Generate code to access dataframe

[kirangadhave]

Each node in the Trrack graph should have a pointer to a dataframe. The dataframes can be saved in Python dict with trrack node ids as keys and names formatted like df_{first 6 digits of trrack id}.

Clicking on the copy button should generate the dataframe in memory if it doesn't exist and copy a code snippet to retrieve the dataframe. E.g. IDE.get_df(<dataframe name>)

The dataframe should be usable in downstream code as usual. On kernel restart the dataframe wouldn't exist in memory, so the code using the variable should fail. So the cell with trrack interaction needs to be re-run. This is expected even in basic python code.

Update

There are two ways to generate the dataframe. Create dynamic dataframe generates a dynamic dataframe. This variable is kept in sync with the current node in the provenance graph. On restarting the kernel running the cell generates the variable.

Another way to create a dataframe is to use copy button on the provenance graph. This dataframe is associated with that specific node.
You must go to the node at least once after the kernel reset.

• add a way to trrack dataframes and recreate them while running the cell. #45
• For binned data, sometimes extra columns are added #46