Read the GBIF enums from the API endpoints

GBIF/Project.toml

@@ -1,7 +1,7 @@
 name = "GBIF"
 uuid = "ee291a33-5a6c-5552-a3c8-0f29a1181037"
 authors = ["Timothée Poisot <timothee.poisot@umontreal.ca>"]
-version = "0.4.2"
+version = "0.4.3"
 
 [deps]
 Dates = "ade2ca70-3891-5945-98fb-dc099432e06a"

GBIF/src/GBIF.jl

@@ -6,90 +6,59 @@ using Dates
 using Tables
 
 const gbifurl = "http://api.gbif.org/v1/"
+
+"""
+    enumerablevalues()
+
+Returns an *array* of values that can be enumerated by the GBIF API.
+"""
+function enumerablevalues()
+    endpoint = GBIF.gbifurl * "enumeration/basic"
+    req = HTTP.get(endpoint)
+    if isequal(200)(req.status)
+        response = JSON.parse(String(req.body))
+        return convert(Vector{String}, response)
+    end
+    return nothing
+end
+
+# We load the keys that can be enumerated when we load the package
+const gbifenumkeys = enumerablevalues()
+
+"""
+    enumeratedvalues(enumerable::String)
+
+For a given enumerable value (given as a string as reported by the output of the `enumerablevalues` function), this function will return an array of possible values.
+"""
+function enumeratedvalues(enumerable::String)
+    if enumerable in GBIF.gbifenumkeys
+        endpoint = GBIF.gbifurl * "enumeration/basic/$(enumerable)"
+        req = HTTP.get(endpoint)
+        if isequal(200)(req.status)
+            response = JSON.parse(String(req.body))
+            return convert(Vector{String}, response)
+        else
+            throw(ErrorException("Unable to enumerate the value $(enumerable)"))
+        end
+    else
+        throw(
+            ArgumentError(
+                "$(enumerable) is not an enumerable entity for GBIF -- see GBIF.enumerablevalues()",
+            ),
+        )
+    end
+end
+
 const gbifenums = Dict(
-    "basisOfRecord" => [
-        "FOSSIL_SPECIMEN",
-        "HUMAN_OBSERVATION",
-        "LITERATURE",
-        "LIVING_SPECIMEN",
-        "MACHINE_OBSERVATION",
-        "MATERIAL_SAMPLE",
-        "OBSERVATION",
-        "PRESERVED_SPECIMEN",
-        "UNKNOWN",
-    ],
-    "occurrenceStatus" => [
-        "PRESENT", "ABSENT",
-    ],
-    "continent" => [
-        "AFRICA",
-        "ANTARCTICA",
-        "ASIA",
-        "EUROPE",
-        "NORTH_AMERICA",
-        "OCEANIA",
-        "SOUTH_AMERICA",
-    ],
-    "establishmentMeans" => [
-        "INTRODUCED",
-        "INVASIVE",
-        "MANAGED",
-        "NATIVE",
-        "NATURALISED",
-        "UNCERTAIN",
-    ],
-    "issue" => [
-        "BASIS_OF_RECORD_INVALID",
-        "CONTINENT_COUNTRY_MISMATCH",
-        "CONTINENT_DERIVED_FROM_COORDINATES",
-        "CONTINENT_INVALID",
-        "COORDINATE_INVALID",
-        "COORDINATE_OUT_OF_RANGE",
-        "COORDINATE_PRECISION_INVALID",
-        "COORDINATE_REPROJECTED",
-        "COORDINATE_REPROJECTION_FAILED",
-        "COORDINATE_REPROJECTION_SUSPICIOUS",
-        "COORDINATE_ROUNDED",
-        "COORDINATE_UNCERTAINTY_METERS_INVALID",
-        "COUNTRY_COORDINATE_MISMATCH",
-        "COUNTRY_DERIVED_FROM_COORDINATES",
-        "COUNTRY_INVALID",
-        "COUNTRY_MISMATCH",
-        "DEPTH_MIN_MAX_SWAPPED",
-        "DEPTH_NON_NUMERIC",
-        "DEPTH_NOT_METRIC",
-        "DEPTH_UNLIKELY",
-        "ELEVATION_MIN_MAX_SWAPPED",
-        "ELEVATION_NON_NUMERIC",
-        "ELEVATION_NOT_METRIC",
-        "ELEVATION_UNLIKELY",
-        "GEODETIC_DATUM_ASSUMED_WGS84",
-        "GEODETIC_DATUM_INVALID",
-        "IDENTIFIED_DATE_INVALID",
-        "IDENTIFIED_DATE_UNLIKELY",
-        "INDIVIDUAL_COUNT_INVALID",
-        "INTERPRETATION_ERROR",
-        "MODIFIED_DATE_INVALID",
-        "MODIFIED_DATE_UNLIKELY",
-        "MULTIMEDIA_DATE_INVALID",
-        "MULTIMEDIA_URI_INVALID",
-        "PRESUMED_NEGATED_LATITUDE",
-        "PRESUMED_NEGATED_LONGITUDE",
-        "PRESUMED_SWAPPED_COORDINATE",
-        "RECORDED_DATE_INVALID",
-        "RECORDED_DATE_MISMATCH",
-        "RECORDED_DATE_UNLIKELY",
-        "REFERENCES_URI_INVALID",
-        "TAXON_MATCH_FUZZY",
-        "TAXON_MATCH_HIGHERRANK",
-        "TAXON_MATCH_NONE",
-        "TYPE_STATUS_INVALID",
-        "ZERO_COORDINATE",
-    ],
+    "basisOfRecord" => enumeratedvalues("BasisOfRecord"),
+    "occurrenceStatus" => enumeratedvalues("OccurrenceStatus"),
+    "continent" => enumeratedvalues("Continent"),
+    "country" => enumeratedvalues("Country"),
+    "establishmentMeans" => enumeratedvalues("EstablishmentMeans"),
+    "issue" => enumeratedvalues("OccurrenceIssue"),
+    "mediaType" => enumeratedvalues("MediaType"),
 )
 
