Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix multi-column select! #404 #405

Merged
merged 4 commits into from
Nov 14, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion Project.toml
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
name = "DataFramesMeta"
uuid = "1313f7d8-7da2-5740-9ea0-a2ca25f37964"
version = "0.15.2"
version = "0.15.3"

[deps]
Chain = "8be319e6-bccf-4806-a6f7-6fae938471bc"
Expand Down
6 changes: 3 additions & 3 deletions src/macros.jl
Original file line number Diff line number Diff line change
Expand Up @@ -859,7 +859,7 @@
```

### Examples

Check failure on line 862 in src/macros.jl

View workflow job for this annotation

GitHub Actions / build

doctest failure in ~/work/DataFramesMeta.jl/DataFramesMeta.jl/src/macros.jl:862-953 ```jldoctest julia> using DataFramesMeta, Statistics julia> df = DataFrame(x = 1:3, y = [2, 1, 2]); julia> globalvar = [2, 1, 0]; julia> @subset(df, :x .> 1) 2×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 2 1 2 │ 3 2 julia> @subset(df, :x .> globalvar) 2×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 2 1 2 │ 3 2 julia> @subset df begin :x .> globalvar :y .== 3 end 0×2 DataFrame julia> df = DataFrame(n = 1:20, x = [3, 3, 3, 3, 1, 1, 1, 2, 1, 1, 2, 1, 1, 2, 2, 2, 3, 1, 1, 2]); julia> g = groupby(df, :x); julia> @subset(g, :n .> mean(:n)) 8×2 DataFrame Row │ n x │ Int64 Int64 ─────┼────────────── 1 │ 12 1 2 │ 13 1 3 │ 15 2 4 │ 16 2 5 │ 17 3 6 │ 18 1 7 │ 19 1 8 │ 20 2 julia> @subset g begin :n .> mean(:n) :n .< 20 end 7×2 DataFrame Row │ n x │ Int64 Int64 ─────┼────────────── 1 │ 12 1 2 │ 13 1 3 │ 15 2 4 │ 16 2 5 │ 17 3 6 │ 18 1 7 │ 19 1 julia> df = DataFrame(a = [1, 2, missing], b = ["x", "y", missing]); julia> @subset(df, :a .== 1) 1×2 DataFrame Row │ a b │ Int64? String? ─────┼───────────────── 1 │ 1 x julia> @subset(df, :a .< 3; view = true) 2×2 SubDataFrame Row │ a b │ Int64? String? ─────┼───────────────── 1 │ 1 x 2 │ 2 y julia> @subset df begin :a .< 3 @kwarg view = true end 2×2 SubDataFrame Row │ a b │ Int64? String? ─────┼───────────────── 1 │ 1 x 2 │ 2 y ``` Subexpression: @subset df begin :x .> globalvar :y .== 3 end Evaluated output: 0×2 DataFrame Row │ x y │ Int64 Int64 ─────┴────────────── Expected output: 0×2 DataFrame diff = Warning: Diff output requires color. 0×2 DataFrameDataFrame Row │ x y │ Int64 Int64 ─────┴──────────────
```jldoctest
julia> using DataFramesMeta, Statistics

Expand Down Expand Up @@ -975,7 +975,7 @@

Use this function as an alternative to placing the `.` to broadcast row-wise operations.

