That's a start

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,13 @@
+# Python-generated files
+__pycache__/
+*.py[oc]
+build/
+dist/
+wheels/
+*.egg-info
+
+# Virtual environments
+.venv
+
+
+.env

.python-version

@@ -0,0 +1 @@
+3.12

app.py

@@ -0,0 +1,432 @@
+import gradio as gr
+import os
+import json
+from typing import List, Tuple, Optional, Dict
+from dataclasses import dataclass, asdict
+from dotenv import load_dotenv
+
+# Load environment variables
+load_dotenv()
+
+# Maximum file size for attachments (10 MB in bytes)
+MAX_ATTACHMENT_SIZE = 10 * 1024 * 1024
+
+
+@dataclass
+class Attachment:
+    """Represents a file attachment with its metadata."""
+    filename: str
+    filepath: str
+    description: str
+    size: int
+
+
+def get_attachments_from_state(attachments_state: List[Dict]) -> List[Attachment]:
+    """
+    Retrieve the current list of attachments for the session.
+
+    Args:
+        attachments_state: The session state containing attachment dictionaries
+
+    Returns:
+        List of Attachment objects
+
+    Note:
+        Session isolation is handled by Gradio's State management.
+        Each user session has its own independent state.
+    """
+    if not attachments_state:
+        return []
+
+    attachments = []
+    for att_dict in attachments_state:
+        if att_dict.get('filepath') and os.path.exists(att_dict.get('filepath')):
+            attachments.append(Attachment(
+                filename=att_dict['filename'],
+                filepath=att_dict['filepath'],
+                description=att_dict['description'],
+                size=att_dict['size']
+            ))
+
+    return attachments
+
+
+def add_attachment_to_pdf(pdf_path: str, attachment: Attachment) -> bool:
+    """
+    Add a single attachment to the PDF file.
+
+    Args:
+        pdf_path: Path to the PDF file
+        attachment: The Attachment object to add
+
+    Returns:
+        True if successful, False otherwise
+
+    TODO: Implement the actual PDF attachment logic here
+    """
+    # ============================================================================
+    # PLACEHOLDER: Insert your logic here to attach the file to the PDF
+    # ============================================================================
+    # Example implementation would:
+    # 1. Load the PDF using a library like pypdf, PyMuPDF, or pikepdf
+    # 2. Embed the file from attachment.filepath as an attachment
+    # 3. Set the attachment description as metadata
+    # 4. Save the PDF back to pdf_path
+
+    print(f"[PLACEHOLDER] Would add attachment to PDF: {attachment.filename}")
+    return True
+
+
+def remove_attachment_from_pdf(pdf_path: str, attachment: Attachment) -> bool:
+    """
+    Remove an attachment from the PDF file.
+
+    Args:
+        pdf_path: Path to the PDF file
+        attachment: The Attachment object to remove
+
+    Returns:
+        True if successful, False otherwise
+
+    TODO: Implement the actual PDF attachment removal logic here
+    """
+    # ============================================================================
+    # PLACEHOLDER: Insert your logic here to remove the attachment from the PDF
+    # ============================================================================
+    # Example implementation would:
+    # 1. Load the PDF using a library like pypdf, PyMuPDF, or pikepdf
+    # 2. Find and remove the embedded file matching attachment.filename
+    # 3. Save the PDF back to pdf_path
+
+    print(f"[PLACEHOLDER] Would remove attachment from PDF: {attachment.filename}")
+    return True
+
+
+def list_pdf_attachments(pdf_path: str) -> List[Attachment]:
+    """
+    List all attachments currently embedded in the PDF file.
+
+    Args:
+        pdf_path: Path to the PDF file
+
+    Returns:
+        List of Attachment objects found in the PDF
+
+    TODO: Implement the actual PDF attachment listing logic here
+    """
+    # ============================================================================
+    # PLACEHOLDER: Insert your logic here to list attachments from the PDF
+    # ============================================================================
+    # Example implementation would:
+    # 1. Load the PDF using a library like pypdf, PyMuPDF, or pikepdf
+    # 2. Iterate through embedded files
+    # 3. Extract filename, description, and size for each
+    # 4. Return as list of Attachment objects
+
+    print(f"[PLACEHOLDER] Would list attachments from PDF: {pdf_path}")
+    return []
+
+
+def attach_files_to_pdf(pdf_path: str, attachments: List[Attachment]) -> str:
+    """
+    Create a new PDF with all attachments embedded.
+
+    Args:
+        pdf_path: Path to the input PDF file
+        attachments: List of Attachment objects to add to the PDF
+
+    Returns:
+        Path to the output PDF file with attachments
+
+    TODO: Implement the actual PDF attachment logic
+    """
+    # This is a placeholder - actual implementation would use a library like pypdf or similar
+    # to embed files as attachments in the PDF
+
+    output_path = pdf_path.replace(".pdf", "_with_attachments.pdf")
+
+    # For now, just copy the original file
+    # In a real implementation, this would:
+    # 1. Load the PDF
+    # 2. Add each attachment with its description as metadata
+    # 3. Save the modified PDF
+
