natural-pdf 0.1.38__py3-none-any.whl → 0.2.0__py3-none-any.whl

natural_pdf/utils/visualization.py

@@ -34,6 +34,10 @@ _BASE_HIGHLIGHT_COLORS = [
 # Default Alpha for highlight fills
 DEFAULT_FILL_ALPHA = 100
 
+# Add quantitative color mapping functionality
+import matplotlib.cm as cm
+import matplotlib.pyplot as plt
+
 
 class ColorManager:
     """
@@ -176,6 +180,208 @@ def create_legend(
     return legend
 
 
+def create_colorbar(
+    values: List[float],
+    colormap: str = "viridis",
+    bins: Optional[Union[int, List[float]]] = None,
+    width: int = 80,
+    height: int = 20,
+    orientation: str = "horizontal",
+) -> Image.Image:
+    """
+    Create a color bar for quantitative data visualization.
+
+    Args:
+        values: List of numeric values to create color bar for
+        colormap: Name of the matplotlib colormap to use
+        bins: Optional binning specification (int for equal bins, list for custom bins)
+        width: Width of the color bar
+        height: Height of the color bar
+        orientation: 'horizontal' or 'vertical'
+
+    Returns:
+        PIL Image with the color bar
+    """
+    import numpy as np
+
+    # Get value range
+    vmin = min(values)
+    vmax = max(values)
+
+    if vmin == vmax:
+        # Handle edge case where all values are the same
+        vmax = vmin + 1
+
+    # Create the colorbar image
+    if orientation == "horizontal":
+        bar_width = width - 40  # Leave space for labels (reduced from 60)
+        bar_height = height
+        total_width = width
+        total_height = height + 40  # Extra space for labels
+    else:
+        bar_width = width
+        bar_height = max(height, 120)  # Ensure minimum height for vertical colorbar
+        total_width = width + 80  # Extra space for labels (increased for larger text)
+        total_height = bar_height + 60  # Extra space for labels
+
+    # Create base image
+    img = Image.new("RGBA", (total_width, total_height), (255, 255, 255, 255))
+    draw = ImageDraw.Draw(img)
+
+    # Try to load a font
+    try:
+        font = ImageFont.truetype("DejaVuSans.ttf", 16)
+    except IOError:
+        try:
+            font = ImageFont.truetype("Arial.ttf", 16)
+        except IOError:
+            # Load default font but try to get a larger size
+            try:
+                font = ImageFont.load_default(size=16)
+            except:
+                font = ImageFont.load_default()
+
+    # Draw the color blocks (5 discrete blocks)
+    if orientation == "horizontal":
+        # Create 5 discrete color blocks
+        num_blocks = 5
+        block_width = bar_width // num_blocks
+
+        for i in range(num_blocks):
+            # Calculate value for this block (center of block)
+            block_start = i / num_blocks
+            block_center = (i + 0.5) / num_blocks
+            value = vmin + block_center * (vmax - vmin)
+
+            # Get color for this block
+            rgb = get_colormap_color(colormap, value, vmin, vmax)
+            color = (*rgb, 255)
+
+            # Calculate block position
+            x_start = 20 + i * block_width
+            x_end = 20 + (i + 1) * block_width
+
+            # Draw filled rectangle for this block
+            draw.rectangle(
+                [(x_start, 10), (x_end, 10 + bar_height)],
+                fill=color,
+                outline=(0, 0, 0, 255),
+                width=1,
+            )
+
+        # Add value labels
+        if bins is not None:
+            # Show bin boundaries
+            if isinstance(bins, int):
+                # Equal-width bins
+                step = (vmax - vmin) / bins
+                tick_values = [vmin + i * step for i in range(bins + 1)]
+            else:
+                # Custom bins
+                tick_values = bins
+
+            for tick_val in tick_values:
+                if vmin <= tick_val <= vmax:
+                    x_pos = int(20 + (tick_val - vmin) / (vmax - vmin) * bar_width)
+                    # Draw tick mark
+                    draw.line(
+                        [(x_pos, 10 + bar_height), (x_pos, 10 + bar_height + 5)],
+                        fill=(0, 0, 0, 255),
+                        width=1,
+                    )
+                    # Draw label
+                    label_text = f"{tick_val:.2f}".rstrip("0").rstrip(".")
+                    text_bbox = draw.textbbox((0, 0), label_text, font=font)
+                    text_width = text_bbox[2] - text_bbox[0]
+                    draw.text(
+                        (x_pos - text_width // 2, 10 + bar_height + 8),
+                        label_text,
+                        fill=(0, 0, 0, 255),
+                        font=font,
+                    )
+        else:
+            # Show min and max values
+            # Min value
+            min_text = f"{vmin:.2f}".rstrip("0").rstrip(".")
+            draw.text((20, 10 + bar_height + 8), min_text, fill=(0, 0, 0, 255), font=font)
+
+            # Max value
+            max_text = f"{vmax:.2f}".rstrip("0").rstrip(".")
+            text_bbox = draw.textbbox((0, 0), max_text, font=font)
+            text_width = text_bbox[2] - text_bbox[0]
+            draw.text(
+                (20 + bar_width - text_width, 10 + bar_height + 8),
+                max_text,
+                fill=(0, 0, 0, 255),
+                font=font,
+            )
+
+    else:  # vertical orientation
+        # Create 5 discrete color blocks
+        num_blocks = 5
+        block_height = bar_height // num_blocks
+
+        for i in range(num_blocks):
+            # Calculate value for this block (center of block, top = max, bottom = min)
+            block_center = (i + 0.5) / num_blocks
+            value = vmax - block_center * (vmax - vmin)
+
+            # Get color for this block
+            rgb = get_colormap_color(colormap, value, vmin, vmax)
+            color = (*rgb, 255)
+
+            # Calculate block position
+            y_start = 30 + i * block_height
+            y_end = 30 + (i + 1) * block_height
+
+            # Draw filled rectangle for this block
+            draw.rectangle(
+                [(10, y_start), (10 + bar_width, y_end)],
+                fill=color,
+                outline=(0, 0, 0, 255),
+                width=1,
+            )
+
+        # Add value labels
+        if bins is not None:
+            # Show bin boundaries
+            if isinstance(bins, int):
+                # Equal-width bins
+                step = (vmax - vmin) / bins
+                tick_values = [vmin + i * step for i in range(bins + 1)]
+            else:
+                # Custom bins
+                tick_values = bins
+
+            for tick_val in tick_values:
+                if vmin <= tick_val <= vmax:
+                    y_pos = int(30 + (vmax - tick_val) / (vmax - vmin) * bar_height)
+                    # Draw tick mark
+                    draw.line(
+                        [(10 + bar_width, y_pos), (10 + bar_width + 5, y_pos)],
+                        fill=(0, 0, 0, 255),
+                        width=1,
+                    )
+                    # Draw label
+                    label_text = f"{tick_val:.2f}".rstrip("0").rstrip(".")
+                    draw.text(
+                        (10 + bar_width + 8, y_pos - 6), label_text, fill=(0, 0, 0, 255), font=font
+                    )
+        else:
+            # Show min and max values
+            # Max value (top)
+            max_text = f"{vmax:.2f}".rstrip("0").rstrip(".")
+            draw.text((10 + bar_width + 8, 30 - 6), max_text, fill=(0, 0, 0, 255), font=font)
+
+            # Min value (bottom)
+            min_text = f"{vmin:.2f}".rstrip("0").rstrip(".")
+            draw.text(
+                (10 + bar_width + 8, 30 + bar_height - 6), min_text, fill=(0, 0, 0, 255), font=font
+            )
+
+    return img
+
+
 def merge_images_with_legend(
     image: Image.Image, legend: Image.Image, position: str = "right"
 ) -> Image.Image:
@@ -262,3 +468,182 @@ def render_plain_page(page, resolution):
     doc.close()
 
