Is it possible to put an optional input to a graph with a default value

Yes, it is possible to define optional inputs with default values in Dagster. When you define a graph, you can specify inputs with default values using Python's standard default argument syntax in the function definition. Here's an example based on the documentation:

python from dagster import graph, op  @op def my_op(x: int):     return x + 1  @graph def my_graph(x: int = 10):     return my_op(x)

In this example, the

my_graph

graph has an input

x

with a default value of

10

. If no value is provided for

x

when the graph is invoked, it will use the default value of

10

. This allows you to have optional inputs for your graphs in Dagster.

@Scout using the "ins={In(...)}" syntax

In Dagster, when you're working with assets or ops that require inputs, you can use the

ins

parameter to specify inputs. However, the

ins

parameter itself does not directly support default values as you would define them in a Python function signature. Instead, the

ins

parameter is used to define the metadata and dependencies for the inputs of an asset or op. If you want to provide a default value for an input when using the

ins

parameter, you would typically handle this within the body of the op or function that is being called. Here's an example of how you might do this:

python from dagster import op, In, graph  @op(ins={"input1": In()}) def my_op(context, input1=None):     if input1 is None:         input1 = "default_value"  # Set the default value here     # Your op logic here  @graph def my_graph():     my_op()  # Called without providing input1, so it will use the default value

In this example,

my_op

is defined with an input

input1

that uses the

In

class to specify its metadata. The default value for

input1

is handled within the body of

my_op

by checking if

input1

is

None

and setting it to a default value if necessary. This approach allows you to define a default behavior for when an input is not provided, but it's important to note that the

ins

parameter itself is not where you define default values. Instead, it's used to specify the input definitions for assets or ops, and the defaulting logic is handled within the op or function.

@Scout Is it possible to put an optional input to a graph with a default value using the "ins={In(...)}" syntax

As of my knowledge cutoff in early 2023, Dagster does not support directly assigning default values to inputs using the

ins={In(...)}

syntax when defining a graph. The

In

class is used to specify metadata about an input, such as its description, dagster_type, and whether it is required, but not to assign a default value. If you want to provide a default value for an input in a graph, you would typically handle this within the body of the ops that make up the graph. Each op can have logic to provide a default value if none is supplied. Here's an example of how you might handle this:

python from dagster import op, In, graph  @op def process_data(data: int):     # Op logic here     return data  @graph def my_graph(input1: int = 10):     process_data(input1)  # When invoking the graph, if you don't provide input1, it will default to 10 result = my_graph.execute_in_process()

In this example,

my_graph

has a parameter

input1

with a default value of

10

. If

input1

is not provided when the graph is invoked, it will use the default value. If you need to use the

ins={In(...)}

syntax for some reason, such as when you are defining a graph without using the decorator syntax, you would still need to handle default values within the ops or by using a config schema that provides defaults. The

In

class itself does not provide a mechanism for default values. Please note that Dagster's features and capabilities are subject to change, and it's always a good idea to consult the latest Dagster documentation or release notes for the most current information.