Adds support for tuples and vectors as arguments to select multiple columns. Prefixing tuples/vectors with a - or ! will exclude the selected columns. Updated docs to use Cols() for column interpolation.

Author: kdpsingh

NEWS.md

@@ -1,5 +1,10 @@
 # TidierData.jl updates
 
+## v0.16.1 - 2024-06-09
+- Adds support for tuples and vectors as arguments to select multiple columns. Prefixing tuples/vectors with a `-` or `!` will exclude the selected columns.
+- The `:` selector from Julia is now available and equivalent to `everything()`
+- `@pivot_longer()` now pivots all columns if no column selectors are provided
+
 ## v0.16.0 - 2024-06-07
 - `unique()`, `mad()`, and `iqr()` are no longer auto-vectorized
 - Bugfix: `@ungroup()` now preserves row-ordering (and is faster)

Project.toml

@@ -1,7 +1,7 @@
 name = "TidierData"
 uuid = "fe2206b3-d496-4ee9-a338-6a095c4ece80"
 authors = ["Karandeep Singh"]
-version = "0.16.0"
+version = "0.16.1"
 
 [deps]
 Chain = "8be319e6-bccf-4806-a6f7-6fae938471bc"

docs/examples/UserGuide/interpolation.jl

@@ -4,7 +4,7 @@
 
 # Note: You can only interpolate values from variables in the parent environment. If you would like to interpolate column names, you have two options: you can either use `across()` or you can use `@aside` with `@pull()` to create variables in the parent environment containing the values of those columns which can then be accessed using interpolatino.
 
-# myvar = :b`, `myvar = (:a, :b)`, and `myvar = [:a, :b]` all refer to *columns* with those names. On the other hand, `myvar = "b"`, `myvar = ("a", "b")` and `myvar = ["a", "b"]` will interpolate those *values*. See below for examples.
+# myvar = :b` and `myvar = Cols(:a, :b)` both refer to *columns* with those names. On the other hand, `myvar = "b"`, `myvar = ("a", "b")` and `myvar = ["a", "b"]` will interpolate the *values*. If you intend to interpolate column names, the preferred way is to use `Cols()` as in the examples below.
 
 using TidierData
 
@@ -20,9 +20,19 @@ myvar = :b
   @select(!!myvar)
 end
 
-# ## Select multiple variables (vector of symbols)
+# ## Select multiple variables
 
-myvars = [:a, :b]
+# You can also use a vector as in `[:a, :b]`, but `Cols()` is preferred because it lets you mix and match numbers.
+
+myvars = Cols(:a, :b)
+
+@chain df begin
+  @select(!!myvars)
+end
+
+# This is the same as this...
+
+myvars = Cols(:a, 2)
 
 @chain df begin
   @select(!!myvars)
@@ -86,7 +96,7 @@ end
 
 # ## Summarize across multiple variables
 
-myvars = [:b, :c]
+myvars = Cols(:b, :c)
 
 @chain df begin
   @summarize(across(!!myvars, (mean, minimum, maximum)))
@@ -103,7 +113,9 @@ end
 
 # ## Group by multiple interpolated variables
 
-myvars = [:a, :b]
+# Once again, you can mix and match column selectors within `Cols()`
+
+myvars = Cols(:a, 2)
 
 @chain df begin
   @group_by(!!myvars)

src/TidierData.jl

@@ -449,6 +449,7 @@ macro group_by(df, exprs...)
 
   tidy_exprs = parse_tidy.(tidy_exprs)
   grouping_exprs = parse_group_by.(exprs)
+  grouping_exprs = parse_tidy.(grouping_exprs)
 
   df_expr = quote
     local any_expressions = any(typeof.($tidy_exprs) .!= QuoteNode)

src/docstrings.jl

@@ -260,6 +260,28 @@ julia> @chain df @select(!(a:b))
    4 │    14
    5 │    15
 