     return image
+
+
+def detect_quantitative_data(values: List[Any]) -> bool:
+    """
+    Detect if a list of values represents quantitative data suitable for gradient coloring.
+
+    Args:
+        values: List of attribute values from elements
+
+    Returns:
+        True if data appears to be quantitative, False otherwise
+    """
+    # Filter out None values
+    numeric_values = []
+    for v in values:
+        if v is not None:
+            try:
+                # Try to convert to float
+                numeric_values.append(float(v))
+            except (ValueError, TypeError):
+                # Not numeric, likely categorical
+                pass
+
+    # If we have fewer than 2 numeric values, treat as categorical
+    if len(numeric_values) < 2:
+        return False
+
+    # If more than 80% of values are numeric and we have >8 unique values, treat as quantitative
+    numeric_ratio = len(numeric_values) / len(values)
+    unique_values = len(set(numeric_values))
+
+    return numeric_ratio > 0.8 and unique_values > 8
+
+
+def get_colormap_color(
+    colormap_name: str, value: float, vmin: float, vmax: float
+) -> Tuple[int, int, int]:
+    """
+    Get a color from a matplotlib colormap based on a normalized value.
+
+    Args:
+        colormap_name: Name of the colormap ('viridis', 'plasma', etc.)
+        value: The value to map to a color
+        vmin: Minimum value in the data range
+        vmax: Maximum value in the data range
+
+    Returns:
+        RGB color tuple (0-255)
+    """
+    # Try to get the colormap from matplotlib
+    try:
+        cmap = cm.get_cmap(colormap_name)
+    except (ValueError, KeyError):
+        # Fallback to viridis if colormap doesn't exist
+        cmap = cm.get_cmap("viridis")
+
+    # Normalize value to [0, 1]
+    if vmax == vmin:
+        t = 0.0
+    else:
+        t = (value - vmin) / (vmax - vmin)
+
+    # Clamp to [0, 1]
+    t = max(0.0, min(1.0, t))
+
+    # Get RGBA color from matplotlib (values are 0-1)
+    rgba = cmap(t)
+
+    # Convert to 0-255 RGB
+    r = int(rgba[0] * 255)
+    g = int(rgba[1] * 255)
+    b = int(rgba[2] * 255)
+
+    return (r, g, b)
+
+
+def apply_bins_to_values(
+    values: List[float], bins: Union[int, List[float]]
+) -> Tuple[List[str], List[float]]:
+    """
+    Apply binning to quantitative values.
+
+    Args:
+        values: List of numeric values
+        bins: Either number of bins (int) or list of bin edges (List[float])
+
+    Returns:
+        Tuple of (bin_labels, bin_values) where bin_values are the centers of bins
+    """
+    if isinstance(bins, int):
+        # Equal-width bins
+        min_val = min(values)
+        max_val = max(values)
+        bin_edges = [min_val + i * (max_val - min_val) / bins for i in range(bins + 1)]
+    else:
+        # Custom bin edges
+        bin_edges = sorted(bins)
+
+    # Create bin labels and centers
+    bin_labels = []
+    bin_centers = []
+    for i in range(len(bin_edges) - 1):
+        start = bin_edges[i]
+        end = bin_edges[i + 1]
+        bin_labels.append(f"{start:.2f}-{end:.2f}")
+        bin_centers.append((start + end) / 2)
+
+    return bin_labels, bin_centers
+
+
+def create_quantitative_color_mapping(
+    values: List[Any], colormap: str = "viridis", bins: Optional[Union[int, List[float]]] = None
+) -> Dict[Any, Tuple[int, int, int, int]]:
+    """
+    Create a color mapping for quantitative data using matplotlib colormaps.
+
+    Args:
+        values: List of values to map to colors
+        colormap: Name of any matplotlib colormap (e.g., 'viridis', 'plasma', 'inferno',
+                 'magma', 'coolwarm', 'RdBu', 'tab10', etc.). See matplotlib.cm for full list.
+        bins: Optional binning specification (int for equal-width bins, list for custom bins)
+
+    Returns:
+        Dictionary mapping values to RGBA colors
+    """
+    # Convert to numeric values, filtering out None/non-numeric
+    numeric_values = []
+    value_to_numeric = {}
+
+    for v in values:
+        if v is not None:
+            try:
+                numeric_val = float(v)
+                numeric_values.append(numeric_val)
+                value_to_numeric[v] = numeric_val
+            except (ValueError, TypeError):
+                pass
+
+    if not numeric_values:
+        # Fallback to categorical if no numeric values
+        return {}
+
+    # Determine min/max for normalization
+    vmin = min(numeric_values)
+    vmax = max(numeric_values)
+
+    # Apply binning if specified
+    if bins is not None:
+        bin_labels, bin_centers = apply_bins_to_values(numeric_values, bins)
+        # Create mapping from original values to bin centers
+        result = {}
+        for orig_val, numeric_val in value_to_numeric.items():
+            # Find which bin this value belongs to
+            if isinstance(bins, int):
+                bin_width = (vmax - vmin) / bins
+                bin_idx = min(int((numeric_val - vmin) / bin_width), bins - 1)
+            else:
+                bin_idx = 0
+                for i, edge in enumerate(bins[1:], 1):
+                    if numeric_val <= edge:
+                        bin_idx = i - 1
+                        break
+                else:
+                    bin_idx = len(bins) - 2
+
+            # Get color for this bin center
+            bin_center = bin_centers[bin_idx]
+            rgb = get_colormap_color(colormap, bin_center, vmin, vmax)
+            result[orig_val] = (*rgb, DEFAULT_FILL_ALPHA)
+
+        return result
+    else:
+        # Continuous gradient mapping
+        result = {}
+        for orig_val, numeric_val in value_to_numeric.items():
+            rgb = get_colormap_color(colormap, numeric_val, vmin, vmax)
+            result[orig_val] = (*rgb, DEFAULT_FILL_ALPHA)
+
+        return result