-# Load the main functions
-
 include("query.jl")
 
 include("types/GBIFTaxon.jl")

GBIF/src/occurrence.jl

@@ -87,7 +87,7 @@ is "ABSENT"; therefore, for the majority of uses, your query will *at least* con
 `"occurrenceStatus" => "PRESENT"`.
 """
 function occurrences(query::Pair...)
-    retrieved, offset, of_max = _internal_occurrences_getter(query...)
+    retrieved, _, of_max = _internal_occurrences_getter(query...)
     if !isnothing(retrieved)
         store = Vector{GBIFRecord}(undef, of_max)
         store[1:length(retrieved)] = retrieved

GBIF/src/query.jl

@@ -1,5 +1,7 @@
 """
-**Checks that the queries for occurrences searches are well formatted**
+    validate_occurrence_query(query::Pair)
+
+Checks that the queries for occurrences searches are well formatted.
 
 This is used internally.
 

Project.toml

@@ -23,7 +23,7 @@ GBIF = "0.4"
 GDAL = "1.5"
 MakieCore = "0.6"
 Reexport = "1.2"
-SimpleSDMDatasets = "0.1"
 SimpleSDMLayers = "0.9"
+SimpleSDMDatasets = "0.1"
 StatsBase = "0.33"
 julia = "1.8"

SimpleSDMDatasets/src/types/datasets.jl

@@ -7,15 +7,21 @@ given by `RasterProvider`s, and the same dataset can have multiple providers.
 abstract type RasterDataset end
 
 # List of datasets
+# TODO write documentation to also mention what they can be used for, e.g. all DEM can
+# provide Elevation
 struct BioClim <: RasterDataset end
+
 struct Elevation <: RasterDataset end
+
 struct MinimumTemperature <: RasterDataset end
 struct MaximumTemperature <: RasterDataset end
 struct AverageTemperature <: RasterDataset end
 struct Precipitation <: RasterDataset end
 struct SolarRadiation <: RasterDataset end
 struct WindSpeed <: RasterDataset end
 struct WaterVaporPressure <: RasterDataset end
+
 struct LandCover <: RasterDataset end
+
 struct HabitatHeterogeneity <: RasterDataset end
 struct Topography <: RasterDataset end

