replace_missing and missing_if

docs/examples/UserGuide/fill_missing.jl

@@ -34,3 +34,18 @@ end
     @fill_missing(a, "down")
 end
 
+# ## `replace_missing()`
+# The `replace_missing` function facilitates the replacement of `missing` values with a specified replacement. 
+
+@chain df begin
+    @mutate(b = replace_missing(b, 2))
+end 
+
+# ## `missing_if()`
+# The `missing_if` function is used to introduce `missing` values under specific conditions. 
+
+@chain df begin
+    @mutate(b = missing_if(b, 5))
+end 
+
+# Both `missing_if` and `replace_missing` are not type specifc.

src/TidierData.jl

@@ -16,7 +16,7 @@ using Reexport
 @reexport using ShiftedArrays: lag, lead
 
 export TidierData_set, across, desc, n, row_number, everything, starts_with, ends_with, matches, if_else, case_when, ntile, 
-      as_float, as_integer, as_string, is_float, is_integer, is_string, @select, @transmute, @rename, @mutate, @summarize, @summarise, @filter,
+      as_float, as_integer, as_string, is_float, is_integer, is_string, missing_if, replace_missing, @select, @transmute, @rename, @mutate, @summarize, @summarise, @filter,
       @group_by, @ungroup, @slice, @arrange, @distinct, @pull, @left_join, @right_join, @inner_join, @full_join,
       @pivot_wider, @pivot_longer, @bind_rows, @bind_cols, @clean_names, @count, @tally, @drop_missing, @glimpse, @separate,
       @unite, @summary, @fill_missing, @slice_sample

src/docstrings.jl

@@ -2441,3 +2441,59 @@ julia> @chain df begin
    5 │    25      5     15
 ```
 """
+
+const docstring_missing_if =
+"""
+    missing_if(x, value)
+
+Replace a specific `value` with `missing` in `x`.
+
+## Arguments
+- `x`: The input value which can be of any type. If `x` is already `missing` or equals `value`, the function will return `missing`. Otherwise, it returns `x` unaltered.
+- `value`: The specific value to be checked against. 
+
+## Examples
+```jldoctest
+julia> df = DataFrame(a = [1, missing, 3, 4], b = ["apple", "apple", "banana", "cherry"])
+
+julia> @chain df begin
+       @mutate(a = missing_if(a, 4), b = missing_if(b, "apple"))
+       end
+4×2 DataFrame
+ Row │ a        b       
+     │ Int64?   String? 
+─────┼──────────────────
+   1 │       1  missing 
+   2 │ missing  missing 
+   3 │       3  banana
+   4 │ missing  cherry
+```
+"""
+
+const docstring_replace_missing =
+"""
+    replace_missing(x, replacement)
+
+Replace `missing` values in `x` with a specified `replacement` value.
+
+# Arguments
+- `x`: The input value which can be of any type. If `x` is `missing`, the function will return `replacement`. Otherwise, it returns `x` unaltered.
+- `replacement`: The value to replace `missing` with in `x`.
+
+# Examples
+```jldoctest
+julia> df = DataFrame(a = [1, missing, 3, 4], b = [4, 5, missing, 8]);
+
+julia> @chain df begin
+       @mutate(a = replace_missing(a, 100), b = replace_missing(b, 35))
+       end
+4×2 DataFrame
+ Row │ a      b     
+     │ Int64  Int64 
+─────┼──────────────
+   1 │     1      4
+   2 │   100      5
+   3 │     3     35
+   4 │     4      8
+```
+"""

src/missings.jl

@@ -115,4 +115,14 @@ macro fill_missing(df, args...)
           fill_missing($(esc(df)), [$(cols_quoted...)], $method)
       end
   end
-end
+end
+
+"""
+$docstring_missing_if
+"""
+missing_if(x, value) = ismissing(x) ? x : (x == value ? missing : x)
+
+"""
+$docstring_replace_missing
+"""
+replace_missing(x, replacement) = ismissing(x) ? replacement : x

src/separate_unite.jl

@@ -61,14 +61,24 @@ end
 $docstring_unite
 """
 macro unite(df, new_col, from_cols, sep)
-    new_col = QuoteNode(new_col)
-
-    if @capture(from_cols, (args__,))
-    elseif @capture(from_cols, [args__])
+    new_col_quoted = QuoteNode(new_col)
+    interpolated_from_cols, _, _ = parse_interpolation(from_cols)
+
+    if @capture(interpolated_from_cols, (args__,)) || @capture(interpolated_from_cols, [args__])
+        args = QuoteNode.(args)
+        from_cols_expr = :[$(args...)]
+    else
+        from_cols_expr = quote
+            if typeof($interpolated_from_cols) <: Tuple
+                collect(Symbol.($interpolated_from_cols))
+
+            else
+                $interpolated_from_cols
+            end
+        end
     end
-
-    args = QuoteNode.(args)
-    var_expr = quote
-         unite($(esc(df)), $new_col, [$(args...)], $sep)
+
+    return quote
+        unite($(esc(df)), $new_col_quoted, $(from_cols_expr), $(esc(sep)))
     end
-end
+end