{natural_pdf-0.1.38.dist-info → natural_pdf-0.2.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: natural-pdf
-Version: 0.1.38
+Version: 0.2.0
 Summary: A more intuitive interface for working with PDFs
 Author-email: Jonathan Soma <jonathan.soma@gmail.com>
 License-Expression: MIT
@@ -26,6 +26,7 @@ Requires-Dist: jenkspy
 Requires-Dist: scipy
 Requires-Dist: ipywidgets>=7.0.0
 Requires-Dist: python-bidi
+Requires-Dist: matplotlib
 Provides-Extra: test
 Requires-Dist: pytest; extra == "test"
 Requires-Dist: pytest-xdist; extra == "test"
@@ -43,7 +44,8 @@ Requires-Dist: uv; extra == "dev"
 Requires-Dist: pipdeptree; extra == "dev"
 Requires-Dist: nbformat; extra == "dev"
 Requires-Dist: jupytext; extra == "dev"
-Requires-Dist: nbclient; extra == "dev"
+Requires-Dist: nbclient==0.10.2; extra == "dev"
+Requires-Dist: jupyter_core==5.7.2; extra == "dev"
 Requires-Dist: ipykernel; extra == "dev"
 Requires-Dist: pre-commit; extra == "dev"
 Requires-Dist: setuptools; extra == "dev"
@@ -52,7 +54,6 @@ Requires-Dist: natural-pdf[ocr-export]; extra == "all"
 Requires-Dist: natural-pdf[deskew]; extra == "all"
 Requires-Dist: natural-pdf[test]; extra == "all"
 Requires-Dist: natural-pdf[search]; extra == "all"
-Requires-Dist: natural-pdf[favorites]; extra == "all"
 Requires-Dist: natural-pdf[export-extras]; extra == "all"
 Requires-Dist: natural-pdf[ai]; extra == "all"
 Provides-Extra: deskew
@@ -75,10 +76,13 @@ Requires-Dist: huggingface_hub>=0.29.3; extra == "ai"
 Requires-Dist: timm; extra == "ai"
 Requires-Dist: doclayout_yolo; extra == "ai"
 Requires-Dist: easyocr; extra == "ai"
+Requires-Dist: openai; extra == "ai"
 Dynamic: license-file
 
 # Natural PDF
 
+[![CI](https://github.com/jsoma/natural-pdf/actions/workflows/ci.yml/badge.svg)](https://github.com/jsoma/natural-pdf/actions/workflows/ci.yml)
+
 A friendly library for working with PDFs, built on top of [pdfplumber](https://github.com/jsvine/pdfplumber).
 
 Natural PDF lets you find and extract content from PDFs using simple code that makes sense.

