Cross Context Functions and Variables

You can access variable and call functions based on their default context (e.g. node variable/function in a node task). Additionally, you can also access the variables or call functions in select few other context.

By default, if a function is not available, node/network task calls the environment function of the same name.

For example, here the sum and array functions are environment functions, while the count is a network function. When you use Nadi IDE, it’ll show you which function is actually being called at the top of the editor.

network load_str("a->b")
network sum(array(count(), 1))

Results:

3

Besides this, you can manually call cross context variable/functions in the following ways:

Env and Network Variables/Functions

You can use env and network variables anywhere in the task system with the dot syntax.

 network load_str("a->b")
env.var = 12;
network.sth = true;

env render("this is {x}", x = network.sth)
network str(env.var)
node array(network.sth, env.var, node.NAME)

Results:

"this is true"
"12"
{
  b = [true, 12, "b"],
  a = [true, 12, "a"]
}

Similary, env and network functions can be called anywhere. These functions cannot be mutable functions (change network internally).

Taking the previous example, if we use env function count, we get an error as the function arguments are different.

network load_str("a->b")
node network.count()
network sum(array(env.count(), 1))

*Error*:

{
  b = 2,
  a = 2
}
Error in function count: Argument 1 (vars [& [bool]]) is required

Node Variables/Functions

You can use node, inputs, output and nodes keywords to access node variables and functions from different contexts. nodes is valid in all tasks, while the other 3 are only valid in a node task and refer to the current node, input nodes and output node respectively.

network load_file("./data/mississippi.net")
env count(nodes._)
node inputs.NAME

Results:

7
{
  lower-mississippi = ["ohio", "upper-mississippi", "missouri", "arkansas", "red"],
  upper-mississippi = [],
  missouri = [],
  arkansas = [],
  red = [],
  ohio = ["tenessee"],
  tenessee = []
}

You can call node functions not just for the node in the context, but also for input nodes, and output node:

Please note that the root node (outlet) of the network doesn’t have output node, so we need to skip that, which can be done through the output._? which is checking for the dummy variable _ in output, which is true if the node has an output.

network load_file("./data/mississippi.net")
node[tenessee -> "lower-mississippi"] inputs.render("{_NAME}")
node[tenessee -> "lower-mississippi"](output._?) output.render("{_NAME}")

Results:

{
  tenessee = [],
  ohio = ["tenessee"],
  lower-mississippi = ["ohio", "upper-mississippi", "missouri", "arkansas", "red"]
}
{
  tenessee = "ohio",
  ohio = "lower-mississippi"
}

You can also use nodes keyword to call the function on each node, it can be used anywhere, but is useful for env and network tasks.

network load_file("./data/mississippi.net")

env nodes.render("Node [{INDEX}] {_NAME}")

Results:

["Node [0] lower-mississippi", "Node [1] upper-mississippi", "Node [2] missouri", "Node [3] arkansas", "Node [4] red", "Node [5] ohio", "Node [6] tenessee"]