navigation.md

Pycrunch navigation

This document intends to describe how to move around the different elements and structures that Crunch exposes

We are going to use the site session variable previously described. That will be our intermediary between Python and Crunch.

Catalogs

Every collection of elements in Crunch is a Shoji Catalog. We can quickly interact with the datasets Catalog by doing:

site.datasets

Catalogs provide some methods to handle individual elements. Let's say we want to interact with the Dataset named "US Elections 2016" we could reference that particular Dataset by doing:

my_dataset = site.datsets.by('name').get("US Elections 2016").entity

Now the variable my_datasets holds a reference to the Dataset we need to interact with. Just like we interacted with the datasets Catalog, we can do it with a given Dataset variables:

my_dataset.variables

In order to visualize, for example, the data contents of our dataset we can make use of the table entity:

my_dataset.table.data

Pycrunch also allows us to interact with data using Pandas. For this we need to know the identifier of the dataset we are interacting with. We can easily get it by doing:

dataset_id = my_dataset.id

To access a Pandas Dataframe of the data:

from pycrunch import pandaslib as crunchpandas
df = crunchpandas.dataframe_from_dataset(site, dataset_id)