### Examples

Check failure on line 978 in src/macros.jl

View workflow job for this annotation

GitHub Actions / build

doctest failure in ~/work/DataFramesMeta.jl/DataFramesMeta.jl/src/macros.jl:978-1008 ```jldoctest julia> using DataFramesMeta julia> df = DataFrame(A=1:5, B=["apple", "pear", "apple", "orange", "pear"]) 5×2 DataFrame Row │ A B │ Int64 String ─────┼─────────────── 1 │ 1 apple 2 │ 2 pear 3 │ 3 apple 4 │ 4 orange 5 │ 5 pear julia> @rsubset df :A > 3 2×2 DataFrame Row │ A B │ Int64 String ─────┼─────────────── 1 │ 4 orange 2 │ 5 pear julia> @rsubset df :A > 3 || :B == "pear" 3×2 DataFrame Row │ A B │ Int64 String ─────┼─────────────── 1 │ 2 pear 2 │ 4 orange 3 │ 5 pear ``` Subexpression: @rsubset df :A > 3 || :B == "pear" Evaluated output: 3×2 DataFrame Row │ A B │ Int64 String ─────┼─────────────── 1 │ 2 pear 2 │ 4 orange 3 │ 5 pear Expected output: 3×2 DataFrame Row │ A B │ Int64 String ─────┼─────────────── 1 │ 2 pear 2 │ 4 orange 3 │ 5 pear diff = Warning: Diff output requires color. 3×2 DataFrame DataFrame Row │ A B B │ Int64 String ─────┼─────────────── String ─────┼─────────────── 1 │ 2 pear pear 2 │ 4 orange orange 3 │ 5 pear
```jldoctest
julia> using DataFramesMeta

Expand Down Expand Up @@ -1127,7 +1127,7 @@
```

