doctra 0.1.1__py3-none-any.whl → 0.2.0__py3-none-any.whl

doctra/cli/main.py

@@ -818,8 +818,8 @@ def info():
 
     # VLM providers
     click.echo("\nVLM Providers:")
-    click.echo("  • Gemini (Google) - gemini-1.5-flash-latest, gemini-1.5-pro")
-    click.echo("  • OpenAI - gpt-4o, gpt-4o-mini, gpt-4-vision-preview")
+    click.echo("  • Gemini (Google) - gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, gemini-2.0-flash")
+    click.echo("  • OpenAI - gpt-5, gpt-5-mini, gpt-4.1, gpt-4.1-mini, gpt-4o")
 
     # Available layout models
     click.echo("\nLayout Detection Models:")

doctra/cli/utils.py

@@ -263,7 +263,7 @@ def create_progress_callback(description: str, total: int):
     """
     Create a progress callback function for use with processing operations.
     
-    Creates a tqdm progress bar and returns a callback function that
+    Creates a beautiful tqdm progress bar and returns a callback function that
     can be used to update the progress during long-running operations.
 
     :param description: Description text for the progress bar
@@ -271,9 +271,18 @@ def create_progress_callback(description: str, total: int):
     :return: Callable progress callback function that takes an integer
              representing the number of completed items
     """
-    from tqdm import tqdm
+    import sys
+    from doctra.utils.progress import create_beautiful_progress_bar, create_notebook_friendly_bar
 
-    pbar = tqdm(total=total, desc=description, leave=True)
+    # Enhanced environment detection
+    is_notebook = "ipykernel" in sys.modules or "jupyter" in sys.modules
+    is_terminal = hasattr(sys.stdout, 'isatty') and sys.stdout.isatty()
+    
+    # Choose appropriate progress bar based on environment
+    if is_notebook:
+        pbar = create_notebook_friendly_bar(total=total, desc=description)
+    else:
+        pbar = create_beautiful_progress_bar(total=total, desc=description, leave=True)
 
     def callback(completed: int):
         pbar.n = completed

doctra/engines/layout/paddle_layout.py

@@ -14,6 +14,7 @@ from paddleocr import LayoutDetection  # pip install paddleocr>=2.7.0.3
 from doctra.utils.pdf_io import render_pdf_to_images
 from doctra.engines.layout.layout_models import LayoutBox, LayoutPage
 from doctra.utils.quiet import suppress_output
+from doctra.utils.progress import create_loading_bar
 
 
 class PaddleLayoutEngine:
@@ -53,8 +54,8 @@ class PaddleLayoutEngine:
         if self.model is not None:
             return
 
-        # Your own friendly progress line
-        with tqdm(total=1, desc=f'Loading PaddleOCR layout model: "{self.model_name}"', leave=True) as bar:
+        # Beautiful loading progress bar
+        with create_loading_bar(f'Loading PaddleOCR layout model: "{self.model_name}"') as bar:
             # Monkey patch tqdm to disable it completely during model loading
             original_tqdm_init = tqdm.__init__
             original_tqdm_update = tqdm.update

doctra/engines/vlm/provider.py

@@ -8,6 +8,7 @@ import outlines
 from pydantic import BaseModel
 from google.genai import Client
 from outlines.inputs import Image
+from anthropic import Anthropic
 # ------------------------------------------------------
 
 def make_model(
@@ -19,12 +20,12 @@ def make_model(
     """
     Build a callable Outlines model for VLM processing.
     
-    Creates an Outlines model instance configured for either Gemini or OpenAI
+    Creates an Outlines model instance configured for Gemini, OpenAI, Anthropic, or OpenRouter
     providers. Only one backend is active at a time, with Gemini as the default.
 
