Added regular expression to detect ANSI style sequences. (#1492)

Author: kmvanbrunt

cmd2/argparse_custom.py

@@ -398,7 +398,7 @@ def __init__(self, value: object, descriptive_data: Sequence[Any], *args: Any) -
         # Make sure all objects are renderable by a Rich table.
         renderable_data = [obj if is_renderable(obj) else str(obj) for obj in descriptive_data]
 
-        # Convert objects with ANSI styles to Rich Text for correct display width.
+        # Convert strings containing ANSI style sequences to Rich Text objects for correct display width.
         self.descriptive_data = ru.prepare_objects_for_rendering(*renderable_data)
 
         # Save the original value to support CompletionItems as argparse choices.

cmd2/cmd2.py

@@ -1191,12 +1191,12 @@ def _completion_supported(self) -> bool:
 
     @property
     def visible_prompt(self) -> str:
-        """Read-only property to get the visible prompt with any ANSI style escape codes stripped.
+        """Read-only property to get the visible prompt with any ANSI style sequences stripped.
 
-        Used by transcript testing to make it easier and more reliable when users are doing things like coloring the
-        prompt using ANSI color codes.
+        Used by transcript testing to make it easier and more reliable when users are doing things like
+        coloring the prompt.
 
-        :return: prompt stripped of any ANSI escape codes
+        :return: the stripped prompt
         """
         return su.strip_style(self.prompt)
 
@@ -4214,7 +4214,7 @@ def _print_documented_command_topics(self, header: str, cmds: list[str], verbose
     def render_columns(self, str_list: list[str] | None, display_width: int = 80) -> str:
         """Render a list of single-line strings as a compact set of columns.
 
-        This method correctly handles strings containing ANSI escape codes and
+        This method correctly handles strings containing ANSI style sequences and
         full-width characters (like those used in CJK languages). Each column is
         only as wide as necessary and columns are separated by two spaces.
 

cmd2/rich_utils.py

@@ -1,5 +1,6 @@
 """Provides common utilities to support Rich in cmd2-based applications."""
 
+import re
 from collections.abc import Mapping
 from enum import Enum
 from typing import (
@@ -28,13 +29,16 @@
 
 from .styles import DEFAULT_CMD2_STYLES
 
+# A compiled regular expression to detect ANSI style sequences.
+ANSI_STYLE_SEQUENCE_RE = re.compile(r"\x1b\[[0-9;?]*m")
+
 
 class AllowStyle(Enum):
     """Values for ``cmd2.rich_utils.ALLOW_STYLE``."""
 
-    ALWAYS = 'Always'  # Always output ANSI style sequences
-    NEVER = 'Never'  # Remove ANSI style sequences from all output
-    TERMINAL = 'Terminal'  # Remove ANSI style sequences if the output is not going to the terminal
+    ALWAYS = "Always"  # Always output ANSI style sequences
+    NEVER = "Never"  # Remove ANSI style sequences from all output
+    TERMINAL = "Terminal"  # Remove ANSI style sequences if the output is not going to the terminal
 
     def __str__(self) -> str:
         """Return value instead of enum name for printing in cmd2's set command."""
@@ -234,7 +238,7 @@ def rich_text_to_string(text: Text) -> str:
     """Convert a Rich Text object to a string.
 
     This function's purpose is to render a Rich Text object, including any styles (e.g., color, bold),
-    to a plain Python string with ANSI escape codes. It differs from `text.plain`, which strips
+    to a plain Python string with ANSI style sequences. It differs from `text.plain`, which strips
     all formatting.
 
     :param text: the text object to convert
@@ -259,7 +263,7 @@ def rich_text_to_string(text: Text) -> str:
 
 
 def string_to_rich_text(text: str) -> Text:
-    r"""Create a Text object from a string which can contain ANSI escape codes.
+    r"""Create a Rich Text object from a string which can contain ANSI style sequences.
 
     This wraps rich.Text.from_ansi() to handle an issue where it removes the
     trailing line break from a string (e.g. "Hello\n" becomes "Hello").
@@ -323,9 +327,9 @@ def prepare_objects_for_rendering(*objects: Any) -> tuple[Any, ...]:
     """Prepare a tuple of objects for printing by Rich's Console.print().
 
     This function converts any non-Rich object whose string representation contains
-    ANSI style codes into a rich.Text object. This ensures correct display width
-    calculation, as Rich can then properly parse and account for the non-printing
-    ANSI codes. All other objects are left untouched, allowing Rich's native
+    ANSI style sequences into a Rich Text object. This ensures correct display width
+    calculation, as Rich can then properly parse and account for these non-printing
+    codes. All other objects are left untouched, allowing Rich's native
     renderers to handle them.
 
     :param objects: objects to prepare
@@ -342,12 +346,10 @@ def prepare_objects_for_rendering(*objects: Any) -> tuple[Any, ...]:
         if isinstance(renderable, ConsoleRenderable):
             continue
 
-        # Check if the object's string representation contains ANSI styles, and if so,
-        # replace it with a Rich Text object for correct width calculation.
         renderable_as_str = str(renderable)
-        renderable_as_text = string_to_rich_text(renderable_as_str)
 
-        if renderable_as_text.plain != renderable_as_str:
-            object_list[i] = renderable_as_text
+        # Check for any ANSI style sequences in the string.
+        if ANSI_STYLE_SEQUENCE_RE.search(renderable_as_str):
+            object_list[i] = string_to_rich_text(renderable_as_str)
 
     return tuple(object_list)

cmd2/string_utils.py

@@ -1,7 +1,7 @@
 """Provides string utility functions.
 
 This module offers a collection of string utility functions built on the Rich library.
-These utilities are designed to correctly handle strings with ANSI escape codes and
+These utilities are designed to correctly handle strings with ANSI style sequences and
 full-width characters (like those used in CJK languages).
 """
 
@@ -94,13 +94,12 @@ def stylize(val: str, style: StyleType) -> str:
 
 
 def strip_style(val: str) -> str:
-    """Strip all ANSI styles from a string.
+    """Strip all ANSI style sequences from a string.
 
     :param val: string to be stripped
     :return: the stripped string
     """
-    text = ru.string_to_rich_text(val)
-    return text.plain
+    return ru.ANSI_STYLE_SEQUENCE_RE.sub("", val)
 
 
 def str_width(val: str) -> int:
@@ -163,4 +162,4 @@ def norm_fold(val: str) -> str:
     """
     import unicodedata
 
-    return unicodedata.normalize('NFC', val).casefold()
+    return unicodedata.normalize("NFC", val).casefold()