Reject negative ignore_rows and ignore_columns in CSVDocumentCleaner

docs-website/docs/pipeline-components/preprocessors/csvdocumentcleaner.mdx

@@ -28,8 +28,8 @@ Use `CSVDocumentCleaner` to clean CSV documents by removing empty rows and colum
 
 ### Parameters
 
-- `ignore_rows`: Number of rows to ignore from the top of the CSV table before processing. If any columns are removed, the same columns will be dropped from the ignored rows.
-- `ignore_columns`: Number of columns to ignore from the left of the CSV table before processing. If any rows are removed, the same rows will be dropped from the ignored columns.
+- `ignore_rows`: Number of rows to ignore from the top of the CSV table before processing. Must be 0 or greater. If any columns are removed, the same columns will be dropped from the ignored rows.
+- `ignore_columns`: Number of columns to ignore from the left of the CSV table before processing. Must be 0 or greater. If any rows are removed, the same rows will be dropped from the ignored columns.
 - `remove_empty_rows`: Whether to remove entirely empty rows.
 - `remove_empty_columns`: Whether to remove entirely empty columns.
 - `keep_id`: Whether to retain the original document ID in the output document.

haystack/components/preprocessors/csv_document_cleaner.py

@@ -43,10 +43,17 @@ def __init__(
         :param remove_empty_rows: Whether to remove rows that are entirely empty.
         :param remove_empty_columns: Whether to remove columns that are entirely empty.
         :param keep_id: Whether to retain the original document ID in the output document.
+        :raises ValueError: If `ignore_rows` or `ignore_columns` is less than 0.
 
         Rows and columns ignored using these parameters are preserved in the final output, meaning
         they are not considered when removing empty rows and columns.
         """
+        if ignore_rows < 0:
+            raise ValueError("ignore_rows must be greater than or equal to 0.")
+
+        if ignore_columns < 0:
+            raise ValueError("ignore_columns must be greater than or equal to 0.")
+
         self.ignore_rows = ignore_rows
         self.ignore_columns = ignore_columns
         self.remove_empty_rows = remove_empty_rows

releasenotes/notes/csvdocumentcleaner-negative-ignore-counts-f0ba474431dcc4ed.yaml

@@ -0,0 +1,7 @@
+---
+upgrade:
+  - |
+    ``CSVDocumentCleaner`` now raises a ``ValueError`` during initialization if ``ignore_rows`` or ``ignore_columns`` is less than 0. This validation is also enforced when loading serialized pipelines containing negative ignore counts. To migrate, update any configuration using negative counts to use ``0`` or omit the parameter.
+fixes:
+  - |
+    Fixed an issue in ``CSVDocumentCleaner`` where negative ``ignore_rows`` or ``ignore_columns`` values silently discarded data.

test/components/preprocessors/test_csv_document_cleaner.py

@@ -2,7 +2,9 @@
 #
 # SPDX-License-Identifier: Apache-2.0
 
-from haystack import Document
+import pytest
+
+from haystack import Document, default_from_dict
 from haystack.components.preprocessors.csv_document_cleaner import CSVDocumentCleaner
 
 
@@ -218,3 +220,25 @@ def test_remove_empty_rows_and_columns_false() -> None:
     result = csv_document_cleaner.run([csv_document])
     cleaned_document = result["documents"][0]
     assert cleaned_document.content == ",B,C\n,,4\n,,\n"
+
+
+@pytest.mark.parametrize("value", [-1, -5])
+def test_negative_ignore_rows_raises(value: int) -> None:
+    with pytest.raises(ValueError, match="ignore_rows must be greater than or equal to 0"):
+        CSVDocumentCleaner(ignore_rows=value)
+
+
+@pytest.mark.parametrize("value", [-1, -5])
+def test_negative_ignore_columns_raises(value: int) -> None:
+    with pytest.raises(ValueError, match="ignore_columns must be greater than or equal to 0"):
+        CSVDocumentCleaner(ignore_columns=value)
+
+
+@pytest.mark.parametrize("param", ["ignore_rows", "ignore_columns"])
+def test_negative_ignore_values_raise_from_dict(param: str) -> None:
+    data = {
+        "type": "haystack.components.preprocessors.csv_document_cleaner.CSVDocumentCleaner",
+        "init_parameters": {param: -1},
+    }
+    with pytest.raises(ValueError, match=f"{param} must be greater than or equal to 0"):
+        default_from_dict(CSVDocumentCleaner, data)