-    :param vlm_provider: VLM provider to use ("gemini" or "openai", default: "gemini")
+    :param vlm_provider: VLM provider to use ("gemini", "openai", or "anthropic", default: "gemini")
     :param vlm_model: Model name to use (defaults to provider-specific defaults)
-    :param api_key: API key for the VLM provider (required for both Gemini and OpenAI)
+    :param api_key: API key for the VLM provider (required for all providers)
     :return: Configured Outlines model instance
     :raises ValueError: If provider is unsupported or API key is missing
     """
@@ -33,9 +34,13 @@ def make_model(
     # Set default models if not provided
     if vlm_model is None:
         if vlm_provider == "gemini":
-            vlm_model = "gemini-1.5-flash-latest"
+            vlm_model = "gemini-2.5-pro"
         elif vlm_provider == "openai":
-            vlm_model = "gpt-4o"
+            vlm_model = "gpt-5"
+        elif vlm_provider == "anthropic":
+            vlm_model = "claude-opus-4-1"
+        elif vlm_provider == "openrouter":
+            vlm_model = "x-ai/grok-4"
 
     if vlm_provider == "gemini":
         if not api_key:
@@ -55,4 +60,27 @@ def make_model(
             vlm_model,
         )
 
-    raise ValueError(f"Unsupported provider: {vlm_provider}. Use 'gemini' or 'openai'.")
+    if vlm_provider == "anthropic":
+        if not api_key:
+            raise ValueError("Anthropic provider requires api_key to be passed to make_model(...).")
+        # Create the Anthropic client and model (exactly like your snippet)
+        client = Anthropic(api_key=api_key)
+        return outlines.from_anthropic(
+            client,
+            vlm_model,
+        )
+
+    if vlm_provider == "openrouter":
+        if not api_key:
+            raise ValueError("OpenRouter provider requires api_key to be passed to make_model(...).")
+        # Create the Anthropic client and model (exactly like your snippet)
+        client = openai.OpenAI(
+            base_url="https://openrouter.ai/api/v1",
+            api_key=api_key,
+        )
+        return outlines.from_openai(
+            client,
+            vlm_model
+        )
+
+    raise ValueError(f"Unsupported provider: {vlm_provider}. Use 'gemini', 'openai', or 'anthropic'.")

doctra/engines/vlm/service.py

@@ -18,6 +18,9 @@ class VLMStructuredExtractor:
         vlm = VLMStructuredExtractor(vlm_provider="gemini", api_key="YOUR_KEY", debug=True)
         chart = vlm.extract_chart("/abs/path/chart.jpg")
         table = vlm.extract_table("/abs/path/table.jpg")
+        
+        # Or with Anthropic:
+        vlm = VLMStructuredExtractor(vlm_provider="anthropic", api_key="YOUR_KEY", debug=True)
     """
 
     def __init__(
@@ -34,9 +37,9 @@ class VLMStructuredExtractor:
         Sets up the VLM model and debug settings for structured data extraction
         from images.
 
-        :param vlm_provider: VLM provider to use ("gemini" or "openai", default: "gemini")
+        :param vlm_provider: VLM provider to use ("gemini", "openai", "anthropic", or "openrouter", default: "gemini")
         :param vlm_model: Model name to use (defaults to provider-specific defaults)
-        :param api_key: API key for the VLM provider (required for both Gemini and OpenAI)
+        :param api_key: API key for the VLM provider (required for all providers)
         :param debug: Whether to enable debug output for error handling (default: True)
         """
         self.model = make_model(

doctra/parsers/structured_pdf_parser.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 import os
 import re
+import sys
 from typing import List, Dict, Any
 from contextlib import ExitStack
 from PIL import Image, ImageDraw, ImageFont
@@ -19,6 +20,7 @@ from doctra.exporters.excel_writer import write_structured_excel
 from doctra.utils.structured_utils import to_structured_dict
 from doctra.exporters.markdown_table import render_markdown_table
 from doctra.exporters.markdown_writer import write_markdown
+from doctra.utils.progress import create_beautiful_progress_bar, create_multi_progress_bars, create_notebook_friendly_bar
 
 
 class StructuredPDFParser:
@@ -30,7 +32,7 @@ class StructuredPDFParser:
     converting visual elements into structured data.
 
     :param use_vlm: Whether to use VLM for structured data extraction (default: False)
-    :param vlm_provider: VLM provider to use ("gemini" or "openai", default: "gemini")
+    :param vlm_provider: VLM provider to use ("gemini", "openai", "anthropic", or "openrouter", default: "gemini")
     :param vlm_model: Model name to use (defaults to provider-specific defaults)
     :param vlm_api_key: API key for VLM provider (required if use_vlm is True)
     :param layout_model_name: Layout detection model name (default: "PP-DocLayout_plus-L")
@@ -66,7 +68,7 @@ class StructuredPDFParser:
         the VLM service for comprehensive document processing.
 
