From abf3ce7785c60d7c5ba3d7bc718f42be5fe867fd Mon Sep 17 00:00:00 2001
From: Jean-Baptiste Braun <jbaptiste.braun@gmail.com>
Date: Fri, 22 May 2026 01:45:14 +0200
Subject: [PATCH] fix(descriptors): silence SyntaxWarning '\d' in TextMatch
 docstring
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The TextMatch class docstring contains an inline regex example
`r"\b\d{3}-\d{3}-\d{4}\b"`. The `r"..."` prefix only marks the *example*
literal as raw — the docstring itself is a plain string, so Python 3.12+
parses `\b` and `\d` inside it and emits
`SyntaxWarning: invalid escape sequence '\d'` on every import.

Mark the docstring itself raw (`r""" ... """`). The rendered help text is
unchanged and the example regex still displays the same way.

Add a regression test that compiles the module under
`warnings.simplefilter("error", SyntaxWarning)` so this can't silently
regress on a future docstring edit.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
---
 src/evidently/descriptors/text_match.py     |  7 ++++++-
 tests/future/descriptors/test_text_match.py | 17 +++++++++++++++++
 2 files changed, 23 insertions(+), 1 deletion(-)

diff --git a/src/evidently/descriptors/text_match.py b/src/evidently/descriptors/text_match.py
index be1edb1ea8..905847d026 100644
--- a/src/evidently/descriptors/text_match.py
+++ b/src/evidently/descriptors/text_match.py
@@ -115,7 +115,12 @@ def exact_match(self, text: str, items: List[str]) -> bool:
 
 
 class TextMatch(Descriptor):
-    """
+    # The docstring contains a regex example with `\b\d{3}-\d{3}-\d{4}\b`. The
+    # `r"..."` prefix only marks the *example* literal as raw — the docstring
+    # itself still parses every backslash, so `\d` triggers
+    # `SyntaxWarning: invalid escape sequence '\d'` on Python 3.12+. Marking the
+    # docstring itself raw silences the warning without changing rendered help.
+    r"""
     Unified text matching descriptor that handles all word/text matching scenarios.
 
     This descriptor replaces multiple legacy text matching features with a single,
diff --git a/tests/future/descriptors/test_text_match.py b/tests/future/descriptors/test_text_match.py
index 761f511ae8..7cb9f93c95 100644
--- a/tests/future/descriptors/test_text_match.py
+++ b/tests/future/descriptors/test_text_match.py
@@ -1,8 +1,25 @@
+import py_compile
+import warnings
+
 import pandas as pd
 import pytest
 
 from evidently.core.datasets import Dataset
 from evidently.descriptors import TextMatch
+import evidently.descriptors.text_match as text_match_module
+
+
+def test_text_match_module_compiles_without_syntax_warning(tmp_path):
+    # Regression: the TextMatch docstring contains an `r"\b\d{3}-..."` regex
+    # example, but the docstring itself wasn't raw, so `\d` triggered
+    # `SyntaxWarning: invalid escape sequence '\d'` on Python 3.12+.
+    # py_compile surfaces SyntaxWarnings as PyCompileErrors when warnings are
+    # promoted to errors, so this is a reliable check that doesn't depend on
+    # whether the module is already cached / imported elsewhere.
+    source_path = text_match_module.__file__
+    with warnings.catch_warnings():
+        warnings.simplefilter("error", SyntaxWarning)
+        py_compile.compile(source_path, cfile=str(tmp_path / "out.pyc"), doraise=True)
 
 
 @pytest.fixture