Convert some data into Panda DataFrames

British Petroleum (BP)

It parse sheet like Primary Energy Consumption (not like Primary Energy - Cons by fuel).

Open: http://www.bp.com/statisticalreview or https://www.bp.com/en/global/corporate/energy-economics/statistical-review-of-world-energy.html

Download Statistical Review of World Energy – all data.

Use:

from shifter_pandas.bp import UNITS_ENERGY, BPDatasource

shifter_ds = BPDatasource("bp-stats-review-2021-all-data.xlsx")

df = shifter_ds.datasource(units_filter=UNITS_ENERGY, regions_filter=["Switzerland"])
df

Swiss Office Federal of Statistics (OFS)

From https://www.bfs.admin.ch/bfs/fr/home/services/recherche/stat-tab-donnees-interactives.html create a stat table.

Click on À propos du tableau

Click on Rendez ce tableau disponible dans votre application

Use:

from shifter_pandas.ofs import OFSDatasource

shifter_ds = OFSDatasource("<URL>")

df = shifter_ds.datasource(<Requête Json>)
df

And replace <URL> and <Requête Json> with the content of the fields of the OFS web page.

Interesting sources

• Parc de motocycles par caractéristiques techniques et émissions
• Bilan démographique selon l'âge et le canton

Our World in Data

Select a publication.

Click Download.

Click Full data (CSV).

Use:

import pandas as pd
from shifter_pandas.wikidata_ import WikidataDatasource

df_owid = pd.read_csv("<file name>")
wdds = WikidataDatasource()
df_wd = wdds.datasource_code(wikidata_id=True, wikidata_name=True, wikidata_type=True)
df = pd.merge(df_owid, df_wd, how="inner", left_on='iso_code', right_on='Code')
df

Interesting sources

• GDP, 1820 to 2018
• Population, 1800 to 2021

World Bank

Open https://data.worldbank.org/

Find a chart

In Download click CSV

Use:

from shifter_pandas.worldbank import wbDatasource

df = wbDatasource("<file name>")
df

Interesting sources

• GDP (current US$)
• GDP (constant 2015 US$)

Wikidata

By providing the wikidata_* parameters, you can ass some data from WikiData.

Careful, the WikiData is relatively slow then the first time you run it il will be slow. We use a cache to make it fast the next times.

You can also get the country list with population and ISO 2 code with:

from shifter_pandas.wikidata_ import (
    ELEMENT_COUNTRY,
    PROPERTY_ISO_3166_1_ALPHA_2,
    PROPERTY_POPULATION,
    WikidataDatasource,
)

shifter_ds = WikidataDatasource()
df = shifter_ds.datasource(
    instance_of=ELEMENT_COUNTRY,
    with_id=True,
    with_name=True,
    properties=[PROPERTY_ISO_3166_1_ALPHA_2, PROPERTY_POPULATION],
    limit=1000,
)
df

Contributing

Install the pre-commit hooks:

pip install pre-commit
pre-commit install --allow-missing-config