batplot 1.8.1__py3-none-any.whl → 1.8.3__py3-none-any.whl

batplot_backup_20251221_101150/cli.py

@@ -0,0 +1,149 @@
+"""CLI entry for batplot.
+
+This module provides the main entry point for the batplot command-line interface.
+It handles version checking, argument delegation, and error handling.
+
+Design principles:
+- Lazy imports: imports happen inside main() to avoid side effects during module load
+- Clean separation: delegates actual work to batplot_main() in batplot.py
+- Non-intrusive: version check is silent and non-blocking (max 2s timeout)
+"""
+
+from __future__ import annotations
+
+import sys
+from typing import Optional
+
+
+def main(argv: Optional[list] = None) -> int:
+	"""
+	Main CLI entry point for batplot.
+	
+	This function is the entry point when users run 'batplot' from the command line.
+	It's also the function called by the 'batplot' command installed via pip/setuptools.
+	
+	HOW IT WORKS:
+	-----------
+	When you type 'batplot file.xy --interactive' in terminal:
+	
+	1. Python's setuptools calls this main() function
+	2. main() performs version check (non-blocking, cached)
+	3. main() parses command-line arguments
+	4. main() delegates to batplot_main() which routes to appropriate mode
+	5. Errors are caught and displayed cleanly (no scary tracebacks)
+	
+	WHY LAZY IMPORTS?
+	----------------
+	We import batplot_main() inside the try block, not at module level.
+	This is called "lazy importing" and has benefits:
+	- Faster startup: Only imports what's needed when needed
+	- Avoids side effects: Some modules might do things at import time
+	- Better error handling: Import errors happen in try/except block
+	
+	Args:
+		argv: Optional command line arguments for testing/programmatic use.
+		      If None, uses sys.argv (normal CLI usage).
+		      Format: ['--mode', 'file.ext', ...]
+		      Note: Do NOT include 'batplot' itself, it's added automatically.
+		
+	Returns:
+		Exit code: 0 for success, non-zero for error.
+		This follows Unix convention where 0 = success, non-zero = error.
+		
+	Examples:
+		>>> # Testing mode (programmatic use)
+		>>> main(['--cv', 'data.mpt'])
+		0
+		
+		>>> # Normal CLI usage (reads from sys.argv)
+		>>> main()
+		0
+	"""
+	# ====================================================================
+	# STEP 1: VERSION CHECK (NON-BLOCKING)
+	# ====================================================================
+	# Check PyPI for newer versions of batplot.
+	# This is done asynchronously and doesn't block the main program.
+	#
+	# Features:
+	# - Non-blocking: 2 second timeout (won't slow down startup)
+	# - Cached: Only checks once per day (saves network requests)
+	# - Optional: Can be disabled with BATPLOT_NO_VERSION_CHECK=1 environment variable
+	# - Silent failure: If check fails, program continues normally
+	#
+	# Why check for updates?
+	# - Users get notified of bug fixes and new features
+	# - Helps maintain an up-to-date installation
+	# ====================================================================
+	try:
+		from . import __version__
+		from .version_check import check_for_updates
+		check_for_updates(__version__)
+	except Exception:
+		# Silently ignore any errors in version checking.
+		# We NEVER want version checking to crash the main program.
+		# If PyPI is down, network is slow, or any other error occurs,
+		# the program should continue normally.
+		pass
+	
+	# ====================================================================
+	# STEP 2: PREPARE ARGUMENTS (TESTING MODE SUPPORT)
+	# ====================================================================
+	# If argv is provided (testing/programmatic mode), temporarily replace
+	# sys.argv so argparse reads from our test arguments instead of real
+	# command line.
+	#
+	# Why? This allows unit tests to call main() with fake arguments without
+	# actually running from command line.
+	#
+	# Example:
+	#   main(['--cv', 'test.mpt'])  # Test with fake arguments
+	#   # sys.argv temporarily becomes ['batplot', '--cv', 'test.mpt']
+	# ====================================================================
+	if argv is not None:
+		old_argv = sys.argv  # Save original for restoration
+		sys.argv = ['batplot'] + list(argv)  # Replace with test arguments
+		
+	try:
+		# ====================================================================
+		# STEP 3: DELEGATE TO MAIN FUNCTION
+		# ====================================================================
+		# Import here (lazy import) to avoid side effects at module import time.
