Changes from session with Ned (#10349)

- Removed `second_row` and `second_column` from the `Table` and `DB_Table`. - Added `first_value` and `last_value` to the `Table` and `DB_Table`. - Fixed bug where negative index access wasn't allowed on `Column`. - Added error if negative index access used on `DB_Column`. Tells user they have to materialize. - Fix argument order for `Table.text_cleanse` and a couple of typo corrections. - Rename `auto_value_type` to `auto_cast` on table and columns.
enso-org · Jun 24, 2024 · e6c8ec7 · e6c8ec7
1 parent 5233390
commit e6c8ec7
Show file tree

Hide file tree

Showing 8 changed files with 205 additions and 126 deletions.
diff --git a/distribution/lib/Standard/Database/0.0.0-dev/src/DB_Column.enso b/distribution/lib/Standard/Database/0.0.0-dev/src/DB_Column.enso
@@ -156,7 +156,8 @@ type DB_Column
              example_at = Examples.integer_column.get 0 -1
     get : Integer -> Any -> Any | Nothing
     get self index:Integer=0 (~default=Nothing) =
-        self.read (..First index+1) . get index default
+        if index < 0 then Error.throw (Unsupported_Database_Operation.Error "Reading backwards from end is not supported in-database. Use `read` to materialize the column.") else
+            self.read (..First index+1) . get index default
 
     ## GROUP Standard.Base.Metadata
        ICON metadata
@@ -1806,14 +1807,15 @@ type DB_Column
 
     ## GROUP Standard.Base.Conversions
        ICON convert
+       ALIAS auto_value_type
        Change the value type of the column to a more specific one, based on its
        contents.
 
        This operation is currently not available in the Database backend.
-    auto_value_type : Boolean -> DB_Column
-    auto_value_type self shrink_types=False =
+    auto_cast : Boolean -> DB_Column
+    auto_cast self shrink_types=False =
         _ = shrink_types
-        Error.throw <| Unsupported_Database_Operation.Error "`DB_Column.auto_value_type` is not supported in the Database backends."
+        Error.throw <| Unsupported_Database_Operation.Error "`DB_Column.auto_cast` is not supported in the Database backends."
 
     ## PRIVATE
        Shares the core CAST logic between `cast` and `parse`.

diff --git a/distribution/lib/Standard/Database/0.0.0-dev/src/DB_Table.enso b/distribution/lib/Standard/Database/0.0.0-dev/src/DB_Table.enso
@@ -172,20 +172,27 @@ type DB_Table
         col = self.get selector if_missing=Nothing
         if Nothing == col then if_missing else col.get index if_missing
 
+    ## ALIAS first cell
+       GROUP Standard.Base.Selections
+       ICON local_scope4
+       Gets the top left value from the table.
+    first_value : Any ! Index_Out_Of_Bounds
+    first_value self = self.at 0 . at 0
+
+    ## ALIAS last cell
+       GROUP Standard.Base.Selections
+       ICON local_scope4
+       Gets the bottom right value from the table.
+    last_value : Any ! Index_Out_Of_Bounds
+    last_value self = self.last_row . at -1
+
     ## ALIAS first field
        GROUP Standard.Base.Selections
        ICON select_column
        Gets the first column.
     first_column : DB_Column ! Index_Out_Of_Bounds
     first_column self = self.at 0
 
-    ## ALIAS second field
-       GROUP Standard.Base.Selections
-       ICON select_column
-       Gets the second column
-    second_column : DB_Column ! Index_Out_Of_Bounds
-    second_column self = self.at 1
-
     ## ALIAS last field
        GROUP Standard.Base.Selections
        ICON select_column
@@ -1164,13 +1171,6 @@ type DB_Table
     first_row self =
         self.read (..First 1) . rows . first
 
-    ## GROUP Standard.Base.Selections
-       ICON select_row
-       Returns the second row of the table.
-    second_row : Row ! Index_Out_Of_Bounds
-    second_row self =
-        self.read (..First 2) . rows . second
-
     ## GROUP Standard.Base.Selections
        ICON select_row
        Returns the last row of the table.
@@ -2556,8 +2556,8 @@ type DB_Table
 