### Examples

Check failure on line 1130 in src/macros.jl

View workflow job for this annotation

GitHub Actions / build

doctest failure in ~/work/DataFramesMeta.jl/DataFramesMeta.jl/src/macros.jl:1130-1204 ```jldoctest julia> using DataFramesMeta, Statistics julia> df = DataFrame(x = 1:3, y = [2, 1, 2]); julia> globalvar = [2, 1, 0]; julia> @subset!(copy(df), :x .> 1) 2×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 2 1 2 │ 3 2 julia> @subset!(copy(df), :x .> globalvar) 2×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 2 1 2 │ 3 2 julia> @subset! copy(df) begin :x .> globalvar :y .== 3 end 0×2 DataFrame julia> df = DataFrame(n = 1:20, x = [3, 3, 3, 3, 1, 1, 1, 2, 1, 1, 2, 1, 1, 2, 2, 2, 3, 1, 1, 2]); julia> g = groupby(copy(df), :x); julia> @subset!(g, :n .> mean(:n)) 8×2 DataFrame Row │ n x │ Int64 Int64 ─────┼────────────── 1 │ 12 1 2 │ 13 1 3 │ 15 2 4 │ 16 2 5 │ 17 3 6 │ 18 1 7 │ 19 1 8 │ 20 2 julia> g = groupby(copy(df), :x); julia> @subset! g begin :n .> mean(:n) :n .< 20 end 7×2 DataFrame Row │ n x │ Int64 Int64 ─────┼────────────── 1 │ 12 1 2 │ 13 1 3 │ 15 2 4 │ 16 2 5 │ 17 3 6 │ 18 1 7 │ 19 1 julia> d = DataFrame(a = [1, 2, missing], b = ["x", "y", missing]); julia> @subset!(d, :a .== 1) 1×2 DataFrame Row │ a b │ Int64? String? ─────┼───────────────── 1 │ 1 x ``` Subexpression: @subset! copy(df) begin :x .> globalvar :y .== 3 end Evaluated output: 0×2 DataFrame Row │ x y │ Int64 Int64 ─────┴────────────── Expected output: 0×2 DataFrame diff = Warning: Diff output requires color. 0×2 DataFrameDataFrame Row │ x y │ Int64 Int64 ─────┴──────────────
```jldoctest
julia> using DataFramesMeta, Statistics

Expand Down Expand Up @@ -1311,7 +1311,7 @@
$ASTABLE_RHS_ORDERBY_DOCS

### Examples

Check failure on line 1314 in src/macros.jl

View workflow job for this annotation

GitHub Actions / build

doctest failure in ~/work/DataFramesMeta.jl/DataFramesMeta.jl/src/macros.jl:1314-1386 ```jldoctest julia> using DataFramesMeta, Statistics julia> d = DataFrame(x = [3, 3, 3, 2, 1, 1, 1, 2, 1, 1], n = 1:10, c = ["a", "c", "b", "e", "d", "g", "f", "i", "j", "h"]); julia> @orderby(d, -:n) 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 10 h 2 │ 1 9 j 3 │ 2 8 i 4 │ 1 7 f 5 │ 1 6 g 6 │ 1 5 d 7 │ 2 4 e 8 │ 3 3 b 9 │ 3 2 c 10 │ 3 1 a julia> @orderby(d, invperm(sortperm(:c, rev = true))) 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 9 j 2 │ 2 8 i 3 │ 1 10 h 4 │ 1 6 g 5 │ 1 7 f 6 │ 2 4 e 7 │ 1 5 d 8 │ 3 2 c 9 │ 3 3 b 10 │ 3 1 a julia> @orderby d begin :x abs.(:n .- mean(:n)) end 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 e 2 │ 1 6 f 3 │ 1 7 g 4 │ 1 9 i 5 │ 1 10 j 6 │ 2 4 d 7 │ 2 8 h 8 │ 3 3 c 9 │ 3 2 b 10 │ 3 1 a julia> @orderby d @byrow :x^2 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 e 2 │ 1 6 f 3 │ 1 7 g 4 │ 1 9 i 5 │ 1 10 j 6 │ 2 4 d 7 │ 2 8 h 8 │ 3 1 a 9 │ 3 2 b 10 │ 3 3 c ``` Subexpression: @orderby d begin :x abs.(:n .- mean(:n)) end Evaluated output: 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 d 2 │ 1 6 g 3 │ 1 7 f 4 │ 1 9 j 5 │ 1 10 h 6 │ 2 4 e 7 │ 2 8 i 8 │ 3 3 b 9 │ 3 2 c 10 │ 3 1 a Expected output: 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 e 2 │ 1 6 f 3 │ 1 7 g 4 │ 1 9 i 5 │ 1 10 j 6 │ 2 4 d 7 │ 2 8 h 8 │ 3 3 c 9 │ 3 2 b 10 │ 3 1 a diff = Warning: Diff output requires color. 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 e d 2 │ 1 6 f g 3 │ 1 7 g f 4 │ 1 9 i j 5 │ 1 10 j h 6 │ 2 4 d e 7 │ 2 8 h i 8 │ 3 3 c b 9 │ 3 2 b c 10 │ 3 1 a

Check failure on line 1314 in src/macros.jl

View workflow job for this annotation

GitHub Actions / build

doctest failure in ~/work/DataFramesMeta.jl/DataFramesMeta.jl/src/macros.jl:1314-1386 ```jldoctest julia> using DataFramesMeta, Statistics julia> d = DataFrame(x = [3, 3, 3, 2, 1, 1, 1, 2, 1, 1], n = 1:10, c = ["a", "c", "b", "e", "d", "g", "f", "i", "j", "h"]); julia> @orderby(d, -:n) 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 10 h 2 │ 1 9 j 3 │ 2 8 i 4 │ 1 7 f 5 │ 1 6 g 6 │ 1 5 d 7 │ 2 4 e 8 │ 3 3 b 9 │ 3 2 c 10 │ 3 1 a julia> @orderby(d, invperm(sortperm(:c, rev = true))) 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 9 j 2 │ 2 8 i 3 │ 1 10 h 4 │ 1 6 g 5 │ 1 7 f 6 │ 2 4 e 7 │ 1 5 d 8 │ 3 2 c 9 │ 3 3 b 10 │ 3 1 a julia> @orderby d begin :x abs.(:n .- mean(:n)) end 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 e 2 │ 1 6 f 3 │ 1 7 g 4 │ 1 9 i 5 │ 1 10 j 6 │ 2 4 d 7 │ 2 8 h 8 │ 3 3 c 9 │ 3 2 b 10 │ 3 1 a julia> @orderby d @byrow :x^2 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 e 2 │ 1 6 f 3 │ 1 7 g 4 │ 1 9 i 5 │ 1 10 j 6 │ 2 4 d 7 │ 2 8 h 8 │ 3 1 a 9 │ 3 2 b 10 │ 3 3 c ``` Subexpression: @orderby d @byrow :x^2 Evaluated output: 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 d 2 │ 1 6 g 3 │ 1 7 f 4 │ 1 9 j 5 │ 1 10 h 6 │ 2 4 e 7 │ 2 8 i 8 │ 3 1 a 9 │ 3 2 c 10 │ 3 3 b Expected output: 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 e 2 │ 1 6 f 3 │ 1 7 g 4 │ 1 9 i 5 │ 1 10 j 6 │ 2 4 d 7 │ 2 8 h 8 │ 3 1 a 9 │ 3 2 b 10 │ 3 3 c diff = Warning: Diff output requires color. 10×3 DataFrame Row │ x n c │ Int64 Int64 String ─────┼────────────────────── 1 │ 1 5 e d 2 │ 1 6 f g 3 │ 1 7 g f 4 │ 1 9 i j 5 │ 1 10 j h 6 │ 2 4 d e 7 │ 2 8 h i 8 │ 3 1 a 9 │ 3 2 b c 10 │ 3 3 cb
```jldoctest
julia> using DataFramesMeta, Statistics