{natural_pdf-0.1.38.dist-info → natural_pdf-0.2.0.dist-info}/RECORD

@@ -1,66 +1,69 @@
-natural_pdf/__init__.py,sha256=qDFJNF8sbEDO-2WSFAxoWEM8updOUP6dB-ckya0kxfs,3275
+natural_pdf/__init__.py,sha256=N4pR0LbuPEnUYFZqbdVqc_FGKldgwPQc1wjJhYKTBBM,3417
 natural_pdf/cli.py,sha256=SkPwhhMM-GhLsj3O1n1Agxz4KOxcZ08sj8hVQSFJB5c,4064
-natural_pdf/analyzers/__init__.py,sha256=M5oD4oEsIBWrzgTaeg4uBYrC4OPd7Mp7tz8ootKN_l8,1134
-natural_pdf/analyzers/guides.py,sha256=D55ov45PE7mhqvEnarn82y5hG6gmDzk7tYw233LnluA,141896
-natural_pdf/analyzers/shape_detection_mixin.py,sha256=Ef1o73QYVXQ2QcQMM_W9XRwY6vaIQHgxzD7etJ6LbiM,62820
+natural_pdf/text_mixin.py,sha256=eFCiHj6Okcw3aum4955BepcI2NPRalkf9UFFVTc_H30,4012
+natural_pdf/analyzers/__init__.py,sha256=3XGoNq3OgiVkZP7tOdeP5XVUl7fDgyztdA8DlOcMLXg,1138
+natural_pdf/analyzers/guides.py,sha256=N8fetR3jrDXzeHtIlbxg8BEbthB_lS0L8yhzVXHqiGQ,143245
+natural_pdf/analyzers/shape_detection_mixin.py,sha256=mgpyJ4jIulz9l9HCqThabJIsLSrXh9BB2AmLxUoHmw0,62584
 natural_pdf/analyzers/text_options.py,sha256=qEkDaYWla0rIM_gszEOsu52q7C_dAfV81P2HLJZM2sw,3333
-natural_pdf/analyzers/text_structure.py,sha256=VfKTsTFrK877sC0grsis9jK3rrgp0Mbp13VWEbukTcs,28437
+natural_pdf/analyzers/text_structure.py,sha256=3WWusi-BI0krUnJxB05DD6XmKj5qRNvQBqH7zOQGm1M,28451
 natural_pdf/analyzers/utils.py,sha256=PYbzJzSAHZ7JsMes84WIrSbA0zkjJGs0CLvIeINsf_k,2100
 natural_pdf/analyzers/layout/__init__.py,sha256=oq1uJ5UkGGMbBKGirV1aRKK3hxAUyjTLywYkPCQH1f0,33
 natural_pdf/analyzers/layout/base.py,sha256=F5xPOJcI65N4nxwm0szvhtbDD6lVMqWDut8PSkTCobU,8349
 natural_pdf/analyzers/layout/docling.py,sha256=4BJYyNVR6VegZGxyisvNIBBRvVk6YKPyDVs7ZdVfzEU,12676
 natural_pdf/analyzers/layout/gemini.py,sha256=ldECVCQ5HNQA3Omjg2NOsTrJXslyYb0vErDncmLIiuE,10510
-natural_pdf/analyzers/layout/layout_analyzer.py,sha256=Ff7OfjMyGIWkfPKl6dHgkFyb-iru83_VDk0gmvyHbbg,15549
-natural_pdf/analyzers/layout/layout_manager.py,sha256=ivMfTPjS14dyISu2o2Q6K48jkftvAOD04aCOtInkZGo,10267
+natural_pdf/analyzers/layout/layout_analyzer.py,sha256=taYM6uE-Lvp35hwv6XvPzrUIjAs9DfroA9Ndjy9bWsg,15553
+natural_pdf/analyzers/layout/layout_manager.py,sha256=j6joIgOkeooxv9Kr9ELomm9-5dWlPvNTMzYa_-pwUAQ,10825
 natural_pdf/analyzers/layout/layout_options.py,sha256=2JENtBMHhP3hP0zpFI5-UP3-t1y49E7oLZnjd9d1eB0,7704
-natural_pdf/analyzers/layout/paddle.py,sha256=44GG1sbaYTgvmtnrckNaCbDaNyw_D7FLLiSvzKP2cbk,23048
+natural_pdf/analyzers/layout/paddle.py,sha256=WDXq-J2XXC0DW2LSe7sDtQstgPIGODnLjv8FCiWDXt0,23604
 natural_pdf/analyzers/layout/pdfplumber_table_finder.py,sha256=Tk0Q7wv7nGYPo69lh6RoezjdepTnMl90SaNIrP29Pwc,5902
-natural_pdf/analyzers/layout/surya.py,sha256=ugRXPIHiLoh65lfbbiXO317TbgdtQ-5kVN1nonEf4ws,9778
+natural_pdf/analyzers/layout/surya.py,sha256=lXgn5bGVv-YIDTQod5NYViuJjYa5tM_qlEUc3dzWe28,9786
 natural_pdf/analyzers/layout/table_structure_utils.py,sha256=_sugFWvVpRK3EimOCrikTDAalGnSaWqiqFbtJw8t-lg,2770
 natural_pdf/analyzers/layout/tatr.py,sha256=cVr0ZyhY2mNLAKZ4DGMm-b7XNJpILKh8x8ZpyDeUhLk,15032
 natural_pdf/analyzers/layout/yolo.py,sha256=2Iz2-WsMy--ftkZQ8j5PGqp_1fTD7Mskl2kNnMUuwCU,8286
-natural_pdf/classification/manager.py,sha256=wyENltPSeWpJNjqzU91-ydJTnACZ_LC1q-ox_tRhMIM,22172
+natural_pdf/classification/manager.py,sha256=BaqBL9GeMvYgoJsiQeI2J8aUKQ5Qxu_ELRvmCWquld8,22172
