Hello, Increasingly I find myself needing to join dataframes on Geographic Information System (GIS) objects and it seems no capability exists to do so in the Julia eco-system. Nor does it seem there is much interest (c.f. Perform spatial join with GeoDataFrames ) . Using python I could do this imp…

GeometryOps.jl offers this in combination with FlexiJoins today! Spatial joins | GeometryOps.jl A VitePress Site

I would assume that [ANN] FlexiJoins.jl: fresh take on joining datasets by @aplavin should be able to do it.

I fear it’s not yet doable within Julia as I know of no package that allows for completely generic predicates. I couldn’t get FlexiJoins predicates to work with custom methods, and I think both DataFrames as FlexiJoins eventually hash the values to do comparisons, which won’t work on geometries. Fo…

[image] bkamins: I would assume that [ANN] FlexiJoins.jl: fresh take on joining datasets by @aplavin should be able to do it. FlexiJoins support optimized spatial joins of points - as in “find pairs of points within the specified distance” or “find the closest match in B for each point from …

I don’t know if this will help [image] Efficiently check if points are contained in polygons Geo To clarify, Rasters.jl polygon ops are also pure julia via PolygonInbounds.jl and other internal code. It’s pretty fast for vectors of points, but not so much for single points…

[image] aplavin: But no optimization for polygon joins is available there, only the naive O(n^2) looping approach. How would one execute the naive version with the above example? Can it make use of intersect/within methods of other packages? [image] aplavin: FlexiJoins support lots of jo…

[image] rocco_sprmnt21: I don’t know if this will help PolygonInbounds.jl is a good example of a specific subset of geospatial predicates, written natively in Julia Overall there are more ( DE-9IM - Wikipedia ) and these are covered by packages like LibGEOS and ArchGDAL (itself linking to GEOS …

[image] evetion: How would one execute the naive version with the above example? Can it make use of intersect/within methods of other packages? FlexiJoins can do joins with arbitrary predicates by just looping over all pairs: julia> using FlexiJoins julia> my_predicate(s, r) = occursin(r, s…

[image] evetion: PolygonInbounds.jl is a good … I’m not sure I understand the remarks in response to my question. Surely also because I misquoted the reference to the discussion “Efficiently check if points are contained in polygons”. What I asked to have clarified is if the use of one of …

Here is an attempt to use FlexiJoins by first finding a bounding box for each geometry. This reduces the quadratic complexity of the query (depending on how annoying the geometries are): # prep data & environment using DataFrames, GeoDataFrames, ArchGDAL, FlexiJoins, IntervalSets dfa = GeoDataFram…

@Dan nice solution! Should definitely outperform the naive O(n^2) one for large datasets, even with the overhead of two sequential joins. Btw, materializing x and y coordinates into separate columns isn’t required: you can pass r -> first(r.geometry) instead of :x in by_pred. The best interface w…

Spatial join with dataframes

General Usage

rocco_sprmnt21 January 20, 2023, 10:55am 5

I don’t know if this will help

1 Like

Topic		Replies	Views
Perform spatial join with GeoDataFrames New to Julia dataframes , geo	3	914	January 20, 2023
[ANN] FlexiJoins.jl: fresh take on joining datasets Package Announcements	33	3631	September 2, 2025
Join dataframe and geotable New to Julia geostat	34	909	April 10, 2024
Conditional left join 2 dataframes when none of the columns are common General Usage dataframes , flexijoins	26	2152	May 26, 2022
[ANN] DataFrameIntervals.jl — joins on intervals of time Package Announcements dataframes	40	2580	October 31, 2022

Spatial join with dataframes

Related topics