Added example for docling

examples/Python3.12/docling-parse-example/README.md

@@ -0,0 +1,45 @@
+## Purpose: Simple RAG (Retrieval-Augmented Generation) Application with Docling-Parse
+
+### Packages used:
+- docling-parse (PDF parsing)
+- reportlab (PDF creation for testing)
+- Pillow (Image processing support)
+- numpy (Numerical operations)
+
+### Functionality:
+This example demonstrates a simplified RAG (Retrieval-Augmented Generation) pipeline:
+
+1. **Document Parsing**: Uses docling-parse v2 to extract text from PDF documents
+2. **Text Chunking**: Splits extracted text into manageable chunks for retrieval
+3. **Keyword-based Retrieval**: Implements simple keyword matching to find relevant chunks
+4. **Result Display**: Shows top matching chunks based on query
+
+### RAG Pipeline Steps:
+1. Parse PDF document with docling-parse
+2. Extract text content from parsed document
+3. Create document chunks (configurable size)
+4. Perform keyword-based retrieval on chunks
+5. Display top relevant results
+
+### Note:
+This is a simplified RAG demonstration. Production RAG systems typically use:
+- Vector embeddings (e.g., sentence-transformers, OpenAI embeddings)
+- Vector databases (e.g., FAISS, Chroma, Pinecone)
+- LLM for generation (e.g., vLLM, Ollama, OpenAI GPT)
+
+### How to run the example:
+```bash
+chmod +x install_test_example.sh
+./install_test_example.sh
+```
+
+Or manually:
+```bash
+python3.12 -m venv .venv
+source .venv/bin/activate
+pip install --extra-index-url https://wheels.developerfirst.ibm.com/ppc64le/linux -r requirements.txt
+python3.12 docling_parse_example.py [optional_pdf_path]
+```
+
+### License:
+It's covered under Apache 2.0 licenses

