search

ipeaGIT / geobr

Easy access to official spatial data sets of Brazil in R and Python
https://ipeagit.github.io/geobr/
786 stars 118 forks source link

Time for a Julia version? :) #264

Closed juliohm closed 4 months ago

juliohm commented 2 years ago

We have developed GeoTables.jl and GADM.jl over the years to download spatial geometries for any place in the world by name. I would be nice to have a GeoBR.jl package in Julia as well. 💯

rafapereirabr commented 2 years ago

Hi Julio. That could be a great idea. Could you please point me to these Julia packages you have created already?

juliohm commented 2 years ago

Here are the two packages I mentioned:

The former uses the latter and you can watch it in action in my JuliaCon 2021 talk:

https://youtu.be/75A6zyn5pIE

A GeoBR.jl would be a nice addition in the sense that it is a data set curated by IBGE specifically for Brazil boundaries, correct? I learned about the project today, didn't have time to investigate it carefully yet.

rafapereirabr commented 2 years ago

Hi @juliohm , thanks for the links. It looks somewhat similar with Python, so assuming it's possible to read spatial data sets in geopackage.gpkg format, building a Julia version of geobr should be relatively simple. We should probably schedule a call to discuss this. Perhps @JoaoCarabetta might wanna join us.

Please mind, however, that geobr is not limited to IBGE data. We cover multiple official spatial data sets from various sources. Naturally, IBGE is the largest producer of spatial data in Brazil but we also cover data from INEP, DataSus, FUNAI, MMA etc.

Another important point you will note in our documentation. geobr does not simply store raw original data sets. One of the most important contributions of geobr is to preprocess the data for various years in order to harmonize attributes, projection and fix eventual topology issues. This is very important to make all data sets in geobr compatible and easy to manipulate for the end user

juliohm commented 2 years ago

That is great @rafapereirabr , let's set a call next week to discuss the most straightforward steps to get geobr working in Julia. We can read gpkg files through GDAL, and that is what GeoTables.jl does behind the scenes. If you store these datasets on a server, we can even setup a GeoBR.jl package that downloads data for specific regions of the world on demand like we did with GADM.jl.

JoaoCarabetta commented 2 years ago

Hey, let me know when you are meeting, I'll try to fit it into my schedule.

But, developing a Julia package should be quite straightforward if you can read gpkg. Just copying the python package structure might solve it.

juliohm commented 2 years ago

How does tomorrow morning sound to you? 11am BRT? My email is on my GitHub profile, appreciate if you can send an invitation with a proposal time.