        This operation is currently not available in the Database backend.
     @columns (Widget_Helpers.make_column_name_multi_selector add_regex=True add_by_type=True)
-    auto_value_types : Vector (Text | Integer | Regex | By_Type) | Text | Integer | Regex | By_Type -> Boolean -> Boolean -> Problem_Behavior -> DB_Table
-    auto_value_types self columns:(Vector (Text | Integer | Regex | By_Type) | Text | Integer | Regex | By_Type)=self.column_names shrink_types:Boolean=False error_on_missing_columns:Boolean=True on_problems:Problem_Behavior=..Report_Warning =
+    auto_cast : Vector (Text | Integer | Regex | By_Type) | Text | Integer | Regex | By_Type -> Boolean -> Boolean -> Problem_Behavior -> DB_Table
+    auto_cast self columns:(Vector (Text | Integer | Regex | By_Type) | Text | Integer | Regex | By_Type)=self.column_names shrink_types:Boolean=False error_on_missing_columns:Boolean=True on_problems:Problem_Behavior=..Report_Warning =
         _ = [columns, shrink_types, error_on_missing_columns, on_problems]
         Error.throw (Unsupported_Database_Operation.Error "DB_Table.auto_value_types is not supported in the Database backends.")
 
@@ -2938,13 +2938,14 @@ type DB_Table
        Applies the specified cleansings to the text in each row of the specified columns
 
        Arguments:
+       - from: The column(s) to cleanse.
        - remove: A vector of the text cleanings to remove from the text. The text cleansings are
            applied in the order they are provided. The same text cleansing can be used multiple
            times. The text cleansings are:
-             - ..Leading_Whitespace: Removes all whitspace from the start of the string.
-             - ..Trailing_Whitespace: Removes all whitspace from the end of the string.
-             - ..Duplicate_Whitespace: Removes all duplicate whitspace from the string replacing it with the first whitespace character of the duplicated block.
-             - ..All_Whitespace: Removes all whitspace from the string.
+             - ..Leading_Whitespace: Removes all whitespace from the start of the string.
+             - ..Trailing_Whitespace: Removes all whitespace from the end of the string.
+             - ..Duplicate_Whitespace: Removes all duplicate whitespace from the string replacing it with the first whitespace character of the duplicated block.
+             - ..All_Whitespace: Removes all whitespace from the string.
              - ..Leading_Numbers: Removes all numbers from the start of the string.
              - ..Trailing_Numbers: Removes all numbers from the end of the string.
              - ..Non_ASCII: Removes all non-ascii characters from the string.
@@ -2953,16 +2954,15 @@ type DB_Table
              - ..Numbers: Removes all numbers characters from the string.
              - ..Punctuation: Removes all characters in the set ,.!?():;'" from the string.
              - ..Symbols: Removes anything that isn't letters, numbers or whitespace from the string.
-       - from: The column(s) to cleanse.
 
        > Example
          Remove leading and trailing spaces from cells.
 
-             table.text_cleanse [..Leading_Whitespace, ..Trailing_Whitespace]
-    @remove make_data_cleanse_vector_selector
+             table.text_cleanse ["Input"] [..Leading_Whitespace, ..Trailing_Whitespace]
     @from (Widget_Helpers.make_column_name_multi_selector add_regex=True add_by_type=True)
-    text_cleanse : Vector Named_Pattern -> Vector (Integer | Text | Regex | By_Type) -> DB_Table
-    text_cleanse self remove from:(Vector (Integer | Text | Regex | By_Type)) =
+    @remove make_data_cleanse_vector_selector
+    text_cleanse : Vector (Integer | Text | Regex | By_Type) -> Vector Named_Pattern -> DB_Table
+    text_cleanse self from:(Vector (Integer | Text | Regex | By_Type)) remove =
         transformer col = col.text_cleanse remove
         Table_Helpers.replace_columns_with_transformed_columns self from transformer
 

diff --git a/distribution/lib/Standard/Table/0.0.0-dev/src/Column.enso b/distribution/lib/Standard/Table/0.0.0-dev/src/Column.enso
@@ -6,6 +6,7 @@ import Standard.Base.Errors.Common.Arithmetic_Error
 import Standard.Base.Errors.Common.Incomparable_Values
 import Standard.Base.Errors.Common.Index_Out_Of_Bounds
 import Standard.Base.Errors.Common.No_Such_Method
+import Standard.Base.Errors.Deprecated.Deprecated
 import Standard.Base.Errors.Illegal_Argument.Illegal_Argument
 import Standard.Base.Errors.Illegal_State.Illegal_State
 import Standard.Base.Internal.Polyglot_Helpers
@@ -1962,6 +1963,7 @@ type Column
 
     ## GROUP Standard.Base.Conversions
        ICON convert
+       ALIAS auto_value_type
        Change the value type of the column to a more specific one, based on its
        contents.
 