+		# batplot_main() will:
+		#   1. Parse command-line arguments using args.parse_args()
+		#   2. Route to appropriate mode handler (XY, EC, Operando, Batch)
+		#   3. Handle mode-specific logic
+		# ====================================================================
+		from .batplot import batplot_main
+		return batplot_main()
+		
+	except ValueError as e:
+		# ====================================================================
+		# STEP 4: ERROR HANDLING
+		# ====================================================================
+		# ValueError is used for user-facing errors:
+		# - Bad command-line arguments
+		# - File not found
+		# - Invalid file format
+		# - Missing required parameters (e.g., --mass for .mpt files)
+		#
+		# We print a clean error message without scary Python traceback.
+		# This makes errors more user-friendly.
+		#
+		# Other exceptions (unexpected bugs) will still show full traceback
+		# for debugging purposes.
+		# ====================================================================
+		print(f"Error: {e}", file=sys.stderr)
+		return 1  # Non-zero exit code indicates error
+		
+	finally:
+		# ====================================================================
+		# CLEANUP: RESTORE ORIGINAL sys.argv
+		# ====================================================================
+		# If we modified sys.argv for testing, restore it now.
+		# This ensures tests don't affect each other.
+		# ====================================================================
+		if argv is not None:
+			sys.argv = old_argv
+
+__all__ = ["main"]

batplot_backup_20251221_101150/color_utils.py

@@ -0,0 +1,547 @@
+"""Shared helpers for color previews and user-defined color management.
+
+COLOR PALETTE SYSTEM OVERVIEW:
+==============================
+This module manages how colors are assigned to multiple curves/lines in batplot plots.
+
+HOW COLOR PALETTES WORK:
+------------------------
+When you have many curves (e.g., 100 files in XY mode, or 50 cycles in EC mode), you want
+each one to have a different color that smoothly transitions across a color palette.
+
+Example: Using 'viridis' palette with 5 curves:
+    Curve 1 → Dark purple (start of viridis)
+    Curve 2 → Blue-purple
+    Curve 3 → Green
+    Curve 4 → Yellow-green
+    Curve 5 → Bright yellow (end of viridis)
+
+The system works by:
+1. Getting a continuous colormap (e.g., 'viridis')
+2. Sampling colors at evenly spaced points along the colormap
+3. Assigning each sampled color to a different curve
+
+For 100 curves, we sample 100 evenly spaced points from the colormap, ensuring each
+curve gets a unique, smoothly varying color.
+
+COLORMAP SOURCES:
+----------------
+1. Matplotlib built-in: 'viridis', 'plasma', 'inferno', 'magma', etc.
+2. cmcrameri scientific colormaps: 'batlow', 'batlowk', 'batloww' (if installed)
+3. Custom colormaps: Defined in _CUSTOM_CMAPS dictionary below
+
+REVERSED COLORMAPS:
+------------------
+Colormaps can be reversed by adding '_r' suffix:
+    'viridis' → normal (dark to bright)
+    'viridis_r' → reversed (bright to dark)
+
+This is useful when you want the color order flipped.
+"""
+
+from __future__ import annotations
+
+from typing import List, Optional, Sequence
+
+import matplotlib.pyplot as plt
+from matplotlib import colors as mcolors
+from matplotlib.colors import LinearSegmentedColormap
+
+from .config import get_user_colors as _cfg_get_user_colors
+from .config import save_user_colors as _cfg_save_user_colors
+
+# ====================================================================================
+# CUSTOM COLORMAP DEFINITIONS
+# ====================================================================================
+# These are custom color palettes designed for scientific visualization.
+# Each colormap is defined as a list of hex color codes that smoothly transition
+# from one color to the next.
+#
+# Format: List of hex color strings, ordered from start to end of colormap
+# Example: ['#02121d', '#053061', ...] means start with very dark blue, end with light yellow
+#
+# These colormaps are registered with matplotlib so they can be used like built-in ones.
