Building Blocks for development data work
bblocks is a growing collection of Python packages designed to simplify the everyday work of analysts and
researchers in the international development sector. Built with usability and modularity in mind, bblocks
provides reliable, reusable components that help standardize and streamline data workflows—from data cleaning
and transformation to country concordance, dataset importation, and integration with ETL pipelines such as Data Commons.
Whether you're wrangling data for a quick analysis or developing a robust, production-ready
pipeline, bblocks offers practical, tested tools that are easy to plug into your workflow and scale with your needs.
Read the documentation for more details on how to use bblocks packages
and the motivation for their creation.
bblocks is an umbrella package that includes several specialized sub-packages,
each designed to address specific data tasks.
bblocks-places: Resolve and standardize place names, including countries, regions, and other geographic entities.bblocks-data-importers: Tools to import data from different international development sources such as IMF, World Bank and many others.
It’s easy to get started with bblocks. Whether you want the full distribution or only specific tools, installation is flexible and straightforward.
This is recommended for most users who want access to all the tools in the bblocks ecosystem.
pip install bblocks[all]Install only the packages you need without unnecessary dependencies. For example, to install the places package:
pip install bblocks[places]You can also install individual packages directly:
pip install bblocks-placesOnce installed, you can start using bblocks packages in your analysis pipelines. Here's a
quick example of fetching some data using th bblocks-data-importers and resolving entity
names with bblocks-places:
from bblocks.data_importers import WorldBank
from bblocks import places
# Fetch World Bank data for series "SI.POV.DDAY" (poverty headcount ratio at $3.00 a day)
wb = WorldBank()
df = wb.get_data(series="SI.POV.DDAY")
# Resolve entities to short names
df["country"] = places.resolve_places(df["entity_name"], to="name_short", not_found="ignore")
# filter for African countries
gdp_africa = places.filter_african_countries(df["country"])
# Keep only relevant columns
gdp_africa = gdp_africa.loc[:, ["year", "country", "value",]]
# preview the results
print(gdp_africa.head())
# Output:
# year country value
# 0 1988 Algeria 12.3
# 1 1995 Algeria 11.8
# 2 2011 Algeria 0
# 3 2000 Angola 27
# ...We welcome contributions to bblocks! If you have ideas for new packages, improvements, or bug fixes,
please check out our contributing guidelines
for details on how to get involved.