+    print(f"Would attach {len(attachments)} files to {pdf_path}")
+    for att in attachments:
+        print(f"  - {att.filename} ({att.size} bytes): {att.description}")
+
+    # Placeholder: copy the original PDF
+    import shutil
+    shutil.copy2(pdf_path, output_path)
+
+    return output_path
+
+
+def validate_file_size(file_path: str) -> Tuple[bool, str]:
+    """Validate that a file is within the size limit."""
+    if not file_path or not os.path.exists(file_path):
+        return False, "File does not exist"
+
+    file_size = os.path.getsize(file_path)
+    if file_size > MAX_ATTACHMENT_SIZE:
+        size_mb = file_size / (1024 * 1024)
+        return False, f"File size ({size_mb:.2f} MB) exceeds 10 MB limit"
+
+    return True, ""
+
+
+def process_pdf_with_attachments(
+    pdf_file,
+    attachments_state: List[Dict]
+) -> Tuple[Optional[str], str]:
+    """
+    Process the PDF file and add attachments to it.
+
+    Args:
+        pdf_file: The uploaded PDF file
+        attachments_state: List of attachment dictionaries (session-specific)
+
+    Returns:
+        Tuple of (output_file_path, status_message)
+
+    Note:
+        Session isolation is handled by Gradio's State management.
+        Each user session has its own independent attachments_state.
+    """
+    if not pdf_file:
+        return None, "Please upload a PDF file"
+
+    if not attachments_state or len(attachments_state) == 0:
+        return None, "Please add at least one attachment"
+
+    # Get attachments for this session
+    attachments = get_attachments_from_state(attachments_state)
+
+    if not attachments:
+        return None, "No valid attachments found"
+
+    try:
+        output_path = attach_files_to_pdf(pdf_file.name, attachments)
+        return output_path, f"Successfully processed PDF with {len(attachments)} attachment(s)"
+    except Exception as e:
+        return None, f"Error processing PDF: {str(e)}"
+
+
+def add_attachment(
+    attachment_file,
+    description: str,
+    current_attachments: List[Dict]
+) -> Tuple[List[Dict], str, str]:
+    """
+    Add a new attachment to the session's list.
+
+    Args:
+        attachment_file: The file to attach
+        description: Description of the attachment
+        current_attachments: Current session-specific list of attachments
+
+    Returns:
+        Tuple of (updated_attachments_list, attachment_list_html, status_message)
+
+    Note:
+        Session isolation is handled by Gradio's State management.
+        Each user session has its own independent current_attachments list.
+        This function only updates the in-memory state for this session.
+    """
+    if not attachment_file:
+        return current_attachments, render_attachments_list(current_attachments), "Please select a file to attach"
+
+    # Validate file size
+    is_valid, error_msg = validate_file_size(attachment_file.name)
+    if not is_valid:
+        return current_attachments, render_attachments_list(current_attachments), f"Error: {error_msg}"
+
+    if not description or description.strip() == "":
+        return current_attachments, render_attachments_list(current_attachments), "Please provide a description for the attachment"
+
+    # Create new attachment
+    filename = os.path.basename(attachment_file.name)
+    file_size = os.path.getsize(attachment_file.name)
+
+    new_attachment = {
+        'filename': filename,
+        'filepath': attachment_file.name,
+        'description': description.strip(),
+        'size': file_size
+    }
+
+    # Add to session-specific list
+    if current_attachments is None:
+        current_attachments = []
+
+    current_attachments.append(new_attachment)
+
+    # ============================================================================
+    # PLACEHOLDER: If you want to immediately attach to PDF on add, insert here
+    # ============================================================================
+    # Example: Uncomment and modify if you want real-time PDF updates
+    # if pdf_file_path:  # You'd need to pass the PDF path as a parameter
+    #     attachment_obj = Attachment(
+    #         filename=filename,
+    #         filepath=attachment_file.name,
+    #         description=description.strip(),
+    #         size=file_size
+    #     )
+    #     add_attachment_to_pdf(pdf_file_path, attachment_obj)
+
+    return (
+        current_attachments,
+        render_attachments_list(current_attachments),
+        f"Added attachment: {filename}"
+    )
+
+
+def remove_attachment(
+    index: int,
+    current_attachments: List[Dict]
+) -> Tuple[List[Dict], str]:
+    """
+    Remove an attachment from the session's list by index.
+
+    Args:
+        index: Index of the attachment to remove
+        current_attachments: Current session-specific list of attachments
+
+    Returns:
+        Tuple of (updated_attachments_list, attachment_list_html)
+
+    Note:
+        Session isolation is handled by Gradio's State management.
+        Each user session has its own independent current_attachments list.
+        This function only updates the in-memory state for this session.
+    """
+    if current_attachments and 0 <= index < len(current_attachments):
+        removed = current_attachments.pop(index)
+
+        # ============================================================================