+# ====================================================================================
+_CUSTOM_CMAPS = {
+    # 'batlow' - Scientific colormap optimized for colorblind accessibility
+    # Colors transition: dark blue → teal → green → yellow
+    'batlow': ['#02121d', '#053061', '#2b7a8b', '#7cbf7b', '#c7e6a2', '#f9f0c3'],
+    
+    # 'batlowk' - Variant with more purple tones
+    # Colors transition: dark purple → purple → brown → yellow
+    'batlowk': ['#150b2d', '#3d2e63', '#5f4f85', '#81718f', '#a6938e', '#cbb58f', '#efd78d'],
+    
+    # 'batloww' - Variant with more blue-green tones
+    # Colors transition: dark blue → blue → teal → green → yellow
+    'batloww': ['#0a1427', '#17385d', '#295f8d', '#4f8fa3', '#7db7a1', '#b2d39a', '#e3e6a8'],
+}
+
+
+def ensure_colormap(name: Optional[str]) -> bool:
+    """
+    Ensure that a named colormap exists and is registered with matplotlib.
+    
+    HOW IT WORKS:
+    ------------
+    This function checks if a colormap is available, and if not, tries to register it.
+    It searches in this order:
+    1. Built-in matplotlib colormaps (viridis, plasma, etc.)
+    2. cmcrameri scientific colormaps (if package is installed)
+    3. Custom colormaps defined in _CUSTOM_CMAPS
+    4. Any other matplotlib-compatible colormap
+    
+    WHY THIS IS NEEDED:
+    -------------------
+    Different colormap sources need to be registered with matplotlib before they can
+    be used. This function ensures the colormap is available regardless of its source.
+    
+    Args:
+        name: Colormap name (e.g., 'viridis', 'batlow', 'viridis_r')
+              '_r' suffix indicates reversed colormap
+    
+    Returns:
+        True if colormap exists and is registered, False otherwise
+    """
+    if not name:
+        return False
+    
+    # Handle reversed colormaps (remove '_r' suffix to get base name)
+    # Example: 'viridis_r' → base = 'viridis', we'll reverse it later if needed
+    base = name[:-2] if name.lower().endswith('_r') else name
+    base_lower = base.lower()
+    
+    # STEP 1: Check if it's already a registered matplotlib colormap
+    if base_lower in plt.colormaps():
+        return True
+    
+    # STEP 2: Try to load from cmcrameri package (scientific colormaps)
+    # cmcrameri is an optional package with colorblind-friendly colormaps
+    try:
+        import cmcrameri.cm as cmc
+        if hasattr(cmc, base_lower):
+            cmap_obj = getattr(cmc, base_lower)
+            try:
+                # Register it with matplotlib so it can be used like built-in colormaps
+                plt.register_cmap(name=base_lower, cmap=cmap_obj)
+            except ValueError:
+                # Already registered, that's fine
+                pass
+            return True
+    except Exception:
+        # cmcrameri not installed or colormap not found, continue to next step
+        pass
+    
+    # STEP 3: Check if it's a custom colormap defined in this module
+    custom = _CUSTOM_CMAPS.get(base_lower)
+    if custom:
+        try:
+            # Create a LinearSegmentedColormap from the list of colors
+            # N=256 means create 256 intermediate colors by interpolating between the given colors
+            # This creates a smooth gradient
+            cmap_obj = LinearSegmentedColormap.from_list(base_lower, custom, N=256)
+            try:
+                # Register with matplotlib
+                plt.register_cmap(name=base_lower, cmap=cmap_obj)
+            except ValueError:
+                # Already registered, that's fine
+                pass
+            return True
+        except Exception:
+            return False
+    
+    # STEP 4: Final fallback - try to get it directly from matplotlib
+    # This handles any other matplotlib-compatible colormap
+    try:
+        plt.get_cmap(base_lower)
+        return True
+    except Exception:
+        return False
+
+
+def _ansi_color_block_from_rgba(rgba) -> str:
+    """Return a two-space block with the given RGBA color."""
+    try:
+        r, g, b, _ = rgba
+        r_i = max(0, min(255, int(round(r * 255))))
+        g_i = max(0, min(255, int(round(g * 255))))
+        b_i = max(0, min(255, int(round(b * 255))))
+        return f"\033[48;2;{r_i};{g_i};{b_i}m  \033[0m"
+    except Exception:
+        return "[??]"
+
+
+def color_block(color: Optional[str]) -> str:
+    """Return a colored block (ANSI) for the supplied color string."""
+    if not color:
+        return "[--]"
+    try:
+        rgba = mcolors.to_rgba(color)
+        return _ansi_color_block_from_rgba(rgba)
+    except Exception:
+        return "[??]"
+
+
+def color_bar(colors: Sequence[str]) -> str:
+    """Return a string of adjacent color blocks."""
