Implement Table.add_group_number, with operations Unique and Equal_Count

CHANGELOG.md

@@ -120,6 +120,7 @@
 - [Added `Table.input` allowing creation of typed tables from vectors of data,
   including auto parsing text columns.][11562]
 - [Enhance Managed_Resource to allow implementation of in-memory caches][11577]
+- [Added `add_group_number` to the in-memory database.[11818]
 - [The reload button clears the HTTP cache.][11673]
 
 [11235]: https://github.com/enso-org/enso/pull/11235
@@ -129,6 +130,7 @@
 [11490]: https://github.com/enso-org/enso/pull/11490
 [11562]: https://github.com/enso-org/enso/pull/11562
 [11577]: https://github.com/enso-org/enso/pull/11577
+[11818]: https://github.com/enso-org/enso/pull/11818
 [11673]: https://github.com/enso-org/enso/pull/11673
 
 #### Enso Language & Runtime

distribution/lib/Standard/Database/0.0.0-dev/src/DB_Table.enso

@@ -25,6 +25,7 @@ import Standard.Table.Columns_To_Add.Columns_To_Add
 import Standard.Table.Columns_To_Keep.Columns_To_Keep
 import Standard.Table.Expression.Expression
 import Standard.Table.Expression.Expression_Error
+import Standard.Table.Grouping_Method.Grouping_Method
 import Standard.Table.Internal.Add_Row_Number
 import Standard.Table.Internal.Column_Naming_Helper.Column_Naming_Helper
 import Standard.Table.Internal.Constant_Column.Constant_Column
@@ -926,6 +927,99 @@ type DB_Table
                 updated_table = renamed_table.updated_columns (renamed_table.internal_columns + [new_column])
                 updated_table.as_subquery
 
+    ## PRIVATE add group column, group id, bucket, tile
+       GROUP Standard.Base.Values
+       ICON column_add
+       Adds a new column to the table enumerating groups of rows, assigning each
+       row to one group number. All rows in each group will get the same number.
+
+       Arguments:
+       - grouping_method: Specifies how to group the rows; see "Grouping
+         Methods", below.
+       - name: The name of the new column. Defaults to "Group".
+       - from: The starting value for the enumeration. Defaults to 0.
+       - step: The amount to increment the enumeration by. Defaults to 1.
+
+       ? Grouping Methods
+
+         The following grouping methods are supported:
+         - `Unique`: Group rows by the specified columns.
+         - Equal_Count: Create the specified number of groups with the same
+           number of rows in each group (except possibly the last one).
+
+       ? Ordering of rows
+
+         Note that the ordering of rows from the original table is preserved in
+         all cases. The grouping and ordering settings can affect how the group
+         numbers are assigned, depending on the grouping method.  The order of
+         the rows itself is not changed by this operation.
+
+       ! Error Conditions
+
+         - If the columns specified in `group_by` or `order_by` are not present
+           in the table, a `Missing_Input_Columns` error is raised.
+         - If the column with the same name as provided `name` already exists,
+           a `Duplicate_Output_Column_Names` problem is reported and the
+           existing column is renamed to avoid the clash.
+         - If grouping on floating point numbers, a `Floating_Point_Equality`
+           problem is reported.
+
+       > Example
+         Assign group numbers based on unique values of the first two columns.
+
+            ## table:
+                x | y | z
+               ---+---+---
+                1 | 0 | 2
+                0 | 1 | 0
+                1 | 2 | 0
+                0 | 1 | 1
+                1 | 0 | 1
+                1 | 2 | 1
+            table = table_builder [['x', [1, 0, 1, 0, 1, 1]], ['y', [0, 1, 2, 1, 0, 2]], ['z' [2, 0, 0, 1, 1, 1]]]
+            table2 = table.add_group_number (..Unique group_by=['x', 'y']) "g"
+            table2.at 'g' . to_vector
+            # => [0, 1, 2, 1, 0, 2]
+            ## table2:
+                x | y | z | g
+               ---+---+---+---
+                1 | 0 | 2 | 0
+                0 | 1 | 0 | 1
+                1 | 2 | 0 | 2
+                0 | 1 | 1 | 1
+                1 | 0 | 1 | 2
+                1 | 2 | 1 | 0
+
+       > Example
+         Divide rows into three groups.
+            ## table:
+                x | y
+               ---+---
+                1 | 5
+                2 | 4
+                3 | 3
+                4 | 2
+                5 | 1
+            table = table_builder [['x', [1, 2, 3, 4, 5]], ['y', [5, 4, 3, 2, 1]]]
+            table2 = tabble.add_group_number (..Equal_Count 3) "g"
+            table2.at 'g' . to_vector
+            # => [0, 0, 1, 1, 2]
+            ## table2:
+                x | y | g
+               ---+---+---
+                1 | 5 | 0
+                2 | 4 | 0
+                3 | 3 | 1
+                4 | 2 | 1
+                5 | 1 | 2
+    @name (Widget.Text_Input display=..Always)
+    @from (Widget.Numeric_Input display=..Always)
+    @group_by (Widget_Helpers.make_column_name_multi_selector display=..When_Modified)
+    @order_by (Widget_Helpers.make_order_by_selector display=..When_Modified)
+    add_group_number self (grouping_method:Grouping_Method=..Unique) (name:Text="Group") (from:Integer=0) (step:Integer=1) (on_problems:Problem_Behavior=..Report_Warning) -> Table =
+        _ = [grouping_method, name, from, step, on_problems]
+        Error.throw (Unsupported_Database_Operation.Error "add_group_number")
+
 