         :param use_vlm: Whether to use VLM for structured data extraction
-        :param vlm_provider: VLM provider to use ("gemini" or "openai")
+        :param vlm_provider: VLM provider to use ("gemini", "openai", "anthropic", or "openrouter")
         :param vlm_model: Model name to use (defaults to provider-specific defaults)
         :param vlm_api_key: API key for VLM provider
         :param layout_model_name: Layout detection model name
@@ -130,12 +132,25 @@ class StructuredPDFParser:
         figures_desc = "Figures (cropped)"
 
         with ExitStack() as stack:
-            charts_bar = stack.enter_context(
-                tqdm(total=chart_count, desc=charts_desc, leave=True)) if chart_count else None
-            tables_bar = stack.enter_context(
-                tqdm(total=table_count, desc=tables_desc, leave=True)) if table_count else None
-            figures_bar = stack.enter_context(
-                tqdm(total=fig_count, desc=figures_desc, leave=True)) if fig_count else None
+            # Enhanced environment detection
+            is_notebook = "ipykernel" in sys.modules or "jupyter" in sys.modules
+            is_terminal = hasattr(sys.stdout, 'isatty') and sys.stdout.isatty()
+            
+            # Use appropriate progress bars based on environment
+            if is_notebook:
+                charts_bar = stack.enter_context(
+                    create_notebook_friendly_bar(total=chart_count, desc=charts_desc)) if chart_count else None
+                tables_bar = stack.enter_context(
+                    create_notebook_friendly_bar(total=table_count, desc=tables_desc)) if table_count else None
+                figures_bar = stack.enter_context(
+                    create_notebook_friendly_bar(total=fig_count, desc=figures_desc)) if fig_count else None
+            else:
+                charts_bar = stack.enter_context(
+                    create_beautiful_progress_bar(total=chart_count, desc=charts_desc, leave=True)) if chart_count else None
+                tables_bar = stack.enter_context(
+                    create_beautiful_progress_bar(total=table_count, desc=tables_desc, leave=True)) if table_count else None
+                figures_bar = stack.enter_context(
+                    create_beautiful_progress_bar(total=fig_count, desc=figures_desc, leave=True)) if fig_count else None
 
             for p in pages:
                 page_num = p.page_index

doctra/parsers/table_chart_extractor.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import os
+import sys
 from typing import List, Dict, Any
 from contextlib import ExitStack
 from pathlib import Path
@@ -9,6 +10,7 @@ from PIL import Image
 from tqdm import tqdm
 
 from doctra.utils.pdf_io import render_pdf_to_images
+from doctra.utils.progress import create_beautiful_progress_bar, create_multi_progress_bars, create_notebook_friendly_bar
 from doctra.engines.layout.paddle_layout import PaddleLayoutEngine
 from doctra.engines.layout.layout_models import LayoutPage
 
@@ -34,7 +36,7 @@ class ChartTablePDFParser:
     :param extract_charts: Whether to extract charts from the document (default: True)
     :param extract_tables: Whether to extract tables from the document (default: True)
     :param use_vlm: Whether to use VLM for structured data extraction (default: False)
-    :param vlm_provider: VLM provider to use ("gemini" or "openai", default: "gemini")
+    :param vlm_provider: VLM provider to use ("gemini", "openai", "anthropic", or "openrouter", default: "gemini")
     :param vlm_model: Model name to use (defaults to provider-specific defaults)
     :param vlm_api_key: API key for VLM provider (required if use_vlm is True)
     :param layout_model_name: Layout detection model name (default: "PP-DocLayout_plus-L")
@@ -64,7 +66,7 @@ class ChartTablePDFParser:
         :param extract_charts: Whether to extract charts from the document
         :param extract_tables: Whether to extract tables from the document
         :param use_vlm: Whether to use VLM for structured data extraction
-        :param vlm_provider: VLM provider to use ("gemini" or "openai")
+        :param vlm_provider: VLM provider to use ("gemini", "openai", "anthropic", or "openrouter")
         :param vlm_model: Model name to use (defaults to provider-specific defaults)
         :param vlm_api_key: API key for VLM provider
         :param layout_model_name: Layout detection model name
@@ -149,10 +151,21 @@ class ChartTablePDFParser:
         table_counter = 1
 