+    blocks = [color_block(col) for col in colors if col]
+    return " ".join(blocks)
+
+
+def palette_preview(name: str, steps: int = 8) -> str:
+    """
+    Return a visual preview of a colormap as colored blocks in the terminal.
+    
+    HOW IT WORKS:
+    ------------
+    This function samples colors from a colormap at evenly spaced intervals and
+    displays them as colored blocks. This lets users see what the colormap looks
+    like before applying it to their data.
+    
+    Example output for 'viridis' with 8 steps:
+        [dark purple block] [purple block] [blue block] [green block] [yellow block] ...
+    
+    SAMPLING METHOD:
+    ---------------
+    For a colormap with N steps:
+    - Step 0: Sample at position 0.0 (start of colormap)
+    - Step 1: Sample at position 1/(N-1)
+    - Step 2: Sample at position 2/(N-1)
+    - ...
+    - Step N-1: Sample at position 1.0 (end of colormap)
+    
+    This gives evenly distributed colors across the entire colormap range.
+    
+    Args:
+        name: Colormap name (e.g., 'viridis', 'plasma', 'batlow')
+        steps: Number of color samples to show (default 8)
+               More steps = more detailed preview but longer output
+    
+    Returns:
+        String of ANSI color codes that display as colored blocks in terminal
+        Empty string if colormap not found
+    """
+    # Ensure colormap is registered
+    ensure_colormap(name)
+    
+    # Try to get the colormap from matplotlib
+    try:
+        cmap = plt.get_cmap(name)
+    except Exception:
+        cmap = None
+        # Fallback: try cmcrameri if it's a batlow variant
+        lower = name.lower()
+        if lower.startswith('batlow'):
+            try:
+                import cmcrameri.cm as cmc
+                if hasattr(cmc, lower):
+                    cmap = getattr(cmc, lower)
+                elif hasattr(cmc, 'batlow'):
+                    cmap = getattr(cmc, 'batlow')
+            except Exception:
+                return ""
+        else:
+            return ""
+    
+    # Ensure steps is at least 1 (avoid division by zero)
+    if steps < 1:
+        steps = 1
+    
+    # Special handling for tab10 to use hardcoded colors (matching EC and CPC interactive)
+    if name.lower() == 'tab10':
+        default_tab10_colors = ['#1f77b4', '#ff7f0e', '#2ca02c', '#d62728', '#9467bd',
+                               '#8c564b', '#e377c2', '#7f7f7f', '#bcbd22', '#17becf']
+        # Use first 'steps' colors from tab10
+        samples = [default_tab10_colors[i % len(default_tab10_colors)] for i in range(steps)]
+    else:
+        # Sample colors at evenly spaced positions along the colormap
+        # Position ranges from 0.0 (start) to 1.0 (end)
+        samples = [
+            mcolors.to_hex(cmap(i / max(steps - 1, 1)))  # Convert to hex color code
+            for i in range(steps)
+        ]
+    
+    # Convert color codes to visual blocks and return as string
+    return color_bar(samples)
+
+
+def _set_cached_colors(fig, colors: List[str]):
+    """
+    Store user colors in figure object for fast access (caching).
+    
+    HOW CACHING WORKS:
+    -----------------
+    Instead of reading from disk every time, we store colors in the figure object.
+    This is faster because:
+    - Reading from disk is slow (file I/O)
+    - Figure object is already in memory (fast access)
+    
+    WHY STORE IN FIGURE OBJECT?
+    ---------------------------
+    The figure object persists throughout the interactive session, so we can
+    cache colors there. This avoids repeated file reads.
+    
+    Args:
+        fig: Matplotlib figure object (where we store the cache)
+        colors: List of color codes to cache
+    """
+    if fig is not None:
+        # setattr() dynamically adds an attribute to an object
+        # This is like: fig._user_colors_cache = list(colors)
+        # We use list() to create a copy (not a reference to original list)
+        setattr(fig, '_user_colors_cache', list(colors))
+
+
+def get_user_color_list(fig=None) -> List[str]:
+    """
+    Return cached user colors (persisted to ~/.batplot).
+    
+    HOW IT WORKS:
+    ------------
+    1. First check if colors are cached in figure object (fast path)
+    2. If not cached, load from disk (~/.batplot/config.json)
+    3. Cache the loaded colors in figure object for next time
+    4. Return the color list
+    
+    WHY CACHING?