     ## ALIAS order_by
        GROUP Standard.Base.Selections

distribution/lib/Standard/Database/0.0.0-dev/src/Feature.enso

@@ -31,6 +31,9 @@ type Feature
     ## PRIVATE
        Catch all for tests that haven't yet been categorized correctly or use multiple features.
     Integration_Tests
+    ## PRIVATE
+       add a group number column to a table.
+    Add_Group_Number
     ## PRIVATE
        add a row number column to a table.
     Add_Row_Number

distribution/lib/Standard/Table/0.0.0-dev/src/Grouping_Method.enso

@@ -0,0 +1,24 @@
+from Standard.Base import all
+import Standard.Base.Errors.Common.Missing_Argument
+
+polyglot java import org.enso.table.operations.AddGroupNumber
+
+## Specifies a method for grouping rows in `add_group_number`.
+type Grouping_Method
+    ## Group rows by the specified columns.
+
+       Arguments:
+       - on: Rows that have the same values for these columns will be grouped
+         together. At least one column must be specified.
+    Unique (on:(Vector | Text | Integer | Regex)=(Missing_Argument.throw "on"))
+
+    ## Create the specified number of groups with the same number of rows in
+       each group (except possibly the last one).
+
+       Arguments
+       - group_count: The number of groups to divide the table into.
+       - order_by: (Optional.) Specifies the order in which rows should be
+         assigned to groups. Only affects the assignment of group numbers, not
+         the ordering of the output rows. Defaults to the order of the rows in
+         the table.
+    Equal_Count (group_count:Integer=(Missing_Argument.throw "group_count")) (order_by:(Vector | Text)=[])

