Merge pull request #26 from EcoJulia/feature/vernacular

✨ vernacular names

Project.toml

@@ -1,7 +1,7 @@
 name = "NCBITaxonomy"
 uuid = "f88b31d2-eb98-4433-b52d-2dd32bc6efce"
 authors = ["Timothée Poisot <[email protected]>"]
-version = "0.0.5"
+version = "0.0.6"
 
 [deps]
 Arrow = "69666777-d1a9-59fb-9406-91d4454c9d45"

docs/src/namefinding.md

@@ -4,6 +4,7 @@
 
 ```@docs
 taxid
+vernacular
 ```
 
 The `taxid` function will return a `NCBITaxon` object, which has two fields:
@@ -28,6 +29,13 @@ the *scientific name* will be returned, no matter what you search
 taxid("cow")
 ```
 
+This may be a good point to note that we can use the `vernacular` function to
+get a list of NCBI-known vernacular names:
+
+```@example taxid
+taxid("cow") |> vernacular
+```
+
 You can pass an additional `fuzzy=true` keyword argument to the `taxid` function
 to perform fuzzy name matching using the Levenshtein distance:
 

src/NCBITaxonomy.jl

@@ -43,4 +43,7 @@ export children, descendants
 include("lineage.jl")
 export lineage, parent, rank
 
+include("vernacular.jl")
+export vernacular
+
 end

src/vernacular.jl

@@ -0,0 +1,15 @@
+"""
+    vernacular(t::NCBITaxon)
+
+This function will return `nothing` if no vernacular name is known, and an array
+of names if found. It searches the "common name" and "genbank common name"
+category of the NCBI taxonomy name table.
+"""
+function vernacular(t::NCBITaxon)
+    names_from_tax = filter(r -> r.tax_id == t.id, NCBITaxonomy.names_table)
+    common_names = filter(r -> r.class == NCBITaxonomy.class_common_name, names_from_tax)
+    genbank_names = filter(r -> r.class == NCBITaxonomy.class_genbank_common_name, names_from_tax)
+    all_names = vcat(common_names.name, genbank_names.name)
+    length(all_names) == 0 && return nothing
+    return unique(all_names)
+end

test/taxid.jl

@@ -38,4 +38,12 @@ module TestTaxid
     chicken = taxid("tchiken"; fuzzy=true, verbose=true)
     @test typeof(chicken) == NCBITaxon
 
+    #Vernacular name
+    chub = vernacular(ncbi"Leuciscus cephalus")
+    @test "European chub" in chub
+    @test "chub" in chub
+
+    # Vernacular missing
+    @test isnothing(vernacular(ncbi"Lamellodiscus elegans"))
+
 end