+julia> @chain df @select(-(a, b))
+5×1 DataFrame
+ Row │ c     
+     │ Int64 
+─────┼───────
+   1 │    11
+   2 │    12
+   3 │    13
+   4 │    14
+   5 │    15
+
+julia> @chain df @select(!(a, b))
+5×1 DataFrame
+ Row │ c     
+     │ Int64 
+─────┼───────
+   1 │    11
+   2 │    12
+   3 │    13
+   4 │    14
+   5 │    15
+
 julia> @chain df begin
          @select(contains("b"), starts_with("c"))
        end
@@ -667,6 +689,34 @@ julia> @chain df begin
    3 │ C         3.0
    4 │ D         4.0
    5 │ E         5.0
+
+julia> @chain df begin
+         @group_by(-(b, c)) # same as `a`
+         @summarize(b = mean(b))
+       end
+5×2 DataFrame
+ Row │ a     b       
+     │ Char  Float64 
+─────┼───────────────
+   1 │ a         1.0
+   2 │ b         2.0
+   3 │ c         3.0
+   4 │ d         4.0
+   5 │ e         5.0
+
+julia> @chain df begin
+         @group_by(!(b, c)) # same as `a`
+         @summarize(b = mean(b))
+       end
+5×2 DataFrame
+ Row │ a     b       
+     │ Char  Float64 
+─────┼───────────────
+   1 │ a         1.0
+   2 │ b         2.0
+   3 │ c         3.0
+   4 │ d         4.0
+   5 │ e         5.0
 ```
 """
 

src/parsing.jl

@@ -1,5 +1,5 @@
 # Not exported
-function parse_tidy(tidy_expr::Union{Expr,Symbol,Number}; # Can be symbol or expression
+function parse_tidy(tidy_expr::Union{Expr,Symbol,Number, QuoteNode}; # Can be symbol or expression
                     autovec::Bool=true, subset::Bool=false, from_across::Bool=false,
                     from_slice::Bool = false)
   if @capture(tidy_expr, across(vars_, funcs_))
@@ -32,10 +32,6 @@ function parse_tidy(tidy_expr::Union{Expr,Symbol,Number}; # Can be symbol or exp
       endindex = QuoteNode(endindex)
     end
     return :(Between($startindex, $endindex))
-  elseif @capture(tidy_expr, names_vect)
-    return Symbol.(names.args)
-  elseif @capture(tidy_expr, -names_vect)
-    return Not(Symbol.(names.args))
   elseif @capture(tidy_expr, (lhs_ = fn_(args__)) | (lhs_ = fn_.(args__)))
     if length(args) == 0
       lhs = QuoteNode(lhs)
@@ -75,13 +71,28 @@ function parse_tidy(tidy_expr::Union{Expr,Symbol,Number}; # Can be symbol or exp
     end
   elseif @capture(tidy_expr, !var_Number)
     return :(Not($var))
+  elseif @capture(tidy_expr, (tuple__,))
+    tuple = parse_tidy.(tuple)
+    return :(Cols($(tuple...)))
+  elseif @capture(tidy_expr, [vec__])
+    vec = parse_tidy.(vec)
+    return :(Cols($(vec...)))
+  elseif @capture(tidy_expr, -[vec__])
+    vec = parse_tidy.(vec)
+    return :(Not(Cols($(vec...)))) # can simpify to Not($(tuple...)) in DataFrames 1.6+
+  elseif @capture(tidy_expr, ![vec__])
+    vec = parse_tidy.(vec)
+    return :(Not(Cols($(vec...)))) # can simpify to Not($(tuple...)) in DataFrames 1.6+
   elseif !subset & @capture(tidy_expr, -fn_(args__)) # negated selection helpers
     return :(Cols(!($(esc(fn))($(args...))))) # change the `-` to a `!` and return
   elseif !subset & @capture(tidy_expr, fn_(args__)) # selection helpers
     if from_across || fn == :Cols # fn == :Cols is to deal with interpolated columns
       return tidy_expr
     elseif fn == :where
       return :(Cols(all.(broadcast($(esc(args...)), eachcol(DataFrame(df_copy))))))
+    elseif fn == :- || fn == :! # for negated selection as in -(A, B), which is internally represnted as function
+      args = parse_tidy.(args)
+      return :(Not(Cols($(args...)))) # can simpify to Not($(tuple...)) in DataFrames 1.6+
     else
       return :(Cols($(esc(tidy_expr))))
     end

src/pivots.jl

@@ -42,7 +42,7 @@ $docstring_pivot_longer
 """
 macro pivot_longer(df, exprs...)
     if length(exprs) == 0
-        exprs = (:(:),)
+        exprs = (:(everything()),)
     end
     exprs = parse_blocks(exprs...)