         with ExitStack() as stack:
-            charts_bar = stack.enter_context(
-                tqdm(total=chart_count, desc=charts_desc, leave=True)) if chart_count else None
-            tables_bar = stack.enter_context(
-                tqdm(total=table_count, desc=tables_desc, leave=True)) if table_count else None
+            # Enhanced environment detection
+            is_notebook = "ipykernel" in sys.modules or "jupyter" in sys.modules
+            is_terminal = hasattr(sys.stdout, 'isatty') and sys.stdout.isatty()
+            
+            # Use appropriate progress bars based on environment
+            if is_notebook:
+                charts_bar = stack.enter_context(
+                    create_notebook_friendly_bar(total=chart_count, desc=charts_desc)) if chart_count else None
+                tables_bar = stack.enter_context(
+                    create_notebook_friendly_bar(total=table_count, desc=tables_desc)) if table_count else None
+            else:
+                charts_bar = stack.enter_context(
+                    create_beautiful_progress_bar(total=chart_count, desc=charts_desc, leave=True)) if chart_count else None
+                tables_bar = stack.enter_context(
+                    create_beautiful_progress_bar(total=table_count, desc=tables_desc, leave=True)) if table_count else None
 
             for p in pages:
                 page_num = p.page_index

doctra/utils/progress.py

@@ -0,0 +1,277 @@
+from __future__ import annotations
+
+import os
+import sys
+from typing import Optional, Dict, Any
+from tqdm import tqdm
+from tqdm.auto import tqdm as tqdm_auto
+
+
+def create_beautiful_progress_bar(
+    total: int,
+    desc: str,
+    leave: bool = True,
+    position: Optional[int] = None,
+    **kwargs
+) -> tqdm:
+    """
+    Create a beautiful and interactive tqdm progress bar with enhanced styling.
+    
+    Features:
+    - Colorful progress bars with gradients
+    - Emoji icons for different operations
+    - Better formatting and spacing
+    - Interactive features
+    - Responsive design
+    
+    :param total: Total number of items to process
+    :param desc: Description text for the progress bar
+    :param leave: Whether to leave the progress bar after completion
+    :param position: Position of the progress bar (for multiple bars)
+    :param kwargs: Additional tqdm parameters
+    :return: Configured tqdm progress bar instance
+    """
+    
+    # Enhanced styling parameters - notebook-friendly format
+    if "ipykernel" in sys.modules:
+        # Simpler format for notebooks to avoid display issues
+        bar_format = "{desc}: {percentage:3.0f}%|{bar}| {n_fmt}/{total_fmt} [{elapsed}<{remaining}]"
+    else:
+        # Full format for terminal
+        bar_format = (
+            "{l_bar}{bar:30}| {n_fmt}/{total_fmt} "
+            "[{elapsed}<{remaining}, {rate_fmt}{postfix}]"
+        )
+    
+    # Color schemes based on operation type
+    color_schemes = {
+        "loading": {"colour": "cyan", "ncols": 100},
+        "charts": {"colour": "green", "ncols": 100},
+        "tables": {"colour": "blue", "ncols": 100},
+        "figures": {"colour": "magenta", "ncols": 100},
+        "ocr": {"colour": "yellow", "ncols": 100},
+        "vlm": {"colour": "red", "ncols": 100},
+        "processing": {"colour": "white", "ncols": 100},
+    }
+    
+    # Determine color scheme based on description
+    desc_lower = desc.lower()
+    if "loading" in desc_lower or "model" in desc_lower:
+        color_scheme = color_schemes["loading"]
+    elif "chart" in desc_lower:
+        color_scheme = color_schemes["charts"]
+    elif "table" in desc_lower:
+        color_scheme = color_schemes["tables"]
+    elif "figure" in desc_lower:
+        color_scheme = color_schemes["figures"]
+    elif "ocr" in desc_lower:
+        color_scheme = color_schemes["ocr"]
+    elif "vlm" in desc_lower:
+        color_scheme = color_schemes["vlm"]
+    else:
+        color_scheme = color_schemes["processing"]
+    
+    # Add emoji icons to descriptions
+    emoji_map = {
+        "loading": "🔄",
+        "charts": "📊",
+        "tables": "📋",
+        "figures": "🖼️",
+        "ocr": "🔍",
+        "vlm": "🤖",
+        "processing": "⚙️",
+    }
+    
+    # Add appropriate emoji to description
+    for key, emoji in emoji_map.items():
+        if key in desc_lower:
+            desc = f"{emoji} {desc}"
+            break
+    else:
+        desc = f"⚙️ {desc}"
+    
+    # Enhanced tqdm configuration
+    tqdm_config = {
+        "total": total,
+        "desc": desc,
+        "leave": leave,
+        "bar_format": bar_format,
+        "ncols": color_scheme["ncols"],
+        "ascii": False,  # Use Unicode characters for better appearance
+        "dynamic_ncols": True,  # Responsive width
+        "smoothing": 0.3,  # Smooth progress updates
+        "mininterval": 0.1,  # Minimum update interval
+        "maxinterval": 1.0,  # Maximum update interval
+        "position": position,
+        **kwargs
+    }
+    
+    # Enhanced environment detection