 natural_pdf/classification/mixin.py,sha256=CXygXXhe_qx1563SmIjiu4uSnZkxCkuRR4fGvLokS2w,9416
 natural_pdf/classification/results.py,sha256=5ha77CxK0GYwkBMJbvUBZkBjsL5GpOveIZDK9nO4j8I,3239
-natural_pdf/collections/mixins.py,sha256=sj76Cn6EdBtb5f-bdAV-1qpdixX8tI4BzPccPiYLI1w,5117
-natural_pdf/collections/pdf_collection.py,sha256=sDVEbFMNME_2OaHIsCoR_W7V1cAATNw4ZRqKWa6nbqA,30131
+natural_pdf/collections/mixins.py,sha256=u4KtnlUZZYQ74e0OXAniOv9RtuA6FhwBxsLMJLjdbpQ,5169
 natural_pdf/core/__init__.py,sha256=QC8H4M3KbXwMFiQORZ0pdPlzx1Ix6oKKQSS7Ib2KEaA,38
 natural_pdf/core/element_manager.py,sha256=DRZvntd99wjXy6KeDjCq5uRhjMftZop9QklOZqlUH8M,55349
-natural_pdf/core/highlighting_service.py,sha256=2tBrrEq6d6hz5f6Yf7z5TysJdlTyuHTURBnQxokJnDM,40645
-natural_pdf/core/page.py,sha256=iWokHLuSrQ71kxB_tTWkCp_O-i72urR4iGFUIzKoH8k,145351
-natural_pdf/core/pdf.py,sha256=5M1gB9psqwJCgE0w7PQ_G1XVa_XCmyNNmluZO7pIyZ4,97112
+natural_pdf/core/highlighting_service.py,sha256=k_SMCINeK4aUwfQLmaiyipCPL8vv33ibrCyqtlni8Bc,67921
+natural_pdf/core/page.py,sha256=nQDUR4eKsUhPmEnofjmJRPITQ1RJoK3ITC0Lrtt4AHw,135510
+natural_pdf/core/page_collection.py,sha256=9ff7IfO04bUkJCBZv__Z9G8A-NY7mR3OujVl54lH-FE,50985
+natural_pdf/core/pdf.py,sha256=q54DyhXwAS_zAmsBd3PsCezu1wyQOYmGmB3iKfP8gAM,101884
+natural_pdf/core/pdf_collection.py,sha256=8tM0qVWS1L5Hwv5cXuZ2X8znAYOjKmlERX62bksDlJU,30144
+natural_pdf/core/render_spec.py,sha256=j77UrHA_g_e0RbAyn-4hkjFtqm_oaTe5KRd_Ii9izf4,12243
 natural_pdf/describe/__init__.py,sha256=kIV7ORmWWB1SAur7nK2aAwR-wHqSedhKfUsaUl4hG0A,586
-natural_pdf/describe/base.py,sha256=CLhZXYQO6SOPUVWLt6VwZ7MK48t_6wgPMyFMLtTCKRc,18166
+natural_pdf/describe/base.py,sha256=Of9WVo9XuShXoeyJr0RN2CpLhF_CeiOjazl-or53RKU,18173
 natural_pdf/describe/elements.py,sha256=JicXC9SJmmasqxalpCXA47-kVwv-6JnR3Xiu778aNHM,12634
 natural_pdf/describe/mixin.py,sha256=rkX14aGrSz7Jvxx8Rbxv3eSfbO-_29DipwpstrV2pDQ,3109
 natural_pdf/describe/summary.py,sha256=cfT4ZQkeatCDAOwWPwhtEVXisNgk6E57fAXAnoRysSU,7645
-natural_pdf/elements/__init__.py,sha256=S8XeiNWJ1WcgnyYKdYV1yxQlAxCCO3FfITT8MQwNbyk,41
-natural_pdf/elements/base.py,sha256=-ZAcc8lb2aSWTKcprwKTvnR6hsDGDm7T8a1Y9V38E_A,52042
-natural_pdf/elements/collections.py,sha256=qtHEaLPxZ6i3zPQsbSOw_KMAr9oDMWR1516ilSMSDeY,133189
+natural_pdf/elements/__init__.py,sha256=ICNikmLeIEuSYypz-KnkBn8xR1hR7rge4hsa1KLkyWY,42
+natural_pdf/elements/base.py,sha256=jEBw5cq4mzgOYeEBrWPml2RBuVmOnwBNA4nTd7pLmMI,52292
+natural_pdf/elements/element_collection.py,sha256=av2YKTxEB5lHYqw1A6aYoN-Uef2qzT9z6ibBAbJMPo4,101322
 natural_pdf/elements/image.py,sha256=zu-P2Y8fRoEXf6IeZU0EYRWsgZ6I_a5vy1FA3VXTGkQ,1424
-natural_pdf/elements/line.py,sha256=mHSeV-ZABY-Cc_K_NpFL53OGtTWlexYDlMvZc8_Vrx8,3845
-natural_pdf/elements/rect.py,sha256=QuQg0Qo7XYQKBac-3Ss0n0ELV6icdPcrygWM2VWzeX8,3325
-natural_pdf/elements/region.py,sha256=s3iFTq6QNiEgSAEV9ywt-3oQW5_swTvB6FNMgANpvmA,151055
-natural_pdf/elements/text.py,sha256=giPJQaXuOBCviQ7QKVx_ZMrKFVpgQAsaCS2-kn-8mp0,20530
+natural_pdf/elements/line.py,sha256=TFn7KXjPT_jUQyQyabU0F7XYU4dC-qadwodJMZF4DCU,3844
+natural_pdf/elements/rect.py,sha256=0lNkVkPkvbRbrFED856RXoUcTcDkeeOIs5xldKGAQT8,3324
+natural_pdf/elements/region.py,sha256=PoT4e2s0gPkMa2Px0LjkThi-Jc8O0_ebl6U7UYADAQk,155289
+natural_pdf/elements/text.py,sha256=IyyU3G4F3OzNZ4Oo0BTK_Wq0p0xFj5EYBWNVL4SZ-BQ,20492
