cd /mnt/github/workspace-hub/pdf-large-reader
pip install -e .

from pdf_large_reader import process_large_pdf, extract_text_only

# Simple text extraction
text = extract_text_only("large_document.pdf")

# Full processing with options
result = process_large_pdf(
    "document.pdf",
    output_format="generator",  # "generator", "list", or "text"
    extract_images=True,
    extract_tables=True
)

# Stream pages for memory efficiency
for page in result:
    print(f"Page {page['page_num']}: {page['text'][:100]}...")

# Extract text
pdf-large-reader extract document.pdf

# Extract with options
pdf-large-reader extract document.pdf --output-format text --extract-images

# Get PDF info
pdf-large-reader info document.pdf

for page in process_large_pdf("doc.pdf", output_format="generator"):
    # Process one page at a time - memory efficient
    handle_page(page)

pages = process_large_pdf("doc.pdf", output_format="list")
# All pages in memory - use for smaller files

text = process_large_pdf("doc.pdf", output_format="text")
# Plain text concatenation

from pdf_large_reader import process_large_pdf
from pathlib import Path

def process_pdf_directory(directory: str):
    """Process all PDFs in a directory efficiently."""
    pdf_dir = Path(directory)

    for pdf_file in pdf_dir.glob("*.pdf"):
        print(f"Processing: {pdf_file.name}")

        # Stream pages to avoid memory issues
        for page in process_large_pdf(str(pdf_file)):
            # Index, analyze, or store each page
            yield {
                "file": pdf_file.name,
                "page": page["page_num"],
                "text": page["text"],
                "images": len(page.get("images", []))
            }

from pdf_large_reader import process_large_pdf
from pdf_large_reader.exceptions import PDFProcessingError

try:
    result = process_large_pdf("document.pdf")
except PDFProcessingError as e:
    print(f"PDF processing failed: {e}")
except FileNotFoundError:
    print("PDF file not found")