+    is_notebook = "ipykernel" in sys.modules or "jupyter" in sys.modules
+    is_terminal = hasattr(sys.stdout, 'isatty') and sys.stdout.isatty()
+    
+    # Add color only for terminal environments (not notebooks)
+    if not is_notebook and is_terminal:
+        tqdm_config["colour"] = color_scheme["colour"]
+    
+    # Use auto tqdm for better Jupyter notebook support
+    if is_notebook:
+        # In notebooks, don't use color to avoid ANSI code issues
+        tqdm_config.pop("colour", None)  # Remove color in notebooks
+        return tqdm_auto(**tqdm_config)
+    else:
+        # In terminal/cmd/powershell, we can use colors
+        return tqdm(**tqdm_config)
+
+
+def create_multi_progress_bars(
+    descriptions: list[str],
+    totals: list[int],
+    positions: Optional[list[int]] = None
+) -> list[tqdm]:
+    """
+    Create multiple beautiful progress bars for concurrent operations.
+    
+    :param descriptions: List of descriptions for each progress bar
+    :param totals: List of totals for each progress bar
+    :param positions: Optional list of positions for each bar
+    :return: List of configured tqdm progress bar instances
+    """
+    if positions is None:
+        positions = list(range(len(descriptions)))
+    
+    bars = []
+    for desc, total, pos in zip(descriptions, totals, positions):
+        bar = create_beautiful_progress_bar(
+            total=total,
+            desc=desc,
+            position=pos,
+            leave=True
+        )
+        bars.append(bar)
+    
+    return bars
+
+
+def update_progress_with_info(
+    bar: tqdm,
+    increment: int = 1,
+    info: Optional[Dict[str, Any]] = None
+) -> None:
+    """
+    Update progress bar with additional information.
+    
+    :param bar: tqdm progress bar instance
+    :param increment: Number to increment the progress
+    :param info: Optional dictionary of information to display
+    """
+    if info:
+        # Format info as postfix
+        postfix_parts = []
+        for key, value in info.items():
+            if isinstance(value, float):
+                postfix_parts.append(f"{key}: {value:.2f}")
+            else:
+                postfix_parts.append(f"{key}: {value}")
+        
+        bar.set_postfix_str(", ".join(postfix_parts))
+    
+    bar.update(increment)
+
+
+def create_loading_bar(desc: str = "Loading", **kwargs) -> tqdm:
+    """
+    Create a special loading progress bar for model initialization.
+    
+    :param desc: Description for the loading operation
+    :param kwargs: Additional tqdm parameters
+    :return: Configured loading progress bar
+    """
+    return create_beautiful_progress_bar(
+        total=1,
+        desc=desc,
+        leave=True,
+        **kwargs
+    )
+
+
+def create_processing_bar(
+    total: int,
+    operation: str,
+    **kwargs
+) -> tqdm:
+    """
+    Create a processing progress bar for data operations.
+    
+    :param total: Total number of items to process
+    :param operation: Type of operation (charts, tables, figures, etc.)
+    :param kwargs: Additional tqdm parameters
+    :return: Configured processing progress bar
+    """
+    desc = f"{operation.title()} (processing)"
+    return create_beautiful_progress_bar(
+        total=total,
+        desc=desc,
+        leave=True,
+        **kwargs
+    )
+
+
+def create_notebook_friendly_bar(
+    total: int,
+    desc: str,
+    **kwargs
+) -> tqdm:
+    """
+    Create a notebook-friendly progress bar with minimal formatting.
+    
+    This function creates progress bars specifically optimized for Jupyter notebooks
+    to avoid display issues and ANSI code problems.
+    
+    :param total: Total number of items to process
+    :param desc: Description text for the progress bar
+    :param kwargs: Additional tqdm parameters
+    :return: Configured notebook-friendly progress bar
+    """
+    # Force notebook mode
+    kwargs["disable"] = False
+    kwargs["ascii"] = True  # Use ASCII characters for better notebook compatibility
+    
+    # Add emoji icons to descriptions (same as beautiful bars)
+    emoji_map = {
+        "loading": "🔄",
+        "charts": "📊",
+        "tables": "📋",
+        "figures": "🖼️",
+        "ocr": "🔍",
+        "vlm": "🤖",
+        "processing": "⚙️",
+    }
+    
+    # Add appropriate emoji to description
+    desc_lower = desc.lower()
+    for key, emoji in emoji_map.items():
+        if key in desc_lower:
+            desc = f"{emoji} {desc}"
+            break
+    else:
+        desc = f"⚙️ {desc}"
+    
+    # Simple format for notebooks
+    bar_format = "{desc}: {percentage:3.0f}%|{bar}| {n_fmt}/{total_fmt}"
+    
+    tqdm_config = {
+        "total": total,
+        "desc": desc,
+        "leave": True,
+        "bar_format": bar_format,
+        "ncols": 80,
+        "ascii": True,
+        "dynamic_ncols": False,  # Fixed width for notebooks
+        "smoothing": 0.1,  # Faster updates
+        "mininterval": 0.05,
+        "maxinterval": 0.5,
+        **kwargs
+    }
+    
+    return tqdm_auto(**tqdm_config)

