Error messages for rename_columns and Vector.duplicates (#9917)

- Improve error message for `rename_columns`.
- Add `length` to `Set` and `Map`.
- Add `duplicates` to `Vector` (and `Array`).

![image](https://github.com/enso-org/enso/assets/4699705/623df253-52e8-4bdc-a69c-ac8dc3ca594e)

CHANGELOG.md

@@ -660,6 +660,7 @@
 - [Added ability to write to Data Links.][9750]
 - [Added `Vector.build_multiple`, and better for support for errors and warnings
   inside `Vector.build` and `Vector.build_multiple`.][9766]
+- [Added `Vector.duplicates`.][9917]
 
 [debug-shortcuts]:
   https://github.com/enso-org/enso/blob/develop/app/gui/docs/product/shortcuts.md#debug
@@ -968,6 +969,7 @@
 [9577]: https://github.com/enso-org/enso/pull/9577
 [9750]: https://github.com/enso-org/enso/pull/9750
 [9766]: https://github.com/enso-org/enso/pull/9766
+[9917]: https://github.com/enso-org/enso/pull/9917
 
 #### Enso Compiler
 

distribution/lib/Standard/Base/0.0.0-dev/src/Data/Array.enso

@@ -343,6 +343,26 @@ type Array
     distinct : (Any -> Any) -> Vector Any
     distinct self (on = x->x) = Array_Like_Helpers.distinct self on
 
+    ## ALIAS duplicates
+       GROUP Selections
+       ICON preparation
+       Returns only non-unique elements within the array.
+
+       Arguments:
+       - on: A projection from the element type to the value of that element
+             which determines the uniqueness criteria.
+
+       The returned duplicate elements are kept in the same order as the
+       first duplicate appeared in the input.
+
+       > Example
+         Removing repeating entries.
+
+             [1, 3, 1, 2, 2, 1].to_array . duplicates == [1, 2].to_array
+    duplicates : (Any -> Any) -> Vector Any
+    duplicates self (on = x->x) =
+        Array_Like_Helpers.duplicates self on
+
     ## ICON dataframe_map_column
        Applies a function to each element of the array, returning the `Vector` of
        results.

distribution/lib/Standard/Base/0.0.0-dev/src/Data/Map.enso

@@ -118,12 +118,17 @@ type Map key value
     not_empty : Boolean
     not_empty self = self.is_empty.not
 
-    ## GROUP Metadata
-       ICON metadata
+    ## ICON metadata
        Returns the number of entries in this map.
     size : Integer
     size self = @Builtin_Method "Map.size"
 
+    ## GROUP Metadata
+       ICON metadata
+       Returns the number of entries in this map.
+    length : Integer
+    length self = self.size
+
     ## GROUP Calculations
        ICON row_add
        Inserts a key-value mapping into this map, overriding any existing

distribution/lib/Standard/Base/0.0.0-dev/src/Data/Set.enso

@@ -47,12 +47,17 @@ type Set
     to_vector : Vector
     to_vector self = self.underlying_map.keys
 
-    ## GROUP Metadata
-       ICON metadata
+    ## ICON metadata
        Returns the number of elements in this set.
     size : Integer
     size self = self.underlying_map.size
 
+    ## GROUP Metadata
+       ICON metadata
+       Returns the number of elements in this set.
+    length : Integer
+    length self = self.size
+
     ## GROUP Logical
        ICON metadata
        Checks if the set is empty.

distribution/lib/Standard/Base/0.0.0-dev/src/Data/Vector.enso

@@ -1243,6 +1243,26 @@ type Vector a
     distinct self (on = x->x) =
         Array_Like_Helpers.distinct self on
 
+    ## ALIAS duplicates
+       GROUP Selections
+       ICON preparation
+       Returns only non-unique elements within the vector.
+
+       Arguments:
+       - on: A projection from the element type to the value of that element
+             which determines the uniqueness criteria.
+
+       The returned duplicate elements are kept in the same order as the
+       first duplicate appeared in the input.
+
+       > Example
+         Removing repeating entries.
+
+             [1, 3, 1, 2, 2, 1] . duplicates == [1, 2]
+    duplicates : (Any -> Any) -> Vector Any
+    duplicates self (on = x->x) =
+        Array_Like_Helpers.duplicates self on
+
     ## ICON convert
        Returns the vector as a `Vector`.
     to_vector : Vector

distribution/lib/Standard/Base/0.0.0-dev/src/Internal/Array_Like_Helpers.enso

@@ -164,6 +164,13 @@ distinct vector on =
                     builder.append item
                     existing.insert key True
 