SimpleSDMDatasets/src/types/futures.jl

@@ -107,6 +107,11 @@ CMIP6Model = Union{
     UKESM1_0_LL,
 }
 
+"""
+    CMIP5Scenario
+
+These scenarios are part of CMIP5. They can be `RCP26` to `RCP85` (with `RCPXX` the scenario).
+"""
 CMIP5Scenario = Union{RCP26, RCP45, RCP60, RCP85}
 CMIP5Model = Union{
     ACCESS1_0,

SimpleSDMDatasets/src/types/providers.jl

@@ -6,7 +6,15 @@ For example, WorldClim2 and CHELSA2 are `RasterProvider`s.
 """
 abstract type RasterProvider end
 
+"""
+    WorldClim2
+
+This provider offers access to the version 2 of the WorldClim data, accessible from `http://www.worldclim.com/version2`.
+"""
 struct WorldClim2 <: RasterProvider end
+
 struct EarthEnv <: RasterProvider end
+
 struct CHELSA1 <: RasterProvider end
-struct CHELSA2 <: RasterProvider end
+
+struct CHELSA2 <: RasterProvider end

dev.jl

@@ -124,4 +124,4 @@ xlims!(ax, -0.2, 1.2)
 ylims!(ax, -0.2, 1.2)
 hidedecorations!(ax)
 
-current_figure()
+current_figure()

docs/make.jl

@@ -61,6 +61,7 @@ makedocs(;
                 "Interface to GBIF data" => "manual/GBIF/index.md",
                 "GBIF data representation" => "manual/GBIF/types.md",
                 "GBIF data retrieval" => "manual/GBIF/data.md",
+                "Enumerated query parameters" => "manual/GBIF/enums.md",
             ],
             "SimpleSDMDatasets.jl" => [
                 "Interface to raster data" => "manual/SimpleSDMDatasets/index.md",

docs/src/manual/GBIF/enums.md

@@ -0,0 +1,12 @@
+# Enumerated values
+
+The GBIF API has a number of controlled vocabularies to perform queries (also called "enumerations"). In order to keep the API and the package in sync, when the package is loaded, we query the API to see what values are enumerable, and what values are acceptable for each of these categories.
+
+```@docs
+GBIF.enumerablevalues
+GBIF.enumeratedvalues
+```
+
+These functions are *not* exported, and are only called once per session to populate
+a dictionary with the accepted values. Note that at the moment, the only enumerated values
+that we store are the one accepted as search argument by the occurrence search endpoint.

