CEP data

[JoaoCarabetta]

Following up our conversation about adding CEP data to geobr.

The current idea is to map CEP numbers to h3 hexagons, so that each hexagon would have a list of CEPs.

API

get_cep_by_geo(latitude, longitude, h3_id)
"""
Gets H3 hexagon in that maps to the coordinates and returns a list of CEP numbers.

The user can use lat, lon or the h3_id to get the CEP numbers

Return
-------
Dataframe:
     h3_id
     h3_geometry
     cep_number
"""

get_geo_by_cep(cep_number)
"""
Gets all H3 hexagons that contains the CEP number. 

Return
-------
Dataframe:
      h3_id
      h3_geometry
      cep_number

Infrastructure

get_cep_by_geo

1. Map latitude, longitude to h3 cell using the H3 package that is available for python and R.
2. Get the selected h3 cell parents
3. Search a file system with the following structure:
   - h3_cell_resolution_4_id
   - h3_cell_resolution_5_id
   - h3_cell_resolution_6_id
   - ... up to a convenient resolution
   On a reasonable resolution, we should save a dataframe that is small enough to be downloaded and filtered.
4. Download df, filter by h3 cell id and return

get_geo_by_cep

1. Search a file system with the following structure:
   - first CEP digit
   - second CEP digit
   - third CEP digit
   - ....
   On a reasonable CEP digit, we should save a dataframe that is small enough to be downloaded and filtered.
2. Download df, filter by h3 cell id and return

@rafapereirabr let me know what you think. It is not easy to build those indexes, but I it would be incredible to have it!