feat: suggest similar columns if column gets accessed that doesnt exi…

…st (#385)

Closes #203

### Summary of Changes

* Added a method `_get_similar_columns` to find columns with a name
similar to a given name.
* Updated `UnknownColumnNameError` to allow for passing name
suggestions.
* Used these suggestions in `get_column`, `keep_only_columns`,
`remove_columns`, `rename_column`, `replace_column`, `transform_column`,
`plot_lineplot`, `plot_scatterplot`.

Co-authored-by: jxnior01 <129027012+jxnior01@users.noreply.github.com>

pyproject.toml

@@ -22,6 +22,7 @@ scikit-learn = "^1.2.0"
 seaborn = "^0.12.2"
 openpyxl = "^3.1.2"
 scikit-image = "^0.21.0"
+levenshtein = "^0.21.1"
 
 [tool.poetry.group.dev.dependencies]
 pytest = "^7.2.1"

src/safeds/data/tabular/containers/_table.py

@@ -7,6 +7,7 @@
 from pathlib import Path
 from typing import TYPE_CHECKING, Any, TypeVar
 
+import Levenshtein
 import matplotlib.pyplot as plt
 import numpy as np
 import openpyxl
@@ -597,7 +598,8 @@ def get_column(self, column_name: str) -> Column:
         Column('b', [2])
         """
         if not self.has_column(column_name):
-            raise UnknownColumnNameError([column_name])
+            similar_columns = self._get_similar_columns(column_name)
+            raise UnknownColumnNameError([column_name], similar_columns)
 
         return Column._from_pandas_series(
             self._data[column_name],
@@ -695,6 +697,34 @@ def get_row(self, index: int) -> Row:
 
         return Row._from_pandas_dataframe(self._data.iloc[[index]], self._schema)
 
+    def _get_similar_columns(self, column_name: str) -> list[str]:
+        """
+        Get all the column names in a Table that are similar to a given name.
+
+        Parameters
+        ----------
+        column_name : str
+            The name to compare the Table's column names to.
+
+        Returns
+        -------
+        similar_columns: list[str]
+            A list of all column names in the Table that are similar or equal to the given column name.
+        """
+        similar_columns = []
+        similarity = 0.6
+        i = 0
+        while i < len(self.column_names):
+            if Levenshtein.jaro_winkler(self.column_names[i], column_name) >= similarity:
+                similar_columns.append(self.column_names[i])
+            i += 1
+            if len(similar_columns) == 4 and similarity < 0.9:
+                similarity += 0.1
+                similar_columns = []
+                i = 0
+
+        return similar_columns
+
     # ------------------------------------------------------------------------------------------------------------------
     # Information
     # ------------------------------------------------------------------------------------------------------------------
@@ -1106,11 +1136,13 @@ def keep_only_columns(self, column_names: list[str]) -> Table:
         1  4
         """
         invalid_columns = []
+        similar_columns: list[str] = []
         for name in column_names:
             if not self._schema.has_column(name):
+                similar_columns = similar_columns + self._get_similar_columns(name)
                 invalid_columns.append(name)
         if len(invalid_columns) != 0:
-            raise UnknownColumnNameError(invalid_columns)
+            raise UnknownColumnNameError(invalid_columns, similar_columns)
 
         clone = self._copy()
         clone = clone.remove_columns(list(set(self.column_names) - set(column_names)))
@@ -1151,11 +1183,13 @@ def remove_columns(self, column_names: list[str]) -> Table:
         1  3
         """
         invalid_columns = []
+        similar_columns: list[str] = []
         for name in column_names:
             if not self._schema.has_column(name):
+                similar_columns = similar_columns + self._get_similar_columns(name)
                 invalid_columns.append(name)
         if len(invalid_columns) != 0:
-            raise UnknownColumnNameError(invalid_columns)
+            raise UnknownColumnNameError(invalid_columns, similar_columns)
 
         transformed_data = self._data.drop(labels=column_names, axis="columns")
         transformed_data.columns = [name for name in self._schema.column_names if name not in column_names]
@@ -1349,7 +1383,8 @@ def rename_column(self, old_name: str, new_name: str) -> Table:
         0  1  2
         """
         if old_name not in self._schema.column_names:
-            raise UnknownColumnNameError([old_name])
+            similar_columns = self._get_similar_columns(old_name)
+            raise UnknownColumnNameError([old_name], similar_columns)
         if old_name == new_name:
             return self
         if new_name in self._schema.column_names:
@@ -1401,7 +1436,8 @@ def replace_column(self, old_column_name: str, new_columns: list[Column]) -> Tab
         0  1    3
         """
         if old_column_name not in self._schema.column_names:
-            raise UnknownColumnNameError([old_column_name])
+            similar_columns = self._get_similar_columns(old_column_name)
+            raise UnknownColumnNameError([old_column_name], similar_columns)
 
         columns = list[Column]()
         for old_column in self.column_names:
@@ -1705,7 +1741,8 @@ def transform_column(self, name: str, transformer: Callable[[Row], Any]) -> Tabl
             items: list = [transformer(item) for item in self.to_rows()]
             result: list[Column] = [Column(name, items)]
             return self.replace_column(name, result)
-        raise UnknownColumnNameError([name])
+        similar_columns = self._get_similar_columns(name)
+        raise UnknownColumnNameError([name], similar_columns)
 
     def transform_table(self, transformer: TableTransformer) -> Table:
         """
@@ -1881,9 +1918,13 @@ def plot_lineplot(self, x_column_name: str, y_column_name: str) -> Image:
         >>> image = table.plot_lineplot("temperature", "sales")
         """
         if not self.has_column(x_column_name) or not self.has_column(y_column_name):
+            similar_columns_x = self._get_similar_columns(x_column_name)
+            similar_columns_y = self._get_similar_columns(y_column_name)
             raise UnknownColumnNameError(
                 ([x_column_name] if not self.has_column(x_column_name) else [])
                 + ([y_column_name] if not self.has_column(y_column_name) else []),
+                (similar_columns_x if not self.has_column(x_column_name) else [])
+                + (similar_columns_y if not self.has_column(y_column_name) else []),
             )
 