+duplicates vector on = Vector.build builder->
+    vector.fold Map.empty current-> item->
+        key = on item
+        count = current.get key 0
+        if count == 1 then builder.append item
+        current.insert key count+1
+
 take vector range = case range of
     ## We are using a specialized implementation for `take Sample`, because
        the default implementation (which needs to be generic for any

distribution/lib/Standard/Table/0.0.0-dev/src/Internal/Table_Helpers.enso

@@ -311,10 +311,18 @@ rename_columns (naming_helper : Column_Naming_Helper) (internal_columns:Vector)
     is_vec_pairs = mapping.is_a Vector && mapping.length > 0 && (mapping.first.is_a Text . not)
     case is_vec_pairs of
         True ->
-            ## Attempt to treat as Map
-            map = Map.from_vector mapping
-            if map.is_error then Error.throw (Illegal_Argument.Error "A mapping Vector must be either a list of names or a list of pairs (old name to new name).") else
-                rename_columns naming_helper internal_columns map case_sensitivity error_on_missing_columns on_problems
+            ## Check all pairs (Integer | Text | Regex => Text )
+            is_valid_row r = r.is_a Vector || r.is_a Pair
+            is_valid_key k = k.is_a Integer || k.is_a Text || k.is_a Regex
+            all_pairs = mapping.all p-> (is_valid_row p) && p.length == 2 && (is_valid_key p.first) && p.second.is_a Text
+            if all_pairs.not then Error.throw (Illegal_Argument.Error "mapping is not a Vector of old name to new name.") else
+                ## Attempt to treat as Map
+                map = Map.from_vector mapping error_on_duplicates=False
+                if map.length == mapping.length then rename_columns naming_helper internal_columns map case_sensitivity error_on_missing_columns on_problems else
+                    duplicates = mapping.duplicates on=_.first . map p->p.first.to_text
+                    duplicate_text = if duplicates.length < 5 then duplicates.to_vector . join ", " else
+                        duplicates.take 3 . to_vector . join ", " + (", ... " + (duplicates.length - 3).to_text + " others")
+                    Error.throw (Illegal_Argument.Error "duplicate old name mappings ("+duplicate_text+").")
         False ->
             unique = naming_helper.create_unique_name_strategy
             problem_builder = Problem_Builder.new error_on_missing_columns=error_on_missing_columns

test/Base_Tests/src/Data/Vector_Spec.enso

@@ -857,6 +857,13 @@ type_spec suite_builder name alter = suite_builder.group name group_builder->
         alter [1, 1.0, 2, 2.0] . distinct . should_equal [1, 2]
         alter [] . distinct . should_equal []
 
+    group_builder.specify "should return a vector containing only duplicate elements" <|
+        alter [1, 3, 1, 2, 2, 1] . duplicates . should_equal [1, 2]
+        alter ["a", "a", "a"] . duplicates . should_equal ["a"]
+        alter ['ś', 's', 's\u0301'] . duplicates . should_equal ['s\u0301']
+        alter [1, 1.0, 2, 2.0] . duplicates . should_equal [1.0, 2.0]
+        alter [] . duplicates . should_equal []
+
     group_builder.specify "should be able to handle distinct on different primitive values" <|
         alter [1, "a"] . distinct . should_equal [1, "a"]
         alter ["a", 1] . distinct . should_equal ["a", 1]

test/Table_Tests/src/Common_Table_Operations/Select_Columns_Spec.enso

@@ -1,4 +1,5 @@
 from Standard.Base import all
+import Standard.Base.Errors.Illegal_Argument.Illegal_Argument
 
 from Standard.Table import Position, Value_Type, Bits
 from Standard.Table.Errors import all
@@ -503,6 +504,29 @@ add_specs suite_builder setup =
             expect_column_names ["lpha", "beta", "gamma", "delta"] <|
                 data.table.rename_columns map
 
+        group_builder.specify "should report invalid input map nicely" <|
+            test_invalid_map map =
+                result = data.table.rename_columns map
+                result.should_fail_with Illegal_Argument
+                result.catch Any . message . should_equal "mapping is not a Vector of old name to new name."
+
+            test_invalid_map [["Alpha"]]
+            test_invalid_map [["Alpha", 1]]
+            test_invalid_map [["Alpha", "Beta", "Delta"]]
+            test_invalid_map [[True, "Beta"]]
+
+        group_builder.specify "should report duplicates in input map nicely" <|
+            test_duplicate_names map message =
+                result = data.table.rename_columns map
+                result.should_fail_with Illegal_Argument
+                result.catch Any . message . should_equal message
+
+            test_duplicate_names [["Alpha", "1"], ["Alpha", "2"]] "duplicate old name mappings (Alpha)."
+            test_duplicate_names [["Alpha", "1"], ["Beta", "2"], ["Gamma", "3"], ["Beta", "4"], ["Alpha", "5"]] "duplicate old name mappings (Beta, Alpha)."
+            test_duplicate_names [["Alpha", "1"], ["Alpha", "2"], ["Alpha", "3"]] "duplicate old name mappings (Alpha)."
+            test_duplicate_names [["Alpha", "1"], ["Beta", "2"], ["Gamma", "3"], ["Beta", "4"], ["Alpha", "5"], ["Gamma","6"], ["Delta","7"], ["Delta","8"]] "duplicate old name mappings (Beta, Alpha, Gamma, Delta)."
+            test_duplicate_names [["Alpha", "1"], ["Beta", "2"], ["Gamma", "3"], ["Beta", "4"], ["Alpha", "5"], ["Gamma","6"], ["Delta","7"], ["Delta","8"], ["Echo","9"], ["Echo","10"]] "duplicate old name mappings (Beta, Alpha, Gamma, ... 2 others)."
+
         group_builder.specify "should correctly handle problems: unmatched names" <|
             weird_name = '.*?-!@#!"'
             map = Map.from_vector [["alpha", "FirstColumn"], ["omicron", "Another"], [weird_name, "Fixed"]]