 natural_pdf/export/mixin.py,sha256=L1q3MIEFWuvie4j4_EmW7GT3NerbZ1as0XMUoqTS7gM,5083
 natural_pdf/exporters/__init__.py,sha256=QffoARekR6WzXEd05oxOytly4qPdBizuIF-SUkeFpig,643
-natural_pdf/exporters/base.py,sha256=XhR1xlkHOh7suOuX7mWbsj1h2o1pZNet-OAS5YCJyeI,2115
+natural_pdf/exporters/base.py,sha256=379sioW_hbkGb21sEVuJhbkkDO5MFsFtTUNO5TgG2YU,2101
 natural_pdf/exporters/hocr.py,sha256=wksvJvWLSxuAfhYzg_0T2_W8eqDoMgAVC-gwZ9FoO_k,19969
 natural_pdf/exporters/hocr_font.py,sha256=1wsGOMj6zoaRN2rxCwrv4MMLGawpNz984WgXpmWekgw,4574
-natural_pdf/exporters/original_pdf.py,sha256=KYW0f9_zdouZq_ZwNGvYnu6WHqv7JWrrEAdPCVmhRV4,6782
-natural_pdf/exporters/paddleocr.py,sha256=RBP03GCk0mLeC7tWtuti8AIUHlpOrtvbWkE2n7Ja7k8,19484
-natural_pdf/exporters/searchable_pdf.py,sha256=G2Tc4tpDXSYIufXJlkA8ppW_3DuzHAaweYKae33pI_c,16290
+natural_pdf/exporters/original_pdf.py,sha256=sJOSq3JmQa3mHK0Nn9w-seO5cY524tIkqV9iXHOMcCE,6782
+natural_pdf/exporters/paddleocr.py,sha256=CyXDy6kIoRu5iquOkn1voVrHiYqCfocqPAIGkRemMJg,19395
+natural_pdf/exporters/searchable_pdf.py,sha256=7RDNTV2jK5b5PhZz-v-kpYGTDCXu8FBgX-Mp9PZAoyg,16328
 natural_pdf/exporters/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 natural_pdf/exporters/data/pdf.ttf,sha256=x4RUIJJaI9iO2DCmOVe4r4Wmao2vjZ_JDoQ2c7LvGlk,572
 natural_pdf/exporters/data/sRGB.icc,sha256=KpLUuuRQt22LCqQhk9-XTXX2Jzjs6_dPAcXnWxKpV5Y,6922
 natural_pdf/extraction/manager.py,sha256=sASPJZ5cWFsl8A4PyTjg2yqkyC00tRl6glfoFA6HcsM,4979
-natural_pdf/extraction/mixin.py,sha256=z0HNRs4x4RoioNjzg3slDeqoHbiPug0HB37bUHehqMY,25066
+natural_pdf/extraction/mixin.py,sha256=ck2e48BYZg5RNderNE0QST6RSn2D6mIZYBw91nMSgp8,24970
 natural_pdf/extraction/result.py,sha256=PDaCCN2LQBbHsZy0_lrQ0ROeMsnmH1WRoXWOjk9M2o4,1825
 natural_pdf/flows/__init__.py,sha256=cUN4A8hTDLZSRr4PO2W_lR4z6hWpbNG8Seox-IIcrLU,277
-natural_pdf/flows/collections.py,sha256=iF8SsfKKb-YVIGi3m-yMRnfKgo_0n_EGhojnYK24h-Q,28493
-natural_pdf/flows/element.py,sha256=mKzk3B7A7sWNvu4CDvAjLr3_ZFLt--ktrSNoLfLpFxU,23940
-natural_pdf/flows/flow.py,sha256=ukkUqXsZmEw-QJEiVqEBLC8ktfBG2Bw56_RR1OEsd24,12802
-natural_pdf/flows/region.py,sha256=jRenBFh2ZmFNklNnGkzCsAM0OfMjbP8fo0p7BiVCl_k,31795
+natural_pdf/flows/collections.py,sha256=ErkHWdX6W_y1SjkcA_bGM0uUYRGPWWpRkHip6LHpej0,25740
+natural_pdf/flows/element.py,sha256=T-9uXsIBe7mIim-mQQMep6Ja5dRfWaYIj8g1ak_Bv8c,24892
+natural_pdf/flows/flow.py,sha256=If4G0feMsO3jZO5T2YXvGhb2UygMgwK3567JD42bSl0,85975
+natural_pdf/flows/region.py,sha256=r_cFtBlmPi7ADN3k8oYA1s_vyz8GeQLCnYcv58Zt5eM,52263
 natural_pdf/ocr/__init__.py,sha256=VY8hhvDPf7Gh2lB-d2QRmghLLyTy6ydxlgo1cS4dOSk,2482
 natural_pdf/ocr/engine.py,sha256=SwNlWydtHbrIghV5JD_j5B4-rnjCMYIWUIEARag-zHw,11839
 natural_pdf/ocr/engine_doctr.py,sha256=ptKrupMWoulZb-R93zr9btoe94JPWU7vlJuN7OBJEIM,17740
@@ -69,38 +72,38 @@ natural_pdf/ocr/engine_paddle.py,sha256=OmZlXVh2SSgNePqb6sMo2Mg5boX7REA4MUY25O7h
 natural_pdf/ocr/engine_surya.py,sha256=lOvSbZk53VKFVxRmqcQzM_0dHVdwTkRGiDZ9AWCgL1Q,5951
 natural_pdf/ocr/ocr_factory.py,sha256=Ix-p1SrV6dchq6YcbbCTf2BPBHSGwu9KBnwnZ_ohOuw,5282
 natural_pdf/ocr/ocr_manager.py,sha256=U8EVzNgeRQxxAbMpCEZhkF7nr_R8Fcvtp28oeV_D-Ms,16229
-natural_pdf/ocr/ocr_options.py,sha256=_BgLjIih6mY3k-AgkdXu9UDD8bykmQX2fpf37tAOhYQ,5146
-natural_pdf/ocr/utils.py,sha256=OxuHwDbHWj6setvnC0QYwMHrAjxGkhmLzWHpMqqGupA,4397