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

@@ -0,0 +1,50 @@
+private
+
+from Standard.Base import all
+import Standard.Base.Errors.Common.Unsupported_Argument_Types
+import Standard.Base.Errors.Illegal_Argument.Illegal_Argument
+
+import project.Column.Column
+import project.Grouping_Method.Grouping_Method
+import project.Internal.Java_Problems
+import project.Internal.Problem_Builder.Problem_Builder
+import project.Internal.Table_Helpers
+import project.Set_Mode.Set_Mode
+import project.Table.Table
+from project.Internal.Add_Row_Number import rename_columns_if_needed
+
+polyglot java import java.lang.ArithmeticException
+polyglot java import org.enso.table.operations.AddGroupNumber
+
+add_group_number (table:Table) (grouping_method:Grouping_Method) (name:Text) (from:Integer) (step:Integer) (on_problems:Problem_Behavior=..Report_Warning) -> Table =
+    problem_builder = Problem_Builder.new error_on_missing_columns=True
+
+    handle_arithmetic_exception _ =
+        Error.throw (Illegal_Argument.Error "The row number has exceeded the 64-bit integer range. BigInteger numbering is currently not supported. Please use a smaller start/step.")
+
+    Panic.catch ArithmeticException handler=handle_arithmetic_exception <| Panic.catch Unsupported_Argument_Types handler=handle_arithmetic_exception <|
+        Java_Problems.with_problem_aggregator on_problems java_problem_aggregator->
+            new_storage = case grouping_method of
+                Grouping_Method.Unique group_by ->
+                    _illegal_if group_by.is_empty "..Unique requires a non-empty 'group_by'" <|
+                        grouping = _prepare_group_by table problem_builder group_by
+                        AddGroupNumber.numberGroupsUnique table.row_count from step grouping java_problem_aggregator
+                Grouping_Method.Equal_Count group_count order_by ->
+                    _illegal_if (group_count < 1) "group_count must be at least 1" <|
+                        ordering = _prepare_ordering table problem_builder order_by
+                        AddGroupNumber.numberGroupsEqualCount table.row_count group_count from step (ordering.at 0) (ordering.at 1) java_problem_aggregator
+            new_column = Column.from_storage name new_storage
+            renamed_table = rename_columns_if_needed table name on_problems Table.new
+            problem_builder.attach_problems_before on_problems <| 
+                renamed_table.set new_column name set_mode=Set_Mode.Add
+
+_prepare_group_by table problem_builder group_by =
+    table.columns_helper.select_columns_helper group_by Case_Sensitivity.Default True problem_builder . map c->c.java_column
+
+_prepare_ordering table problem_builder order_by =
+    ordering = Table_Helpers.resolve_order_by table.columns order_by problem_builder
+    ordering_columns = ordering.map c->c.column.java_column
+    directions = ordering.map c->c.associated_selector.direction.to_sign
+    [ordering_columns, directions]
+
+_illegal_if b msg ~cont = if b then Error.throw (Illegal_Argument.Error msg) else cont

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

@@ -168,6 +168,17 @@ make_join_condition_selector table display:Display=..Always cache=Nothing =
     item_editor = Single_Choice display=display values=names
     Vector_Editor item_editor=item_editor item_default="(..Equals "+table.column_names.first.pretty+")" display=display
 
+## PRIVATE
+make_grouping_method_selector table:Table display:Display=..Always -> Widget =
+    column_selector = make_column_name_selector table display=Display.Always
+    columns_selector = Vector_Editor item_editor=column_selector item_default=table.column_names.first.pretty display=display
+
+    unique = Option "Unique" "..Unique" [["on", columns_selector]]
+    equal_count = Option "Equal Count" "..Equal_Count" [["order_by", columns_selector]]
+    names=[unique, equal_count]
+
+    Single_Choice display=display values=names
+
 ## PRIVATE
    Make a column name selector.
 make_order_by_selector : Table -> Display -> Boolean -> Widget

distribution/lib/Standard/Table/0.0.0-dev/src/Main.enso

@@ -31,6 +31,8 @@ export project.Extensions.Table_Conversions.parse_to_table
 export project.Extensions.Table_Conversions.to_table
 export project.Extensions.Table_Conversions.write_table
 
+export project.Grouping_Method.Grouping_Method
+
 export project.Headers.Headers
 
 export project.Join_Condition.Join_Condition

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

@@ -35,6 +35,8 @@ import project.Delimited.Delimited_Format.Delimited_Format
 import project.Expression.Expression
 import project.Expression.Expression_Error
 import project.Extensions.Table_Conversions
+import project.Grouping_Method.Grouping_Method
+import project.Internal.Add_Group_Number
 import project.Internal.Add_Row_Number
 import project.Internal.Add_Running
 import project.Internal.Aggregate_Column_Helper
@@ -2323,6 +2325,99 @@ type Table
     add_row_number self (name:Text="Row") (from:Integer=0) (step:Integer=1) (group_by:(Vector | Text | Integer | Regex)=[]) (order_by:(Vector | Text)=[]) (on_problems:Problem_Behavior=..Report_Warning) =
         Incomparable_Values.handle_errors <| Add_Row_Number.add_row_number self name from step group_by order_by on_problems
 