recipes.jl

@@ -1,3 +1,5 @@
+using SimpleSDMLayers
+
 @shorthands bivariate
 @shorthands bivariatelegend
 @shorthands trivariate
@@ -6,7 +8,7 @@
 """
 test 1
 """
-@recipe function plot(layer::T) where {T<:SimpleSDMLayer}
+@recipe function plot(layer::T) where {T <: SimpleSDMLayer}
     seriestype --> :heatmap
     if get(plotattributes, :seriestype, :heatmap) in [:heatmap, :contour]
         eltype(layer) <: AbstractFloat || throw(
@@ -32,20 +34,20 @@ test 2
 @recipe function plot(
     l1::FT,
     l2::ST;
-    classes::Int=3,
-    p0=colorant"#e8e8e8ff",
-    p1=colorant"#64acbeff",
-    p2=colorant"#c85a5aff",
-    quantiles=true,
-) where {FT<:SimpleSDMLayer,ST<:SimpleSDMLayer}
+    classes::Int = 3,
+    p0 = colorant"#e8e8e8ff",
+    p1 = colorant"#64acbeff",
+    p2 = colorant"#c85a5aff",
+    quantiles = true,
+) where {FT <: SimpleSDMLayer, ST <: SimpleSDMLayer}
     eltype(l1) <: Number || throw(
         ArgumentError(
-            "Plotting is only supported for layers with number values ($(eltype(l1)))"
+            "Plotting is only supported for layers with number values ($(eltype(l1)))",
         ),
     )
     eltype(l2) <: Number || throw(
         ArgumentError(
-            "Plotting is only supported for layers with number values ($(eltype(l2)))"
+            "Plotting is only supported for layers with number values ($(eltype(l2)))",
         ),
     )
     seriestype --> :scatter
@@ -130,21 +132,21 @@ end
 test 2
 """
 @recipe function plot(
-    x::FT, y::ST, z::TT; quantiles=true, simplex=false, red="", green="", blue=""
-) where {FT<:SimpleSDMLayer,ST<:SimpleSDMLayer,TT<:SimpleSDMLayer}
+    x::FT, y::ST, z::TT; quantiles = true, simplex = false, red = "", green = "", blue = "",
+) where {FT <: SimpleSDMLayer, ST <: SimpleSDMLayer, TT <: SimpleSDMLayer}
     eltype(x) <: Number || throw(
         ArgumentError(
-            "Plotting is only supported for layers with number values ($(eltype(x)))"
+            "Plotting is only supported for layers with number values ($(eltype(x)))",
         ),
     )
     eltype(y) <: Number || throw(
         ArgumentError(
-            "Plotting is only supported for layers with number values ($(eltype(y)))"
+            "Plotting is only supported for layers with number values ($(eltype(y)))",
         ),
     )
     eltype(z) <: Number || throw(
         ArgumentError(
-            "Plotting is only supported for layers with number values ($(eltype(z)))"
+            "Plotting is only supported for layers with number values ($(eltype(z)))",
         ),
     )
     SimpleSDMLayers._layers_are_compatible(x, y)
@@ -206,26 +208,33 @@ test 2
                 y0 = (row_number - 1) * h
                 push!(shapes_x, [x0, x0 + a / 2, x0 + a])
                 push!(shapes_y, [y0, y0 + h, y0])
-                push!(shapes_colors, rgb_from_xy(x0 + a / 2, y0 + h / 2; simplex=simplex))
+                push!(shapes_colors, rgb_from_xy(x0 + a / 2, y0 + h / 2; simplex = simplex))
                 if row_number > 1
                     push!(shapes_x, [x0, x0 + a / 2, x0 + a])
                     push!(shapes_y, [y0, y0 - h, y0])
-                    push!(shapes_colors, rgb_from_xy(x0 + a / 2, y0 - h / 2; simplex=simplex))
+                    push!(
+                        shapes_colors,
+                        rgb_from_xy(x0 + a / 2, y0 - h / 2; simplex = simplex),
+                    )
                 end
             end
         end
         _fs = get(plotattributes, :annotationfontsize, 14)
         @series begin
             seriestype := :shape
-            annotations := [(-0.05, 0.0, (red, _fs, :left, 60.0)), (1.0, -0.05, (green, _fs, :right, 0.0)), (0.5+0.05, sqrt(3)/2, (blue, _fs, :left, -60.0))]
+            annotations := [
+                (-0.05, 0.0, (red, _fs, :left, 60.0)),
+                (1.0, -0.05, (green, _fs, :right, 0.0)),
+                (0.5 + 0.05, sqrt(3) / 2, (blue, _fs, :left, -60.0)),
+            ]
             markersize := 0
             seriescolor := hcat(shapes_colors...)
             shapes_x, shapes_y
         end
     end
 end
 
-function rgb_from_xy(x, y; simplex=false)
+function rgb_from_xy(x, y; simplex = false)
     b = 2y / sqrt(3)
     g = (2x - b) / 2
     r = 1 - (b + g)

src/SpeciesDistributionToolkit.jl

@@ -39,6 +39,4 @@ include("pseudoabsences.jl")
 export WithinRadius, SurfaceRangeEnvelope, RandomSelection
 export pseudoabsencemask
 
-
-
 end # module SpeciesDistributionToolkit