doctra/version.py

@@ -1,2 +1,2 @@
 """Version information for Doctra."""
-__version__ = '0.1.1'
+__version__ = '0.2.0'

{doctra-0.1.1.dist-info → doctra-0.2.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: doctra
-Version: 0.1.1
+Version: 0.2.0
 Summary: Parse, extract, and analyze documents with ease
 Home-page: https://github.com/AdemBoukhris457/Doctra
 Author: Adem Boukhris
@@ -241,6 +241,8 @@ Provides-Extra: openai
 Requires-Dist: openai>=1.0.0; extra == "openai"
 Provides-Extra: gemini
 Requires-Dist: google-generativeai>=0.3.0; extra == "gemini"
+Provides-Extra: anthropic
+Requires-Dist: anthropic>=0.40.0; extra == "anthropic"
 Provides-Extra: dev
 Requires-Dist: pytest>=6.0; extra == "dev"
 Requires-Dist: pytest-cov>=2.0; extra == "dev"
@@ -329,7 +331,7 @@ parser = StructuredPDFParser()
 # Parser with VLM for structured data extraction
 parser = StructuredPDFParser(
     use_vlm=True,
-    vlm_provider="openai",  # or "gemini"
+    vlm_provider="openai",  # or "gemini" or "anthropic" or "openrouter"
     vlm_api_key="your_api_key_here"
 )
 
@@ -344,7 +346,7 @@ parser = StructuredPDFParser(
     # VLM Settings
     use_vlm=True,
     vlm_provider="openai",
-    vlm_model="gpt-4o",
+    vlm_model="gpt-5",
     vlm_api_key="your_api_key",
     
     # Layout Detection Settings
@@ -406,7 +408,7 @@ parser = ChartTablePDFParser(
     # VLM Settings
     use_vlm=True,
     vlm_provider="openai",
-    vlm_model="gpt-4o",
+    vlm_model="gpt-5",
     vlm_api_key="your_api_key",
     
