Nodes in python¶
If you want to work with Library nodes from python code, use the interactive module. The interactive module is intended for experimentation, scripting, and test and aims to make it convenient to work in IPython or similar.
The code example below demonstrates how to load the interactive module.
from sympathy.app import interactive
Interactive relies on sympathy.api in order to produce port data such as ADAF or Table, but using sympathy.api explicitly is not required.
Loading the Library¶
from sympathy.app import interactive library = interactive.load_library()
load_library may produce warnings similarly to the ones produced when
running Sympathy for Data as GUI or CLI.
When the Library has been loaded you are ready to begin loading nodes. The nodes
can be loaded by nodeid or name and if no match is found the method will also
attempt to do a fuzzy match of the provided name. If more than one node is
matched, then a
KeyError is produced listing which nodes that match the
given name. Matching the node name is often good enough, it should certainly be
unique within a library and is easy to read compared to the full nodeid.
random_table = library.node('Random Table')
Working with configurations¶
There are two different ways of configuring nodes: graphical and programmatic. When working interactively it is often a good start to use the graphical interface, the programmatic interface is more useful for automation and tests. Some nodes have very complex configurations that can be hard to get right and, for those cases, the graphical interface is recommended.
The code example below demonstrates how to launch the configuration GUI for a Random Table node. The node remembers its configuration and the changes will have effect when the node is executed and in other cases when its configuration is used.
random_table = library.node('Random Table') random_table.configure()
The code example below demonstrates how to set the column_entries attributes of a Random Table node to the value 3. This change will make the node produce 3 random columns of the default column_length which is 1000, when executed.
random_table = library.node('Random Table') random_table.parameters.attributes.column_entries.value = 3
The parameters, when accessed via attributes, have a similar interface node_context.parameters wrapped in sympathy.api.parameters or ParameterRoot, but allows you to index the elements using dot notation. This way is more convenient when used from the CLI since it allows for code completion. If you instead wish to work with the same interface as is used by nodes, then use random_table.parameters.data.
Working with nodes¶
Nodes store the changes made during configure and when the parameters are changed. They produce a list of data elements when executed and expect a list of data elements as input, this makes it possible to easily connect the data between nodes. Note that the ordering of inputs and outputs is important and should match the declaration order in the node definition.
The code example below demonstrates how to use the result produced by one node as input for another.
random_table = library.node('Random Table') rt_output = random_table.execute() item_to_list = library.node('Item to List') itl_output = item_to_list.execute(rt_output) assert(itl_output == rt_output)
The code example below demonstrates how to use the result produced by multiple nodes as input for another.
random_table0 = library.node('Random Table') rt_output0 = random_table.execute() random_table1 = library.node('Random Table') rt_output1 = random_table.execute() vjoin_table = library.node('VJoin Table') vj_output = vjoin_table.execute(rt_output0 + rt_output1)