Add @subset

docs/src/index.md

@@ -12,7 +12,7 @@ in C#.
 In addition, DataFramesMeta provides 
 
 * `@orderby`, for sorting data frames
-* `@subset` and `@subset!`, for keeping rows of a DataFrame matching a given condition
+* `@subset` and `@subset!`, for keeping rows of a data frame matching a given condition
 * `@by`, for grouping and combining a data frame in a single step
 * `@with`, for working with the columns of a data frame with high performance and 
   convenient syntax

src/macros.jl

@@ -472,7 +472,7 @@ and
 
 !!! note
     `@subset` treats `missing` values as `false` when filtering rows.
-    Unlike `DataFrames.subset` and other boolean operations with
+    Unlike `DataFrames.subset` and other Boolean operations with
     `missing`, `@subset` will *not* error on missing values, and
     will only keep `true` values.
 
@@ -631,7 +631,7 @@ and
 
 !!! note
     `@subset!` treats `missing` values as `false` when filtering rows.
-    Unlike `DataFrames.subset!` and other boolean operations with
+    Unlike `DataFrames.subset!` and other Boolean operations with
     `missing`, `@subset!` will *not* error on missing values, and
     will only keep `true` values.
 

src/parsing.jl

@@ -183,7 +183,10 @@ end
 # We need wrap_byrow as a keyword argument here in case someone
 # uses `@transform df @byrow begin ... end`, which we
 # deal with outside of this function.
-function fun_to_vec(ex::Expr; gensym_names::Bool=false, outer_flags=nothing, no_dest::Bool=false)
+function fun_to_vec(ex::Expr;
+                    gensym_names::Bool=false,
+                    outer_flags::Union{NamedTuple, Nothing}=nothing,
+                    no_dest::Bool=false)
     # classify the type of expression
     # :x # handled via dispatch
     # cols(:x) # handled as though above
@@ -333,19 +336,20 @@ function replace_dotted!(e, membernames)
     Expr(:., x_new, y_new)
 end
 
-"""
-    create_args_vector(args...) -> vec, wrap_byrow
-
-Given multiple arguments which can be any type
-of expression-like object (`Expr`, `QuoteNode`, etc.),
-puts them into a single array, removing line numbers.
-"""
 function create_args_vector(args...)
     create_args_vector(Expr(:block, args...))
 end
 
 """
    create_args_vector(arg) -> vec, outer_flags
+
+Given an expression return a vector of operations
+and a `NamedTuple` of the macro-flags that appear
+in the expression.
+
+If a `:block` expression, returns the `args` of
+the block as an array. If a simple expression,
+wrap the expression in a one-element vector.
 """
 function create_args_vector(arg)
     arg, outer_flags = extract_macro_flags(MacroTools.unblock(arg))

test/dataframes.jl

@@ -700,17 +700,17 @@ end
 
     @test d.y == [true]
 
-    d = @where(df, @linenums_macro begin end)
+    d = @subset(df, @linenums_macro begin end)
 
     @test nrow(d) == 1
 
-    d = @where df begin
+    d = @subset df begin
         @byrow @linenums_macro_byrow begin end
     end
 
     @test nrow(d) == 1
 
-    d = @where df @byrow begin
+    d = @subset df @byrow begin
         @linenums_macro_byrow begin end
     end
 

test/deprecated.jl

@@ -143,5 +143,20 @@ end
     @test d ≅ @where(df, :A .> 1, :B .> 1)
 end
 
+@testset "@where with a grouped data frame" begin
+    df = DataFrame(
+        g = [1, 1, 1, 2, 2],
+        i = 1:5,
+        t = ["a", "b", "c", "c", "e"],
+        y = [:v, :w, :x, :y, :z],
+        c = [:g, :quote, :body, :transform, missing]
+    )
+
+    gd = groupby(df, :g)
+
+    @test @where(gd, :i .== first(:i)) ≅ df[[1, 4], :]
+    @test @where(gd, cols(:i) .> mean(cols(:i)), :t .== "c") ≅ df[[3], :]
+    @test @where(gd, :c .== :g) ≅ df[[], :]
+end
 
 end # module

test/grouping.jl

@@ -357,19 +357,4 @@ end
 	@test @select(g, :a, @byrow t = :a ^ 2).t ≅ d.a .^ 2
 end
 
-@testset "@where with a grouped data frame" begin
-    df = DataFrame(
-        g = [1, 1, 1, 2, 2],
-        i = 1:5,
-        t = ["a", "b", "c", "c", "e"],
-        y = [:v, :w, :x, :y, :z],
-        c = [:g, :quote, :body, :transform, missing]
-    )
-
-    gd = groupby(df, :g)
-
-    @test @where(gd, :i .== first(:i)) ≅ df[[1, 4], :]
-    @test @where(gd, cols(:i) .> mean(cols(:i)), :t .== "c") ≅ df[[3], :]
-    @test @where(gd, :c .== :g) ≅ df[[], :]
-end
 end # module

test/subset.jl

@@ -79,7 +79,9 @@ end
 
     x = [2, 1, 0, 0]
 
-    @test @subset!(copy(df), :A .> 1) == df[(df.A .> 1) .=== true,:]
+    df2 = copy(df)
+    @test @subset!(df2, :A .> 1) === df2
+    @test df2 == df[(df.A .> 1) .=== true,:]
     @test @subset!(copy(df), :B .> 1) == df[df.B .> 1,:]
     @test @subset!(copy(df), :A .> x) == df[(df.A .> x) .=== true,:]
     @test @subset!(copy(df), :B .> x) ≅ df[df.B .> x,:]
@@ -140,4 +142,20 @@ end
     @test d ≅ @subset!(copy(df), :A .> 1, :B .> 1)
 end
 
-end # module
+@testset "@subset with a grouped data frame" begin
+    df = DataFrame(
+        g = [1, 1, 1, 2, 2],
+        i = 1:5,
+        t = ["a", "b", "c", "c", "e"],
+        y = [:v, :w, :x, :y, :z],
+        c = [:g, :quote, :body, :transform, missing]
+    )
+
+    gd = groupby(df, :g)
+
+    @test @subset(gd, :i .== first(:i)) ≅ df[[1, 4], :]
+    @test @subset(gd, cols(:i) .> mean(cols(:i)), :t .== "c") ≅ df[[3], :]
+    @test @subset(gd, :c .== :g) ≅ df[[], :]
+end
+
+end # module