fix(ansi): handle CRLF line endings in Text.from_ansi

rich/ansi.py

@@ -132,6 +132,7 @@ def decode(self, terminal_text: str) -> Iterable[Text]:
         Yields:
             Text: Marked up Text.
         """
+        terminal_text = terminal_text.replace("\r\n", "\n")
         for line in re.split(r"(?<=\n)", terminal_text):
             yield self.decode_line(line.rstrip("\n"))
 

tests/test_ansi.py

@@ -102,3 +102,16 @@ def test_decode_newlines():
     assert Text.from_ansi("Hello\nWorld\n").plain == "Hello\nWorld\n"
     assert Text.from_ansi("Hello\nWorld\n\n").plain == "Hello\nWorld\n\n"
     assert Text.from_ansi("\nHello\nWorld\n\n").plain == "\nHello\nWorld\n\n"
+
+
+def test_decode_crlf():
+    """Test CRLF line endings are handled correctly.
+    Regression test for https://github.com/Textualize/rich/issues/4090
+    """
+    assert Text.from_ansi("Hello\r\nWorld\r\n").plain == "Hello\nWorld\n"
+    assert Text.from_ansi("Hello\r\n").plain == "Hello\n"
+    assert Text.from_ansi("\r\nHello\r\n").plain == "\nHello\n"
+    assert Text.from_ansi("Hello\r\n\r\nWorld").plain == "Hello\n\nWorld"
+    # Mixed line endings
+    assert Text.from_ansi("Hello\nWorld\r\n").plain == "Hello\nWorld\n"
+    assert Text.from_ansi("Hello\r\nWorld\n").plain == "Hello\nWorld\n"