Expand Down Expand Up @@ -1406,7 +1406,7 @@

Use this function as an alternative to placing the `.` to broadcast row-wise operations.

### Examples

Check failure on line 1409 in src/macros.jl

View workflow job for this annotation

GitHub Actions / build

doctest failure in ~/work/DataFramesMeta.jl/DataFramesMeta.jl/src/macros.jl:1409-1446 ```jldoctest julia> using DataFramesMeta julia> df = DataFrame(x = [8,8,-8,7,7,-7], y = [-1, 1, -2, 2, -3, 3]) 6×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 8 -1 2 │ 8 1 3 │ -8 -2 4 │ 7 2 5 │ 7 -3 6 │ -7 3 julia> @rorderby df abs(:x) (:x * :y^3) Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 7 -3 2 │ -7 3 3 │ 7 2 4 │ 8 -1 5 │ 8 1 6 │ -8 -2 julia> @rorderby df :y == 2 ? -:x : :y 6×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 7 2 2 │ 7 -3 3 │ -8 -2 4 │ 8 -1 5 │ 8 1 6 │ -7 3 ``` Subexpression: @rorderby df abs(:x) (:x * :y^3) Evaluated output: 6×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 7 -3 2 │ -7 3 3 │ 7 2 4 │ 8 -1 5 │ 8 1 6 │ -8 -2 Expected output: Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 7 -3 2 │ -7 3 3 │ 7 2 4 │ 8 -1 5 │ 8 1 6 │ -8 -2 diff = Warning: Diff output requires color. 6×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 7 -3 2 │ -7 3 3 │ 7 2 4 │ 8 -1 5 │ 8 1 6 │ -8 -2
```jldoctest
julia> using DataFramesMeta

Expand Down Expand Up @@ -1931,7 +1931,7 @@
function rselect_helper(x, args...)
x, exprs, outer_flags, kw = get_df_args_kwargs(x, args...; wrap_byrow = true)

t = (fun_to_vec(ex; gensym_names=false, outer_flags=outer_flags) for ex in exprs)
t = (fun_to_vec(ex; gensym_names=false, outer_flags=outer_flags, allow_multicol=true) for ex in exprs)
quote
$select($x, $(t...); $(kw...))
end
Expand Down Expand Up @@ -1984,7 +1984,7 @@
function select!_helper(x, args...)
x, exprs, outer_flags, kw = get_df_args_kwargs(x, args...; wrap_byrow = false)

t = (fun_to_vec(ex; gensym_names = false, outer_flags = outer_flags) for ex in exprs)
t = (fun_to_vec(ex; gensym_names = false, outer_flags = outer_flags, allow_multicol=true) for ex in exprs)
quote
$select!($x, $(t...); $(kw...))
end
Expand Down Expand Up @@ -2120,7 +2120,7 @@
function rselect!_helper(x, args...)
x, exprs, outer_flags, kw = get_df_args_kwargs(x, args...; wrap_byrow = true)

t = (fun_to_vec(ex; gensym_names=false, outer_flags=outer_flags) for ex in exprs)
t = (fun_to_vec(ex; gensym_names=false, outer_flags=outer_flags, allow_multicol=true) for ex in exprs)
quote
$select!($x, $(t...); $(kw...))
end
Expand Down Expand Up @@ -2529,7 +2529,7 @@
should be used instead of `@distinct(df, [:x,:y])` to avoid unexpected behaviors.

### Examples

Check failure on line 2532 in src/macros.jl

View workflow job for this annotation

GitHub Actions / build

doctest failure in ~/work/DataFramesMeta.jl/DataFramesMeta.jl/src/macros.jl:2532-2552 ```jldoctest julia> using DataFramesMeta; julia> df = DataFrame(x = 1:10, y = 10:-1:1); julia> @distinct(df, :x .+ :y) 1×2 DataFrame Row │ x y │ Int64 Int64 ─────┼─────────────── 1 │ 1 10 julia> @distinct df begin :x .+ :y end 1×2 DataFrame Row │ x y │ Int64 Int64 ─────┼─────────────── 1 │ 1 10 ``` Subexpression: @distinct(df, :x .+ :y) Evaluated output: 1×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 1 10 Expected output: 1×2 DataFrame Row │ x y │ Int64 Int64 ─────┼─────────────── 1 │ 1 10 diff = Warning: Diff output requires color. 1×2 DataFrame Row │ x y y │ Int64 Int64 ─────┼─────────────── Int64 ─────┼────────────── 1 │ 1 1 10

Check failure on line 2532 in src/macros.jl

View workflow job for this annotation

GitHub Actions / build

doctest failure in ~/work/DataFramesMeta.jl/DataFramesMeta.jl/src/macros.jl:2532-2552 ```jldoctest julia> using DataFramesMeta; julia> df = DataFrame(x = 1:10, y = 10:-1:1); julia> @distinct(df, :x .+ :y) 1×2 DataFrame Row │ x y │ Int64 Int64 ─────┼─────────────── 1 │ 1 10 julia> @distinct df begin :x .+ :y end 1×2 DataFrame Row │ x y │ Int64 Int64 ─────┼─────────────── 1 │ 1 10 ``` Subexpression: @distinct df begin :x .+ :y end Evaluated output: 1×2 DataFrame Row │ x y │ Int64 Int64 ─────┼────────────── 1 │ 1 10 Expected output: 1×2 DataFrame Row │ x y │ Int64 Int64 ─────┼─────────────── 1 │ 1 10 diff = Warning: Diff output requires color. 1×2 DataFrame Row │ x y y │ Int64 Int64 ─────┼─────────────── Int64 ─────┼────────────── 1 │ 1 1 10
```jldoctest
julia> using DataFramesMeta;