+natural_pdf/ocr/ocr_options.py,sha256=dX0b8pb6rllKWpweZuIqWBoN16mmibycrUhUTsOcCw0,5144
+natural_pdf/ocr/utils.py,sha256=Nks2t8WQ2n4zSRc8ugbvnqNQEYfahVwCbFKkjUI2l2Y,4427
 natural_pdf/qa/__init__.py,sha256=2u2KJcA71g1I0HnLD-j6yvDw1moAjo9kkLhhfoYRURM,166
-natural_pdf/qa/document_qa.py,sha256=EduwpmUs8Oz35GrCfLw3312F_ngxIpWZLM8KNvasdrM,19887
+natural_pdf/qa/document_qa.py,sha256=z3ACNdjFA5Il-I2QkxIhZRny030TBKrmcRbUcyVU4IA,20606
 natural_pdf/qa/qa_result.py,sha256=8_jL5MJAHR4LcjGVe5lVsFizxWieF6VI86DWaqetYxs,2167
 natural_pdf/search/__init__.py,sha256=araouqM-l_m0VlluKf6i9BybAsHnfCuh39M0-xEI3jA,4273
 natural_pdf/search/lancedb_search_service.py,sha256=dfz5IiMIcAc3KFzkBDF6Ab_JDLpLHqW6DO1JDkPPu1k,14458
 natural_pdf/search/numpy_search_service.py,sha256=GwPwnX_wxBPFHe-bKS5upMRZLHj8PjLQ2d84lZygzHg,10331
 natural_pdf/search/search_options.py,sha256=sq_e8_jSROicD94b_xtDtLnjEr_Zsy4icjzPkK0a8QA,3566
-natural_pdf/search/search_service_protocol.py,sha256=Dl-Q-CrutkhZwI69scbW9EWPeYM63qxB60_EA7YqIYo,6699
+natural_pdf/search/search_service_protocol.py,sha256=u8pbuWP96fnQEe6mnreY9DrdiDAHP6ZCY7phvSbFlP8,6697
 natural_pdf/search/searchable_mixin.py,sha256=hqQ_AuID5eTGRCtKYdFLZ1zF35y73uk3x1M1VW9Il8U,23514
 natural_pdf/selectors/__init__.py,sha256=oZGeqSv53EqmIZOhcnawuaGGlRg1h79vArXuZCWKm4A,123
-natural_pdf/selectors/parser.py,sha256=W1gZuBhGy2uHqCoExzCAFbsiVMKYSgUfCc9cr4rO1V0,37540
+natural_pdf/selectors/parser.py,sha256=uWo0K4uWJFbD4kTXz9fOcPwEjs7cGR9Mfpm1jm7qKUM,38824
 natural_pdf/tables/__init__.py,sha256=sCvCGbGsL6BiqlNxAYfVv003bIDLI11FmjHhaWfcU6w,104
-natural_pdf/tables/result.py,sha256=hrGIWDkImpdxsGzugcQKU-qrTgHwwfOigJDFdYl8aUc,3994
+natural_pdf/tables/result.py,sha256=lfhLs5OxZ2IRLNndb8zjOQBk1SPjHx4KePzI7GlRkMg,5478
 natural_pdf/templates/__init__.py,sha256=jYBxzfi73vew0f6yhIh1MlRxw4F_TVN2hKQR0YXOFe0,20
 natural_pdf/utils/__init__.py,sha256=s3M8FggaK1P3EBYn6R_-HgSDjNc9C73gyKe1hihtNWg,43
 natural_pdf/utils/bidi_mirror.py,sha256=jJEES0xDrMfo5Me8kHMxHv4COS51PitnYi2EvKv3HCE,1151
-natural_pdf/utils/debug.py,sha256=RN7H3E6ph-GtxubCW6psW7TO8o2BxcNLiEzByTVR9fk,995
-natural_pdf/utils/highlighting.py,sha256=EIY6ihVGtUTS_DjWyxpnr_UXpcR4btC1KhSGQ9VUfKg,698
+natural_pdf/utils/debug.py,sha256=Epwie_jmRgknUSaEoxEyvr1lBXpfYTFOe2UQh_zSj_0,1026
+natural_pdf/utils/highlighting.py,sha256=c9SvvPaJDI9bWXzq1A7zdh_0s3C4GCMngrJdkL2AMeM,719
 natural_pdf/utils/identifiers.py,sha256=P7n6owcubnF8oAMa_UfYtENmIaJQdH_AMC9Jbs2bWXo,1117
-natural_pdf/utils/layout.py,sha256=nCmXiflatJoh-uFtDzfNnlYuRu7Pu9voJWUek6rEMzI,752
+natural_pdf/utils/layout.py,sha256=tJRRzwUVP0EeqqbGzr9yOuE5qFvhjZ9A44BuItmKGaU,753
 natural_pdf/utils/locks.py,sha256=7HJqV0VsNcOfISnbw8goCKWP5ck11uSJo6T_x9XIPKI,215
-natural_pdf/utils/packaging.py,sha256=e7U2wWvpunlAWpPFexNkD_c4dYbPp5LcKo7og4bNGvk,22411
+natural_pdf/utils/packaging.py,sha256=TM0jafwS5yVbTGC-RMi4TyWunf9cUUo9h5J6rMzkT-o,22444
 natural_pdf/utils/reading_order.py,sha256=u7XyVZdKMPMK0CL1C7xFogKnZ92b0JKT068KFjQWe18,7437
-natural_pdf/utils/text_extraction.py,sha256=HYWlYGPfafwzsuMyfL5oQhvcD4NobbvC_aCpcDs9I64,13901
-natural_pdf/utils/visualization.py,sha256=olDkWtuVzP0NxRg0CP0DL-eXNCY7Bs-SH-2Xn-cjbo0,9370
+natural_pdf/utils/text_extraction.py,sha256=CCwPTmMoTgtQt2P00X_ADIf6ZGNfxvjCO9FO0_HqG40,13900
