This project saves a little scraper to save data published in Official ONPE Page related to Peruvian 2021 presidential elections.
It works pretty simple, we have public access to election results for each "mesa", where a "mesa" is a chunk of about 200-250 citizens with Peruvian nationality. To access to one "mesa" we can fetch the results from:
"https://api.resultadossep.eleccionesgenerales2021.pe/mesas/detalle/<MESA_NUMBER>"
Where <MESA_NUMBER>
is a six digits number and refers to one "mesa" results.
Then, we only obtain all the available "mesa"s.
With the script called main.py
, which contains an algorithm to fetch data in a concurrent way, we can extract in partitioned CSV format files.
To run, first, you need to install pdm and run the following:
pdm install
pdm run python main.py
You can use the extracted data, it is saved in dataset/dataset_*.csv
. Each csv 20000 rows, where each row is a "mesa" result.
Each "mesa" have the next columns:
ubigeo
place
address
department
province
district
copy_code
observation
description
candidates
total_citizens
partido_politico_nacional_peru_libre_total_valids
partido_politico_nacional_peru_libre_total_emiteds
partido_politico_nacional_peru_libre_code
partido_politico_nacional_peru_libre_congresal
partido_politico_nacional_peru_libre_list_code
fuerza_popular_total_valids
fuerza_popular_total_emiteds
fuerza_popular_code
fuerza_popular_congresal
fuerza_popular_list_code
total_votos_validos_total_valids
total_votos_validos_total_emiteds
total_votos_validos_code
total_votos_validos_congresal
votos_en_blanco_total_valids
votos_en_blanco_total_emiteds
votos_en_blanco_code
votos_en_blanco_congresal
votos_nulos_total_valids
votos_nulos_total_emiteds
votos_nulos_code
votos_nulos_congresal
votos_impugnados_total_valids
votos_impugnados_total_emiteds
votos_impugnados_code
votos_impugnados_congresal
total_votos_emitidos_total_valids
total_votos_emitidos_total_emiteds
total_votos_emitidos_code
total_votos_emitidos_congresa
I know, we need to document each column, but it will keep as a To-do. BTW, I think some columns have an intuitive name. Please, If you want/can help us with the naming and description, only create an Issue.
In the analytics.ipynb
notebook you can an example of manipulation of the partitioned data.