examples/Python3.12/docling-parse-example/docling_parse_example.py

@@ -0,0 +1,209 @@
+import sys
+from pathlib import Path
+from typing import List, Dict
+from docling_parse.pdf_parser import DoclingPdfParser
+
+def parse_pdf_with_docling(pdf_path: str) -> str:
+    """
+    Parse a PDF document using docling-parse and extract text content.
+
+    Args:
+        pdf_path: Path to the PDF file to parse
+
+    Returns:
+        Extracted text content from the PDF
+    """
+    try:
+        # Create parser instance and load the PDF document
+        parser = DoclingPdfParser()
+        doc = parser.load(pdf_path)
+
+        # Extract text content from all pages
+        text_content = ""
+
+        for page_num in range(1, doc.number_of_pages() + 1):
+            page = doc.get_page(page_num)
+
+            # Extract text from word cells
+            if hasattr(page, 'word_cells') and page.word_cells:
+                for cell in page.word_cells:
+                    text_content += cell.text + " "
+                text_content += "\n"
+
+        return text_content.strip() if text_content else "No text content extracted"
+
+    except Exception as e:
+        return f"Error parsing PDF: {str(e)}"
+
+
+def create_document_chunks(text: str, chunk_size: int = 500) -> List[str]:
+    """
+    Split document text into smaller chunks for RAG processing.
+
+    Args:
+        text: Full document text
+        chunk_size: Size of each chunk in characters
+
+    Returns:
+        List of text chunks
+    """
+    # Simple chunking by character count
+    chunks = []
+    words = text.split()
+    current_chunk = []
+    current_size = 0
+
+    for word in words:
+        word_size = len(word) + 1  # +1 for space
+        if current_size + word_size > chunk_size and current_chunk:
+            chunks.append(" ".join(current_chunk))
+            current_chunk = [word]
+            current_size = word_size
+        else:
+            current_chunk.append(word)
+            current_size += word_size
+
+    if current_chunk:
+        chunks.append(" ".join(current_chunk))
+
+    return chunks
+
+
+def simple_keyword_search(chunks: List[str], query: str) -> List[Dict[str, any]]:
+    """
+    Simple keyword-based retrieval (mock RAG retrieval).
+    In a real RAG system, this would use embeddings and vector search.
+
+    Args:
+        chunks: List of document chunks
+        query: Search query
+
+    Returns:
+        List of relevant chunks with scores
+    """
+    results = []
+    query_lower = query.lower()
+    query_words = set(query_lower.split())
+
+    for idx, chunk in enumerate(chunks):
+        chunk_lower = chunk.lower()
+        # Simple scoring: count matching words
+        matches = sum(1 for word in query_words if word in chunk_lower)
+
+        if matches > 0:
+            results.append({
+                "chunk_id": idx,
+                "text": chunk,
+                "score": matches,
+                "preview": chunk[:200] + "..." if len(chunk) > 200 else chunk
+            })
+
+    # Sort by score (descending)
+    results.sort(key=lambda x: x["score"], reverse=True)
+    return results
+
+
+def create_sample_pdf():
+    """
+    Create a sample PDF document with RAG-relevant content.
+    """
+    try:
+        from reportlab.pdfgen import canvas
+        from reportlab.lib.pagesizes import letter
+
+        pdf_path = "rag_sample_document.pdf"
+        c = canvas.Canvas(pdf_path, pagesize=letter)
+
+        # Add content about AI and machine learning
+        y_position = 750
+        content = [
+            "Introduction to Artificial Intelligence",
+            "",
+            "Artificial Intelligence (AI) is the simulation of human intelligence",
+            "processes by machines, especially computer systems. These processes",
+            "include learning, reasoning, and self-correction.",
+            "",
+            "Machine Learning is a subset of AI that provides systems the ability",
+            "to automatically learn and improve from experience without being",
+            "explicitly programmed. Deep learning is a subset of machine learning.",
+            "",
+            "Natural Language Processing (NLP) is a branch of AI that helps",
+            "computers understand, interpret and manipulate human language.",
+            "NLP draws from many disciplines, including computer science and",
+            "computational linguistics.",
+        ]
+
+        for line in content:
+            c.drawString(50, y_position, line)
+            y_position -= 20
+
+        c.showPage()
+        c.save()
+
+        print(f"✓ Created sample PDF: {pdf_path}")
+        return pdf_path
+
+    except ImportError:
+        print("✗ reportlab not available, skipping PDF creation")
+        return None
+
+
+if __name__ == "__main__":
+    print("=" * 70)
+    print("Simple RAG Application with Docling-Parse")
+    print("=" * 70)
+
+    # Step 1: Get or create PDF document
+    if len(sys.argv) > 1:
+        pdf_path = sys.argv[1]
+        print(f"\n[1/5] Using provided PDF: {pdf_path}")
+    else:
+        print("\n[1/5] Creating sample PDF document...")
+        pdf_path = create_sample_pdf()
+        if not pdf_path:
+            print("\n✗ No PDF provided and couldn't create sample PDF")
+            print("Usage: python docling_parse_example.py [path_to_pdf]")
+            sys.exit(1)
+
+    # Check if file exists
+    if not Path(pdf_path).exists():
+        print(f"\n✗ Error: PDF file not found: {pdf_path}")
+        sys.exit(1)
+
+    # Step 2: Parse PDF with docling-parse
+    print("\n[2/5] Parsing PDF with docling-parse...")
+    document_text = parse_pdf_with_docling(pdf_path)
+    print(f"✓ Extracted {len(document_text)} characters")
+
+    # Step 3: Create document chunks
+    print("\n[3/5] Creating document chunks for RAG...")
+    chunks = create_document_chunks(document_text, chunk_size=300)
+    print(f"✓ Created {len(chunks)} chunks")
+
+    # Step 4: Demonstrate retrieval
+    print("\n[4/5] Performing keyword-based retrieval...")
+    query = "machine learning"
+    print(f"Query: '{query}'")
+
+    results = simple_keyword_search(chunks, query)
+    print(f"✓ Found {len(results)} relevant chunks")
+
+    # Step 5: Display results
+    print("\n[5/5] Top Retrieved Chunks:")
+    print("=" * 70)
+
+    for i, result in enumerate(results[:3], 1):  # Show top 3
+        print(f"\nChunk #{result['chunk_id']} (Score: {result['score']})")
+        print("-" * 70)
+        print(result['preview'])
+
+    print("\n" + "=" * 70)
+    print("RAG Example completed successfully!")
+    print("=" * 70)
+    print("\nNote: This is a simplified RAG demonstration.")
+    print("Production RAG systems would use:")
+    print("  - Vector embeddings (e.g., sentence-transformers)")
+    print("  - Vector databases (e.g., FAISS, Chroma)")
+    print("  - LLM for generation (e.g., vLLM, Ollama)")
+    print("=" * 70)
+