     # Layout Detection Settings
@@ -545,7 +547,7 @@ parser = StructuredPDFParser(
     use_vlm=True,
     vlm_provider="openai",
     vlm_api_key="your_openai_api_key",
-    vlm__model="gpt-4o",
+    vlm__model="gpt-5",
     layout_model_name="PP-DocLayout_plus-L",
     dpi=300,  # Higher DPI for better quality
     min_score=0.5,  # Higher confidence threshold
@@ -623,4 +625,4 @@ parser.display_pages_with_boxes("document.pdf")
 - **Pandas**: Data manipulation
 - **OpenPyXL**: Excel file generation
 - **Google Generative AI**: For Gemini VLM integration
-- **OpenAI**: For GPT-4 VLM integration
+- **OpenAI**: For GPT-5 VLM integration

{doctra-0.1.1.dist-info → doctra-0.2.0.dist-info}/RECORD

@@ -1,20 +1,20 @@
 doctra/__init__.py,sha256=-Pkx0Vh4Hz3EQvLaxlL6Mo4lVig59FTN5LvUcxThn4U,519
-doctra/version.py,sha256=tiyU44F2UjL7SjuX_aL4-UZ5m_D9WOmXSUZQnjpR8PM,60
+doctra/version.py,sha256=oXtS5MRUB2QfE2Q8GOIq0p_iwA9QH5_2LxFqVKJlb_I,60
 doctra/cli/__init__.py,sha256=4PTujjYRShOOUlZ7PwuWckShPWLC4v4CYIhJpzgyv1k,911
-doctra/cli/main.py,sha256=O3Bgov3rtf58AJHmuojJaptrH17X1mw19iTplId3gGo,35327
-doctra/cli/utils.py,sha256=QuttjEtBiFrOHmqZz4mjbf3GWZe26lYChPwa23Loz_4,11314
+doctra/cli/main.py,sha256=aRxV0yMtswwXKcBrIE7rxMvZCsFSjCVrE5rIqKzYGOY,35368
+doctra/cli/utils.py,sha256=IghiUZQCOmXODC5-5smHGz2KeV4xqbP4avmA1Mggln0,11800
 doctra/engines/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 doctra/engines/layout/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 doctra/engines/layout/layout_models.py,sha256=vuTzjWd3FD-SkFPngktmUVhOJ6Xvff6ufwFEq796PQs,3162
-doctra/engines/layout/paddle_layout.py,sha256=Yf6_OtBq_RSup8CSDofJUZxM_bJMBlCC0eSv5ib1uNk,9364
+doctra/engines/layout/paddle_layout.py,sha256=N9Bzt6372BfWUtQspYqh6PpYWOndjoIYET0_OJU85cs,9405
 doctra/engines/ocr/__init__.py,sha256=h6bFiveGXdI59fsKzCqOXki3C74DCndEmvloOtMqnR0,133
 doctra/engines/ocr/api.py,sha256=YOBKDLExXpvSiOsc_TDJasaMPxzdVx1llQCtYlsruWo,1280
 doctra/engines/ocr/path_resolver.py,sha256=2_7Nsekt3dCDU3oVsgdr62iMrlAhbGNfYwgh4G7S3pA,1492
 doctra/engines/ocr/pytesseract_engine.py,sha256=Imz2uwju6himkBiS8CH7DLxBRe-LtmMYZiOdb_6PoQw,2911
 doctra/engines/vlm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 doctra/engines/vlm/outlines_types.py,sha256=qL-G6MNiA5mxp1qAPVEFhOANp4NqVt_MQKseJCr_xXE,970
-doctra/engines/vlm/provider.py,sha256=ws-04Jhuvg0a3vXzz8cfMWIiwldoIFs3i_qSb2Q6enA,2137
-doctra/engines/vlm/service.py,sha256=cONhekqKfGo2fe-2g7YT89BHxytdjGhCSFyU3sJUzWI,4966
+doctra/engines/vlm/provider.py,sha256=njkz99NXZQjkPlRKeje9M_tlaktXyw3VnpFT7enNalk,3216
