, d = 4), timeout = 100)
res <- recv_aio(s2, 2L, timeout = 100)
res

send_aio(s1, data.frame(a = 1, b = 2), data.frame(c = 3, d = 4), timeout = 100)
recv_aio(s2, 2L, timeout = 100, keep.raw = FALSE)

close(s1)
close(s2)

# }