+natural_pdf/utils/visualization.py,sha256=zhZEHgYnZFuX7YxTHXF8Y3D97uHp2beTKMaC-JkCFwk,22364
 natural_pdf/widgets/__init__.py,sha256=QTVaUmsw__FCweFYZebwPssQxxUFUMd0wpm_cUbGZJY,181
 natural_pdf/widgets/viewer.py,sha256=KW3JogdR2TMg2ECUMYp8hwd060hfg8EsYBWxb5IEzBY,24942
-natural_pdf-0.1.38.dist-info/licenses/LICENSE,sha256=9zfwINwJlarbDmdh6iJV4QUG54QSJlSAUcnC1YiC_Ns,1074
-optimization/memory_comparison.py,sha256=F90D_5WhliSGAct_lyx93xd4q4F-jeo8QpGyDr8tmNw,6543
-optimization/pdf_analyzer.py,sha256=xf6h-FNlqCpsm8NriXcs_bQZOB8eQkxgGGKVRL_jgCM,19347
+natural_pdf-0.2.0.dist-info/licenses/LICENSE,sha256=9zfwINwJlarbDmdh6iJV4QUG54QSJlSAUcnC1YiC_Ns,1074
+optimization/memory_comparison.py,sha256=0i_foFSRmppj-fY069qjwH36s_zkx-1L2ASAAlepWzA,6541
+optimization/pdf_analyzer.py,sha256=HjrmTgu2qchxPeDckc5kjgxppGwd40UESrYS9Myj7pY,19352
 optimization/performance_analysis.py,sha256=RjAqeE3YS1r_7qTWkY6Ng5YMbb6MXJXfXX6LoVjg_xQ,13035
 optimization/test_cleanup_methods.py,sha256=PmLOL4MRgvV0j_DW9W1TS8MsGGgu57QCuq6_5y7zK3s,6209
 optimization/test_memory_fix.py,sha256=A3knK74fNhvHknDbLhbTmA276x1ifl-3ivJ_7BhVSTI,6170
@@ -115,8 +118,8 @@ tools/bad_pdf_eval/llm_enrich.py,sha256=mCh4KGi1HmIkzGjj5rrHz1Osd7sEX1IZ_FW08H1t
 tools/bad_pdf_eval/llm_enrich_with_retry.py,sha256=XUtPF1hUvqd3frDXT0wDTXoonuAivhjM5vgFdZ-tm0A,9373
 tools/bad_pdf_eval/reporter.py,sha256=e1g__mkSB4q02p3mGWOwMhvFs7F2HJosNBxup0-LkyU,400
 tools/bad_pdf_eval/utils.py,sha256=hR95XQ7qf7Cu6BdyX0L7ggGVx-ah5sK0jHWblTJUUic,4896
-natural_pdf-0.1.38.dist-info/METADATA,sha256=7a2BfP1oBRbUDUm_9t-3jCsw9BGjIiGyoFwGQyDvcVo,6739
-natural_pdf-0.1.38.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-natural_pdf-0.1.38.dist-info/entry_points.txt,sha256=1R_KMv7g60UBBpRqGfw7bppsMNGdayR-iJlb9ohEk_8,81
-natural_pdf-0.1.38.dist-info/top_level.txt,sha256=oZlRzSc3nZ9sV3L6kD_Di734Pp62ANrm46imFVa51qQ,58
-natural_pdf-0.1.38.dist-info/RECORD,,
+natural_pdf-0.2.0.dist-info/METADATA,sha256=XQDU_jfYdUorv20eFGGyVLl48REcto9zLXWzG-gLtxw,6951
+natural_pdf-0.2.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+natural_pdf-0.2.0.dist-info/entry_points.txt,sha256=1R_KMv7g60UBBpRqGfw7bppsMNGdayR-iJlb9ohEk_8,81
+natural_pdf-0.2.0.dist-info/top_level.txt,sha256=oZlRzSc3nZ9sV3L6kD_Di734Pp62ANrm46imFVa51qQ,58
+natural_pdf-0.2.0.dist-info/RECORD,,

optimization/memory_comparison.py

@@ -4,7 +4,7 @@ Memory comparison script to measure the effectiveness of the character duplicati
 
 This script compares memory usage before and after the optimization by:
 1. Testing with a text-heavy PDF
-2. Measuring detailed memory usage patterns  
+2. Measuring detailed memory usage patterns
 3. Calculating memory savings
 """
 

optimization/pdf_analyzer.py

@@ -14,7 +14,7 @@ import sys
 from pathlib import Path
 
 import natural_pdf as npdf
-from natural_pdf.elements.collections import ElementCollection
+from natural_pdf.elements.element_collection import ElementCollection
 
 
 def analyze_pdf(
@@ -386,7 +386,7 @@ def analyze_pdf(
             # 7. Render page as image
             print(f"\n🖼️  RENDERING PAGE AS IMAGE:")
             try:
-                img = page.to_image(resolution=144)
+                img = page.render(resolution=144)
                 print(f"Image: {img.width}x{img.height} pixels")
 
                 # Save image in output folder