+    ## PRIVATE add group column, group id, bucket, tile
+       GROUP Standard.Base.Values
+       ICON column_add
+       Adds a new column to the table enumerating groups of rows, assigning each
+       row to one group number. All rows in each group will get the same number.
+
+       Arguments:
+       - grouping_method: Specifies how to group the rows; see "Grouping
+         Methods", below.
+       - name: The name of the new column. Defaults to "Group".
+       - from: The starting value for the enumeration. Defaults to 0.
+       - step: The amount to increment the enumeration by. Defaults to 1.
+
+       ? Grouping Methods
+
+         The following grouping methods are supported:
+         - `Unique`: Group rows by the specified columns.
+         - Equal_Count: Create the specified number of groups with the same
+           number of rows in each group (except possibly the last one).
+
+       ? Ordering of rows
+
+         Note that the ordering of rows from the original table is preserved in
+         all cases. The grouping and ordering settings can affect how the group
+         numbers are assigned, depending on the grouping method.  The order of
+         the rows itself is not changed by this operation.
+
+       ! Error Conditions
+
+         - If the columns specified in `group_by` or `order_by` are not present
+           in the table, a `Missing_Input_Columns` error is raised.
+         - If the column with the same name as provided `name` already exists,
+           a `Duplicate_Output_Column_Names` problem is reported and the
+           existing column is renamed to avoid the clash.
+         - If grouping on floating point numbers, a `Floating_Point_Equality`
+           problem is reported.
+
+       > Example
+         Assign group numbers based on unique values of the first two columns.
+
+            ## table:
+                x | y | z
+               ---+---+---
+                1 | 0 | 2
+                0 | 1 | 0
+                1 | 2 | 0
+                0 | 1 | 1
+                1 | 0 | 1
+                1 | 2 | 1
+            table = table_builder [['x', [1, 0, 1, 0, 1, 1]], ['y', [0, 1, 2, 1, 0, 2]], ['z' [2, 0, 0, 1, 1, 1]]]
+            table2 = table.add_group_number (..Unique group_by=['x', 'y']) "g"
+            table2.at 'g' . to_vector
+            # => [0, 1, 2, 1, 0, 2]
+            ## table2:
+                x | y | z | g
+               ---+---+---+---
+                1 | 0 | 2 | 0
+                0 | 1 | 0 | 1
+                1 | 2 | 0 | 2
+                0 | 1 | 1 | 1
+                1 | 0 | 1 | 2
+                1 | 2 | 1 | 0
+
+       > Example
+         Divide rows into three groups.
+            ## table:
+                x | y
+               ---+---
+                1 | 5
+                2 | 4
+                3 | 3
+                4 | 2
+                5 | 1
+            table = table_builder [['x', [1, 2, 3, 4, 5]], ['y', [5, 4, 3, 2, 1]]]
+            table2 = tabble.add_group_number (..Equal_Count 3) "g"
+            table2.at 'g' . to_vector
+            # => [0, 0, 1, 1, 2]
+            ## table2:
+                x | y | g
+               ---+---+---
+                1 | 5 | 0
+                2 | 4 | 0
+                3 | 3 | 1
+                4 | 2 | 1
+                5 | 1 | 2
+    @grouping_method (Widget_Helpers.make_grouping_method_selector display=..Always)
+    @name (Widget.Text_Input display=..Always)
+    @from (Widget.Numeric_Input display=..Always)
+    @group_by (Widget_Helpers.make_column_name_multi_selector display=..When_Modified)
+    @order_by (Widget_Helpers.make_order_by_selector display=..When_Modified)
+    add_group_number self (grouping_method:Grouping_Method=(Missing_Argument.throw "grouping_method")) (name:Text="Group") (from:Integer=0) (step:Integer=1) (on_problems:Problem_Behavior=..Report_Warning) -> Table =
+        Incomparable_Values.handle_errors <| Add_Group_Number.add_group_number self grouping_method name from step on_problems
+
     ## ALIAS add column, expression, formula, new column, update column
        GROUP Standard.Base.Values
        ICON column_add