Expand Down
61 changes: 61 additions & 0 deletions test/multicol.jl
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,67 @@ df = DataFrame(A = 1, AA = 2, B = 3)
@test t == DataFrame(AA = 2, B = 3)
end

@testset "rselect_multi" begin
df = DataFrame(A = 1, AA = 2, B = 3)

t = @rselect df Not(:A)
@test t == DataFrame(AA = 2, B = 3)

t = @rselect df All()
@test t == DataFrame(A = 1, AA = 2, B = 3)

t = @rselect df Cols(r"A")
@test t == DataFrame(A = 1, AA = 2)

t = @rselect df Between(:AA, :B)
@test t == DataFrame(AA = 2, B = 3)
end

@testset "select!_multi" begin
df = DataFrame(A = 1, AA = 2, B = 3)

@select! df Not(:A)
@test df == DataFrame(AA = 2, B = 3)

df = DataFrame(A = 1, AA = 2, B = 3)

@select! df All()
@test df == DataFrame(A = 1, AA = 2, B = 3)

df = DataFrame(A = 1, AA = 2, B = 3)

@select! df Cols(r"A")
@test df == DataFrame(A = 1, AA = 2)

df = DataFrame(A = 1, AA = 2, B = 3)

@select! df Between(:AA, :B)
@test df == DataFrame(AA = 2, B = 3)
end

@testset "rselect!_multi" begin
df = DataFrame(A = 1, AA = 2, B = 3)

@rselect! df Not(:A)
@test df == DataFrame(AA = 2, B = 3)

df = DataFrame(A = 1, AA = 2, B = 3)

@rselect! df All()
@test df == DataFrame(A = 1, AA = 2, B = 3)

df = DataFrame(A = 1, AA = 2, B = 3)

@rselect! df Cols(r"A")
@test df == DataFrame(A = 1, AA = 2)

df = DataFrame(A = 1, AA = 2, B = 3)

@rselect! df Between(:AA, :B)
@test df == DataFrame(AA = 2, B = 3)
end


@testset "othermacros_multi" begin
df = DataFrame(A = 1, AA = 2, B = 3)

Expand Down
Loading