+    -----------
+    - Fast: Memory access is much faster than disk I/O
+    - Efficient: Only reads from disk once per session
+    - Persistent: Colors are saved to disk, so they persist between sessions
+    
+    Args:
+        fig: Matplotlib figure object (optional, for caching)
+    
+    Returns:
+        List of color codes (hex strings like '#FF0000' or named colors like 'red')
+    """
+    # Check if colors are already cached in figure object
+    if fig is not None and hasattr(fig, '_user_colors_cache'):
+        # Return cached colors (fast path - no disk access)
+        # list() creates a copy so caller can't modify the cached version
+        return list(getattr(fig, '_user_colors_cache'))
+    
+    # Not cached - load from disk
+    colors = list(_cfg_get_user_colors())
+    # Cache for next time
+    _set_cached_colors(fig, colors)
+    return colors
+
+
+def _save_user_colors(colors: List[str], fig=None) -> List[str]:
+    """
+    Save user colors to disk and cache, removing duplicates and empty entries.
+    
+    HOW IT WORKS:
+    ------------
+    1. Remove empty/None colors (filter out invalid entries)
+    2. Remove duplicates (keep only first occurrence of each color)
+    3. Save cleaned list to disk (~/.batplot/config.json)
+    4. Update cache in figure object
+    5. Return cleaned list
+    
+    WHY CLEAN THE LIST?
+    ------------------
+    - Empty strings would cause errors when trying to use them as colors
+    - Duplicates waste space and confuse users
+    - Clean data = better user experience
+    
+    Args:
+        colors: List of color codes (may contain duplicates or empty strings)
+        fig: Matplotlib figure object (optional, for caching)
+    
+    Returns:
+        Cleaned list (no duplicates, no empty entries)
+    """
+    cleaned: List[str] = []  # Type annotation: cleaned is a list of strings
+    for col in colors:
+        # Skip empty/None colors
+        if not col:
+            continue
+        # Only add if not already in cleaned list (removes duplicates)
+        if col not in cleaned:
+            cleaned.append(col)
+    # Save to disk (persists between sessions)
+    _cfg_save_user_colors(cleaned)
+    # Update cache (fast access for current session)
+    _set_cached_colors(fig, cleaned)
+    return cleaned
+
+
+def add_user_color(color: str, fig=None) -> List[str]:
+    """Append a user color (if not already present)."""
+    colors = get_user_color_list(fig)
+    if color and color not in colors:
+        colors.append(color)
+        colors = _save_user_colors(colors, fig)
+    return colors
+
+
+def remove_user_color(index: int, fig=None) -> List[str]:
+    """Remove a user color by 0-based index."""
+    colors = get_user_color_list(fig)
+    if 0 <= index < len(colors):
+        colors.pop(index)
+        colors = _save_user_colors(colors, fig)
+    return colors
+
+
+def clear_user_colors(fig=None) -> None:
+    _save_user_colors([], fig)
+
+
+def resolve_color_token(token: str, fig=None) -> str:
+    """
+    Translate color references like '2' or 'u3' into actual color codes.
+    
+    HOW IT WORKS:
+    ------------
+    Users can reference saved colors in two ways:
+    1. By number: '2' means the 2nd saved color (1-based indexing)
+    2. By 'u' prefix: 'u3' means the 3rd saved color (1-based indexing)
+    
+    Examples:
+        '2' → Returns colors[1] (2nd color, but 0-based index is 1)
+        'u3' → Returns colors[2] (3rd color, but 0-based index is 2)
+        'red' → Returns 'red' (not a reference, so return as-is)
+        '#FF0000' → Returns '#FF0000' (not a reference, so return as-is)
+    
+    WHY TWO FORMATS?