@@ -1990,15 +1992,54 @@ type Column
              elements are no longer than 255 characters, the column will get a
              max length of 255. Otherwise, the column size limit will stay
              unchanged.
-    auto_value_type : Boolean -> Column
-    auto_value_type self shrink_types=False =
+    auto_cast : Boolean -> Column
+    auto_cast self shrink_types=False =
         new_value_type = case shrink_types of
             False -> self.inferred_precise_value_type
             True ->
                 Storage.to_value_type self.java_column.getStorage.inferPreciseTypeShrunk
         # We run with Report_Error because we do not expect any problems.
         self.cast new_value_type on_problems=Problem_Behavior.Report_Error
 
+    ## PRIVATE
+       GROUP Standard.Base.Conversions
+       ICON convert
+
+       Deprecated: Use `auto_cast` instead.
+
+       Change the value type of the column to a more specific one, based on its
+       contents.
+
+       Arguments:
+       - shrink_types: If set `True`, smaller types will be chosen if possible,
+         according to the rules below. Defaults to `False`.
+
+       ? Auto Type Selection Rules
+
+         - If a `Mixed` column can be assigned a single type, like `Char` or
+           `Integer`, that will be used.
+         - Text columns are not parsed. To do that, use the `parse` method.
+         - If a `Float` column contains only integers, it will be converted to
+           an Integer column.
+         - If a `Decimal` column contains only integers that could fit in a
+           64-bit integer storage, it will be converted to an Integer column.
+         - If `shrink_types` is `False` (default), no other transformations are
+           applied.
+         - However, if `shrink_types` is set to `True`, then:
+           - Integer columns will be assigned the smallest size that can fit all
+             values (down to 16-bit integers; converting to the `Byte` type has
+             to be done manually through `cast`).
+           - If all elements in a text column have the same length, the type
+             will become fixed length.
+           - Otherwise, if a text column is variable length, but all text
+             elements are no longer than 255 characters, the column will get a
+             max length of 255. Otherwise, the column size limit will stay
+             unchanged.
+    auto_value_type : Boolean -> Column
+    auto_value_type self shrink_types=False =
+        Warning.attach (Deprecated.Warning "Standard.Table.Column.Column" "auto_value_type" "Deprecated: `auto_value_type` has been replaced by `auto_cast`.") <|
+            self.auto_cast shrink_types
+
     ## ALIAS transform column
        ICON column_add
 
@@ -2191,11 +2232,12 @@ type Column
     @index (self-> Numeric_Input minimum=0 maximum=self.length-1)
     get : Integer -> Any -> Any | Nothing
     get self index:Integer=0 (~default=Nothing) =
-        valid_index = (index >= 0) && (index < self.length)
-        if valid_index.not then default else
-            storage = self.java_column.getStorage
-            if storage.isNothing index then Nothing else
-                java_to_enso <| storage.getItem index
+        if index < 0 && index >= -self.length then self.get (self.length + index) default else
+            valid_index = (index >= 0) && (index < self.length)
+            if valid_index.not then default else
+                storage = self.java_column.getStorage
+                if storage.isNothing index then Nothing else
+                    java_to_enso <| storage.getItem index
 
     ## ICON data_input
        Returns a column containing rows of this column.

diff --git a/distribution/lib/Standard/Table/0.0.0-dev/src/Internal/Table_Helpers.enso b/distribution/lib/Standard/Table/0.0.0-dev/src/Internal/Table_Helpers.enso
@@ -296,7 +296,7 @@ read_name_map_from_table column_map:Table = case column_map.column_count of
         if col.value_type.is_text then col.to_vector else
             Error.throw (Illegal_Argument.Error "Expected a table with one or two columns of text values.")
     2 ->
-        if column_map.first_column.value_type.is_text && column_map.second_column.value_type.is_text then Map.from_vector column_map.rows else
+        if column_map.first_column.value_type.is_text && (column_map.at 1).value_type.is_text then Map.from_vector column_map.rows else
             Error.throw (Illegal_Argument.Error "Expected a table with one or two columns of text values.")
     _ -> Error.throw (Illegal_Argument.Error "Expected a table with one or two columns of text values.")