+        # PLACEHOLDER: If you want to immediately remove from PDF, insert here
+        # ============================================================================
+        # Example: Uncomment and modify if you want real-time PDF updates
+        # if pdf_file_path:  # You'd need to pass the PDF path as a parameter
+        #     attachment_obj = Attachment(
+        #         filename=removed['filename'],
+        #         filepath=removed['filepath'],
+        #         description=removed['description'],
+        #         size=removed['size']
+        #     )
+        #     remove_attachment_from_pdf(pdf_file_path, attachment_obj)
+
+        return current_attachments, render_attachments_list(current_attachments)
+
+    return current_attachments, render_attachments_list(current_attachments)
+
+
+def render_attachments_list(attachments: Optional[List[Dict]]) -> str:
+    """Render the list of attachments as HTML."""
+    if not attachments or len(attachments) == 0:
+        return "<p style='color: #666; font-style: italic;'>No attachments added yet</p>"
+
+    html = "<div style='font-family: sans-serif;'>"
+    for idx, att in enumerate(attachments):
+        size_mb = att['size'] / (1024 * 1024)
+        html += f"""
+        <div style='border: 1px solid #ddd; padding: 12px; margin: 8px 0; border-radius: 6px; background: #f9f9f9;'>
+            <div style='display: flex; justify-content: space-between; align-items: start;'>
+                <div style='flex: 1;'>
+                    <strong style='color: #333;'>📎 {att['filename']}</strong>
+                    <span style='color: #666; font-size: 0.9em;'> ({size_mb:.2f} MB)</span>
+                    <p style='margin: 8px 0 0 0; color: #555;'>{att['description']}</p>
+                </div>
+            </div>
+        </div>
+        """
+    html += "</div>"
+    return html
+
+
+def main():
+    """Create and launch the Gradio interface."""
+
+    with gr.Blocks(title="Incorporate data in PDF") as demo:
+        gr.Markdown("""
+        # Incorporate Data in PDF
+
+        Upload a PDF file and add attachments with descriptions. The attachments will be embedded in the PDF file.
+        """)
+
+        # State to store attachments
+        attachments_state = gr.State([])
+
+        # 1. File Upload Section (full width at top)
+        gr.Markdown("### 1. Upload PDF File")
+        pdf_input = gr.File(
+            label="Select PDF File",
+            file_types=[".pdf"],
+            type="filepath"
+        )
+
+        # 2. Middle section with two columns
+        with gr.Row():
+            # Left column: List of current attachments
+            with gr.Column(scale=1):
+                gr.Markdown("### 2. Current Attachments")
+                attachments_display = gr.HTML(
+                    value="<p style='color: #666; font-style: italic;'>No attachments added yet</p>"
+                )
+
+            # Right column: Form to add new attachments
+            with gr.Column(scale=1):
+                gr.Markdown("### 3. Add New Attachment")
+                gr.Markdown("*Maximum file size: 10 MB per attachment*")
+
+                attachment_file = gr.File(
+                    label="Select File to Attach",
+                    file_types=None,  # Allow any file type
+                    type="filepath"
+                )
+
+                attachment_description = gr.Textbox(
+                    label="Attachment Description",
+                    placeholder="Enter a description for this attachment...",
+                    lines=2
+                )
+
+                add_btn = gr.Button("Add Attachment", variant="primary")
+                add_status = gr.Markdown("")
+
+        # 3. Download Section (full width at bottom)
+        gr.Markdown("### 4. Process and Download")
+
+        process_btn = gr.Button("Process PDF with Attachments", variant="primary", size="lg")
+        process_status = gr.Markdown("")
+
+        output_file = gr.File(
+            label="Download PDF with Attachments",
+            visible=False
+        )
+
+        # Event handlers
+        add_btn.click(
+            fn=add_attachment,
+            inputs=[attachment_file, attachment_description, attachments_state],
+            outputs=[attachments_state, attachments_display, add_status]
+        ).then(
+            lambda: (None, ""),  # Clear the file input and description
+            outputs=[attachment_file, attachment_description]
+        )
+
+        process_btn.click(
+            fn=process_pdf_with_attachments,
+            inputs=[pdf_input, attachments_state],
+            outputs=[output_file, process_status]
+        ).then(
+            lambda x: gr.File(visible=x is not None),
+            inputs=[output_file],
+            outputs=[output_file]
+        )
+
+    demo.launch(theme=gr.themes.Soft())
+
+
+if __name__ == "__main__":
+    main()

pyproject.toml

@@ -0,0 +1,16 @@
+[project]
+name = "pdf-incorporate"
+version = "0.1.0"
+description = "Incorporate data in machine readable format within PDF files"
+readme = "README.md"
+requires-python = ">=3.12"
+dependencies = [
+    "gradio[mcp]>=6.2.0",
+    "requests>=2.28",
+    "python-dotenv>=1.2.1",
+    "tiktoken>=0.12.0",
+    "parxy",
+]
+
+[tool.uv.sources]
+parxy = { git = "https://github.com/OneOffTech/parxy/", branch = "refactor-pdf-helpers" }