+    ---------------
+    - '2' is shorter and easier to type
+    - 'u3' is more explicit (clearly indicates user color)
+    - Both are 1-based (user-friendly) but converted to 0-based (Python indexing)
+    
+    Args:
+        token: Color reference string (e.g., '2', 'u3', 'red', '#FF0000')
+        fig: Matplotlib figure object (optional, for accessing cached colors)
+    
+    Returns:
+        Actual color code (hex string or named color), or original token if not a reference
+    """
+    # Empty token - return as-is
+    if not token:
+        return token
+    
+    # Remove whitespace
+    stripped = token.strip()
+    idx = None  # Will hold the 0-based index if token is a reference
+    
+    # Check if token is 'u' prefix format (e.g., 'u3')
+    # stripped[1:] gets everything after the first character
+    # .isdigit() checks if it's all digits
+    if stripped.lower().startswith('u') and stripped[1:].isdigit():
+        # Convert 'u3' → index 2 (3rd color, but 0-based)
+        idx = int(stripped[1:]) - 1
+    # Check if token is just a number (e.g., '2')
+    elif stripped.isdigit():
+        # Convert '2' → index 1 (2nd color, but 0-based)
+        idx = int(stripped) - 1
+    
+    # If we found a valid index, look up the color
+    if idx is not None:
+        colors = get_user_color_list(fig)
+        # Check if index is valid (within bounds of color list)
+        if 0 <= idx < len(colors):
+            return colors[idx]  # Return the actual color code
+    
+    # Not a reference, or invalid index - return token as-is
+    return token
+
+
+def print_user_colors(fig=None) -> None:
+    """Print saved colors with indices and color blocks."""
+    colors = get_user_color_list(fig)
+    if not colors:
+        print("No saved user colors.")
+        return
+    print("Saved colors:")
+    for idx, color in enumerate(colors, 1):
+        print(f"  {idx}: {color_block(color)} {color}")
+
+
+def manage_user_colors(fig=None) -> None:
+    """Interactive submenu for editing user-defined colors."""
+    while True:
+        colors = get_user_color_list(fig)
+        print("\n\033[1mUser color list:\033[0m")
+        if colors:
+            for idx, color in enumerate(colors, 1):
+                print(f"  {idx}: {color_block(color)} {color}")
+        else:
+            print("  (empty)")
+        print("Options: a=add colors, d=delete numbers, c=clear, q=back")
+        choice = input("User colors> ").strip().lower()
+        if not choice:
+            continue
+        if choice == 'q':
+            break
+        if choice == 'a':
+            line = input("Enter colors (space-separated names/hex codes) or q: ").strip()
+            if not line or line.lower() == 'q':
+                continue
+            # List comprehension: splits line by spaces, keeps only non-empty tokens
+            # Example: "red blue #FF0000" → ['red', 'blue', '#FF0000']
+            # The 'if tok' part filters out empty strings (from multiple spaces)
+            new_colors = [tok for tok in line.split() if tok]
+            if new_colors:
+                colors = get_user_color_list(fig)
+                added = 0
+                for col in new_colors:
+                    if col not in colors:
+                        colors.append(col)
+                        added += 1
+                _save_user_colors(colors, fig)
+                print(f"Added {added} color(s).")
+            continue
+        if choice == 'd':
+            if not colors:
+                print("No colors to delete.")
+                continue
+            line = input("Enter number(s) to delete (e.g., 1 or 1,3,5): ").strip()
+            if not line:
+                continue
+            tokens = line.replace(',', ' ').split()
+            indices = []
+            for tok in tokens:
+                if tok.isdigit():
+                    idx = int(tok) - 1
+                    if 0 <= idx < len(colors):
+                        indices.append(idx)
+                    else:
+                        print(f"Index out of range: {tok}")
+                else:
+                    print(f"Invalid entry: {tok}")
+            if indices:
+                # Sort indices in reverse order (largest first)
+                # WHY? When deleting multiple items, we must delete from end to start.
+                # If we delete index 1 first, then index 3 becomes index 2, and we'd delete the wrong item!
+                # Example: colors = ['red', 'blue', 'green', 'yellow']
+                #          Delete indices [1, 3] (blue and yellow)
+                #          If we delete 1 first: ['red', 'green', 'yellow'] (index 3 is now out of bounds!)
+                #          If we delete 3 first: ['red', 'blue', 'green'] (then delete 1: ['red', 'green'] ✓)
+                for idx in sorted(indices, reverse=True):
+                    colors.pop(idx)  # Remove color at this index
+                _save_user_colors(colors, fig)
+                print("Updated color list.")
+            continue
+        if choice == 'c':
+            confirm = input("Clear all saved colors? (y/n): ").strip().lower()
+            if confirm == 'y':
+                clear_user_colors(fig)
+                print("Cleared saved colors.")
+            continue
+        print("Unknown option.")
+
+
+__all__ = [
+    'add_user_color',
+    'clear_user_colors',
+    'color_bar',
+    'color_block',
+    'manage_user_colors',
+    'palette_preview',
+    'print_user_colors',
+    'remove_user_color',
+    'resolve_color_token',
+    'get_user_color_list',
+]