Fixed Text.from_ansi() removing trailing line break.

Author: kmvanbrunt

CHANGELOG.md

@@ -19,6 +19,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Fixed extraction of recursive exceptions https://github.com/Textualize/rich/pull/3772
 - Fixed padding applied to Syntax https://github.com/Textualize/rich/pull/3782
 - Fixed `Panel` title missing the panel background style https://github.com/Textualize/rich/issues/3569
+- Fixed `Text.from_ansi()` removing trailing line break. https://github.com/Textualize/rich/issues/3577
 
 ### Added
 

CONTRIBUTORS.md

@@ -94,3 +94,4 @@ The following people have contributed to the development of Rich:
 - [Jonathan Helmus](https://github.com/jjhelmus)
 - [Brandon Capener](https://github.com/bcapener)
 - [Alex Zheng](https://github.com/alexzheng111)
+- [Kevin Van Brunt](https://github.com/kmvanbrunt)

rich/text.py

@@ -326,6 +326,26 @@ def from_ansi(
         )
         decoder = AnsiDecoder()
         result = joiner.join(line for line in decoder.decode(text))
+
+        # AnsiDecoder's use of str.splitlines() discards trailing line breaks.
+        # If the original string ends with recognized line break character,
+        # then restore the missing newline.
+        # Source: https://docs.python.org/3/library/stdtypes.html#str.splitlines
+        line_break_chars = {
+            "\n",  # Line Feed
+            "\r",  # Carriage Return
+            "\v",  # Vertical Tab
+            "\f",  # Form Feed
+            "\x1c",  # File Separator
+            "\x1d",  # Group Separator
+            "\x1e",  # Record Separator
+            "\x85",  # Next Line (NEL)
+            "\u2028",  # Line Separator
+            "\u2029",  # Paragraph Separator
+        }
+        if text and text[-1] in line_break_chars:
+            result.append("\n")
+
         return result
 
     @classmethod

tests/test_ansi.py

@@ -32,6 +32,42 @@ def test_decode():
     assert lines == expected
 
 
+def test_from_ansi_ending_newline():
+    """Ensures that Text.from_ansi() converts a trailing line break to a
+    newline character instead of removing it.
+    """
+
+    # Line breaks recognized by str.splitlines().
+    # Source: https://docs.python.org/3/library/stdtypes.html#str.splitlines
+    line_breaks = {
+        "\n",  # Line Feed
+        "\r",  # Carriage Return
+        "\r\n",  # Carriage Return + Line Feed
+        "\v",  # Vertical Tab
+        "\f",  # Form Feed
+        "\x1c",  # File Separator
+        "\x1d",  # Group Separator
+        "\x1e",  # Record Separator
+        "\x85",  # Next Line (NEL)
+        "\u2028",  # Line Separator
+        "\u2029",  # Paragraph Separator
+    }
+
+    # Test all line breaks
+    for lb in line_breaks:
+        input_string = f"Text{lb}"
+        expected_output = input_string.replace(lb, "\n")
+        assert Text.from_ansi(input_string).plain == expected_output
+
+    # Test string without trailing line break
+    input_string = "No trailing\nline break"
+    assert Text.from_ansi(input_string).plain == input_string
+
+    # Test empty string
+    input_string = ""
+    assert Text.from_ansi(input_string).plain == input_string
+
+
 def test_decode_example():
     ansi_bytes = b"\x1b[01m\x1b[KC:\\Users\\stefa\\AppData\\Local\\Temp\\tmp3ydingba:\x1b[m\x1b[K In function '\x1b[01m\x1b[Kmain\x1b[m\x1b[K':\n\x1b[01m\x1b[KC:\\Users\\stefa\\AppData\\Local\\Temp\\tmp3ydingba:3:5:\x1b[m\x1b[K \x1b[01;35m\x1b[Kwarning: \x1b[m\x1b[Kunused variable '\x1b[01m\x1b[Ka\x1b[m\x1b[K' [\x1b[01;35m\x1b[K-Wunused-variable\x1b[m\x1b[K]\n    3 | int \x1b[01;35m\x1b[Ka\x1b[m\x1b[K=1;\n      |     \x1b[01;35m\x1b[K^\x1b[m\x1b[K\n"
     ansi_text = ansi_bytes.decode("utf-8")
@@ -45,7 +81,7 @@ def test_decode_example():
         console.print(text)
     result = capture.get()
     print(repr(result))
-    expected = "\x1b[1mC:\\Users\\stefa\\AppData\\Local\\Temp\\tmp3ydingba:\x1b[0m In function '\x1b[1mmain\x1b[0m':\n\x1b[1mC:\\Users\\stefa\\AppData\\Local\\Temp\\tmp3ydingba:3:5:\x1b[0m \x1b[1;35mwarning: \x1b[0munused variable '\x1b[1ma\x1b[0m' \n[\x1b[1;35m-Wunused-variable\x1b[0m]\n    3 | int \x1b[1;35ma\x1b[0m=1;\n      |     \x1b[1;35m^\x1b[0m\n"
+    expected = "\x1b[1mC:\\Users\\stefa\\AppData\\Local\\Temp\\tmp3ydingba:\x1b[0m In function '\x1b[1mmain\x1b[0m':\n\x1b[1mC:\\Users\\stefa\\AppData\\Local\\Temp\\tmp3ydingba:3:5:\x1b[0m \x1b[1;35mwarning: \x1b[0munused variable '\x1b[1ma\x1b[0m' \n[\x1b[1;35m-Wunused-variable\x1b[0m]\n    3 | int \x1b[1;35ma\x1b[0m=1;\n      |     \x1b[1;35m^\x1b[0m\n\n"
     assert result == expected
 
 
@@ -55,7 +91,7 @@ def test_decode_example():
         # https://github.com/Textualize/rich/issues/2688
         (
             b"\x1b[31mFound 4 errors in 2 files (checked 18 source files)\x1b(B\x1b[m\n",
-            "Found 4 errors in 2 files (checked 18 source files)",
+            "Found 4 errors in 2 files (checked 18 source files)\n",
         ),
         # https://mail.python.org/pipermail/python-list/2007-December/424756.html
         (b"Hallo", "Hallo"),