         fig = plt.figure()
@@ -1935,9 +1976,13 @@ def plot_scatterplot(self, x_column_name: str, y_column_name: str) -> Image:
         >>> image = table.plot_scatterplot("temperature", "sales")
         """
         if not self.has_column(x_column_name) or not self.has_column(y_column_name):
+            similar_columns_x = self._get_similar_columns(x_column_name)
+            similar_columns_y = self._get_similar_columns(y_column_name)
             raise UnknownColumnNameError(
                 ([x_column_name] if not self.has_column(x_column_name) else [])
                 + ([y_column_name] if not self.has_column(y_column_name) else []),
+                (similar_columns_x if not self.has_column(x_column_name) else [])
+                + (similar_columns_y if not self.has_column(y_column_name) else []),
             )
 
         fig = plt.figure()

src/safeds/exceptions/_data.py

@@ -16,8 +16,19 @@ class UnknownColumnNameError(KeyError):
         The name of the column that was tried to be accessed.
     """
 
-    def __init__(self, column_names: list[str]):
-        super().__init__(f"Could not find column(s) '{', '.join(column_names)}'")
+    def __init__(self, column_names: list[str], similar_columns: list[str] | None = None):
+        class _UnknownColumnNameErrorMessage(
+            str,
+        ):  # This class is necessary for the newline character in a KeyError exception. See https://stackoverflow.com/a/70114007
+            def __repr__(self) -> str:
+                return str(self)
+
+        error_message = f"Could not find column(s) '{', '.join(column_names)}'."
+
+        if similar_columns is not None and len(similar_columns) > 0:
+            error_message += f"\nDid you mean '{similar_columns}'?"
+
+        super().__init__(_UnknownColumnNameErrorMessage(error_message))
 
 
 class NonNumericColumnError(Exception):

tests/safeds/data/tabular/containers/_table/test_get_similar_columns.py

@@ -0,0 +1,46 @@
+import pytest
+from safeds.data.tabular.containers import Table
+from safeds.exceptions._data import UnknownColumnNameError
+
+
+@pytest.mark.parametrize(
+    ("table", "column_name", "expected"),
+    [
+        (Table({"column1": ["col1_1"], "x": ["y"], "cilumn2": ["cil2_1"]}), "col1", ["column1"]),
+        (
+            Table(
+                {
+                    "column1": ["col1_1"],
+                    "col2": ["col2_1"],
+                    "col3": ["col2_1"],
+                    "col4": ["col2_1"],
+                    "cilumn2": ["cil2_1"],
+                },
+            ),
+            "clumn1",
+            ["column1", "cilumn2"],
+        ),
+        (
+            Table({"column1": ["a"], "column2": ["b"], "column3": ["c"]}),
+            "notexisting",
+            [],
+        ),
+        (
+            Table({"column1": ["col1_1"], "x": ["y"], "cilumn2": ["cil2_1"]}),
+            "x",
+            ["x"],
+        ),
+        (Table({}), "column1", []),
+    ],
+    ids=["one similar", "two similar/ dynamic increase", "no similar", "exact match", "empty table"],
+)
+def test_should_get_similar_column_names(table: Table, column_name: str, expected: list[str]) -> None:
+    assert table._get_similar_columns(column_name) == expected
+
+
+def test_should_raise_error_if_column_name_unknown() -> None:
+    with pytest.raises(
+        UnknownColumnNameError,
+        match=r"Could not find column\(s\) 'col3'.\nDid you mean '\['col1', 'col2'\]'?",
+    ):
+        raise UnknownColumnNameError(["col3"], ["col1", "col2"])

tests/safeds/exceptions/test_unknown_column_name_error.py

@@ -0,0 +1,42 @@
+import pytest
+from safeds.exceptions import UnknownColumnNameError
+
+
+@pytest.mark.parametrize(
+    ("column_names", "similar_columns", "expected_error_message"),
+    [
+        (["column1"], [], r"Could not find column\(s\) 'column1'\."),
+        (["column1", "column2"], [], r"Could not find column\(s\) 'column1, column2'\."),
+        (["column1"], ["column_a"], r"Could not find column\(s\) 'column1'\.\nDid you mean '\['column_a'\]'\?"),
+        (
+            ["column1", "column2"],
+            ["column_a"],
+            r"Could not find column\(s\) 'column1, column2'\.\nDid you mean '\['column_a'\]'\?",
+        ),
+        (
+            ["column1"],
+            ["column_a", "column_b"],
+            r"Could not find column\(s\) 'column1'\.\nDid you mean '\['column_a', 'column_b'\]'\?",
+        ),
+        (
+            ["column1", "column2"],
+            ["column_a", "column_b"],
+            r"Could not find column\(s\) 'column1, column2'\.\nDid you mean '\['column_a', 'column_b'\]'\?",
+        ),
+    ],
+    ids=[
+        "one_unknown_no_suggestions",
+        "two_unknown_no_suggestions",
+        "one_unknown_one_suggestion",
+        "two_unknown_one_suggestion",
+        "one_unknown_two_suggestions",
+        "two_unknown_two_suggestions",
+    ],
+)
+def test_empty_similar_columns(
+    column_names: list[str],
+    similar_columns: list[str],
+    expected_error_message: str,
+) -> None:
+    with pytest.raises(UnknownColumnNameError, match=expected_error_message):
+        raise UnknownColumnNameError(column_names, similar_columns)