Connect modular pieces of Python code ("blocks") into a processing dag pipeline. Blocks are an improvement on libraries; if you have a library, you still need to build an application. Blocks are pieces of an application, you just have to connect them.
See the examples directory in the sier2-tutorial repository for examples.
A block is a self-contained piece of code with input and output parameters.
Blocks can be connected to each other using a Dag to create
a dag of blocks.
More precisely, output parameters in one block can be connected to input parameters in another block. The connections need not be one-to-one: parameters in multiple blocks can be connected to parameters in a single block; conversely, parameters in a single block can be connected to parameters in multiple blocks.
Block parameters use param, which not only implement triggering and watching of events, but allow parameters to be named and documented.
A typical block implementation looks like this.
from sier2 import Block
class Increment(Block):
"""A block that adds one to the input value."""
in_int = param.Integer(label='The input', doc='An integer')
out_int = param.Integer(label='The output', doc='The incremented value')
def execute(self):
self.out_int = self.in_int + 1See the examples in examples (Python scripts) and examples-panel (scripts that use Panel as a UI).
To build the documentation from the repository root directory:
docs/make html