+doctra/engines/vlm/service.py,sha256=uD4BXz3u7B_3iq-xU3MTdDDyjrj1Jm8MDeJU1KXHTZc,5121
 doctra/exporters/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 doctra/exporters/excel_writer.py,sha256=U5Eb5SF7_ll1QveUapSWSkCRt3OEoisKEVUQ_7X8Wjo,7762
 doctra/exporters/image_saver.py,sha256=zsPoQ0CwoE643ui4iZMdXk96kv5mU8L_zC2JfF22N1A,1639
@@ -22,8 +22,8 @@ doctra/exporters/markdown_table.py,sha256=4_OJIwG_WoIPYBzJx1njy_3tNVdkK6QKSP-P9r
 doctra/exporters/markdown_writer.py,sha256=L7EjF2MB8jYX7XkZ3a3NeeEC8gnb0qzRPTzIN9tdfuw,1027
 doctra/parsers/__init__.py,sha256=8M6LVzcWGpuTIK_1SMXML3ll7zK1CTHXGI5qXvqdm-A,206
 doctra/parsers/layout_order.py,sha256=W6b-T11H907RZ2FaZwNvnYhmvH11rpUzxC5yLkdf28k,640
-doctra/parsers/structured_pdf_parser.py,sha256=g0k9XsSJRVnJg4omrEC1Ef1MWZZ3Ve2OnXjMoc6IScU,19953
-doctra/parsers/table_chart_extractor.py,sha256=A-rjazOmx6d_8CbZXdebE4NsYYqiQP0wQktTfCp_pwI,12669
+doctra/parsers/structured_pdf_parser.py,sha256=4T4zYZWbqqtRua_TPSRmjT1tOc1RE-XSMOLC5fVFJk0,21070
+doctra/parsers/table_chart_extractor.py,sha256=-pyJFYzFVdxDwCD1z5BHhT6qyf9BljbfccKGoMJiD90,13591
 doctra/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 doctra/utils/bbox.py,sha256=R2-95p0KiWvet3TH27TQVvCar7WJg6z0u3L21iEDF-A,674
 doctra/utils/constants.py,sha256=ZWOvNDrvETbQ_pxHiX7vUW4J5Oj8_qnov0QacUOBizI,189
@@ -31,10 +31,11 @@ doctra/utils/file_ops.py,sha256=3IS0EQncs6Kaj27fcg2zxQX3xRSvtItIsyKGLYgeOgw,815
 doctra/utils/io_utils.py,sha256=L1bWV4-ybs2j_3ZEN7GfQVgdC73JKVECVnpwKbP0dy0,219
 doctra/utils/ocr_utils.py,sha256=Doa1uYBg3kRgRYd2aPq9fICHgHfrM_efdhZfI7jl6OM,780
 doctra/utils/pdf_io.py,sha256=c8EY47Z1iqVtlLFHS_n0qGuXJ5ERFaMUd84ivXV0b9E,706
+doctra/utils/progress.py,sha256=GSjHkNulwqX-Uh_QNP-g-nZH6F-zAwQC120KeTRkRlo,8752
 doctra/utils/quiet.py,sha256=5XPS-1CtJ0sVk6qgSQctdhr_wR8mP1xoJLoUbmkXROA,387
 doctra/utils/structured_utils.py,sha256=EdNhCUDLKvYcLqXbTGveNtIRGyQ3yzYhTh-zy_awwM4,1450
-doctra-0.1.1.dist-info/licenses/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
-doctra-0.1.1.dist-info/METADATA,sha256=NI9kwistJwaI6sYQ0vp5df-4302_EpSG7QUnk2OpocM,26751
-doctra-0.1.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-doctra-0.1.1.dist-info/top_level.txt,sha256=jI7E8jHci2gP9y0GYaWxlg9jG0O5n3FjHJJPLXDXMds,7
-doctra-0.1.1.dist-info/RECORD,,
+doctra-0.2.0.dist-info/licenses/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
+doctra-0.2.0.dist-info/METADATA,sha256=zxVmrkHUI4puc1D8fdUFbRb8WQAL0M4X92v-UaZswPI,26862
+doctra-0.2.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+doctra-0.2.0.dist-info/top_level.txt,sha256=jI7E8jHci2gP9y0GYaWxlg9jG0O5n3FjHJJPLXDXMds,7
+doctra-0.2.0.dist-info/RECORD,,