examples/Python3.12/docling-parse-example/install_test_example.sh

@@ -0,0 +1,67 @@
+#!/bin/bash
+# -------------------------------
+# Function to detect Linux distro
+# -------------------------------
+detect_distro() {
+    if [ -f /etc/os-release ]; then
+        . /etc/os-release
+        echo $ID
+    elif [ -f /etc/redhat-release ]; then
+        echo "rhel"
+    elif [ -f /etc/debian_version ]; then
+        echo "debian"
+    else
+        echo "unknown"
+    fi
+}
+
+DISTRO=$(detect_distro)
+echo "Detected distribution: $DISTRO"
+echo "Installing prerequisites..."
+
+# -------------------------------
+# Install system dependencies
+# -------------------------------
+case $DISTRO in
+    "fedora"|"rhel"|"centos"|"rocky"|"almalinux")
+        if command -v dnf >/dev/null 2>&1; then
+            sudo dnf install -y python3.12 python3.12-devel python3-pip 
+        else
+            sudo yum install -y python3.12 python3.12-devel python3-pip 
+        fi
+        ;;
+    "ubuntu"|"debian")
+        export DEBIAN_FRONTEND=noninteractive  
+        sudo apt update
+        sudo apt install -y python3.12 python3.12-dev python3-pip python3.12-venv 
+        ;;
+    "sles")
+        sudo zypper refresh
+        sudo zypper install -y python312 python312-pip
+
+        ;;
+    *)
+        echo "Unsupported distribution: $DISTRO"
+        exit 1
+        ;;
+esac
+
+
+
+python3.12 -m venv .venv
+source .venv/bin/activate
+
+python3.12 -m pip install --no-cache --prefer-binary --extra-index-url https://wheels.developerfirst.ibm.com/ppc64le/linux -r requirements.txt
+
+WORKDIR=$(pwd)
+
+cd $WORKDIR
+
+python3.12 docling_parse_example.py
+
+echo "\n ==== Running tests ==== \n"
+
+python3.12 sub-test1.py
+python3.12 sub-test2.py
+python3.12 sub-test3.py
+

examples/Python3.12/docling-parse-example/requirements.txt

@@ -0,0 +1,4 @@
+docling-parse==5.8.0+ppc64le1
+reportlab==4.2.5
+Pillow==11.1.0
+numpy==2.4.4+ppc64le1

examples/Python3.12/docling-parse-example/sub-test1.py

@@ -0,0 +1,36 @@
+import unittest
+import importlib.metadata
+
+class TestDoclingParseLibrary(unittest.TestCase):
+    def test_docling_parse_import(self):
+        """Check if docling-parse can be imported"""
+        try:
+            import docling_parse
+        except ImportError:
+            self.fail("docling-parse is not installed")
+
+    def test_docling_parse_version(self):
+        """Verify docling-parse version"""
+        version = importlib.metadata.version("docling-parse")
+        assert "5.8.0" in version, f"Expected docling-parse 5.8.0, got {version}"
+
+    def test_pdf_parser_import(self):
+        """Check if DoclingPdfParser can be imported"""
+        try:
+            from docling_parse.pdf_parser import DoclingPdfParser
+            self.assertIsNotNone(DoclingPdfParser, "DoclingPdfParser should not be None")
+        except ImportError as e:
+            self.fail(f"Failed to import DoclingPdfParser: {e}")
+
+    def test_pdf_parser_instantiation(self):
+        """Check if DoclingPdfParser can be instantiated"""
+        try:
+            from docling_parse.pdf_parser import DoclingPdfParser
+            parser = DoclingPdfParser()
+            self.assertIsNotNone(parser, "Parser instance should not be None")
+        except Exception as e:
+            self.fail(f"Failed to instantiate DoclingPdfParser: {e}")
+
+if __name__ == "__main__":
+    unittest.main()
+