# Django ORM Comparison - Know both frameworks
# Django model (contrast with SQLAlchemy)
from django.db import models

class Department(models.Model):
    name = models.CharField(max_length=50)

class Employee(models.Model):
    name = models.CharField(max_length=100)
    email = models.EmailField(unique=True)
    department = models.ForeignKey(Department, on_delete=models.CASCADE)

# Django query (similar but different syntax)
Employee.objects.filter(department__name="HR").select_related('department')

# Async ORM - Modern Python requirement
# Requires SQLAlchemy 1.4+ and asyncpg
from sqlalchemy.ext.asyncio import create_async_engine, AsyncSession

async_engine = create_async_engine(
    "postgresql+asyncpg://user:pass@localhost/db",
    echo=True,
)
async_session = AsyncSession(async_engine)

async with async_session.begin():
    result = await async_session.execute(
        select(Employee).where(Employee.name == "Alice")
    )
    employee = result.scalar_one()

# Testing Strategies - Interview differentiator
from unittest import mock

# Mock database for unit tests
with mock.patch('sqlalchemy.create_engine') as mock_engine:
    mock_conn = mock.MagicMock()
    mock_engine.return_value.connect.return_value = mock_conn
    
    # Test your ORM-dependent code
    create_employee("Test", "test@company.com")
    mock_conn.execute.assert_called()

# Production Monitoring - Track slow queries
from sqlalchemy import event

@event.listens_for(engine, "before_cursor_execute")
def before_cursor(conn, cursor, statement, params, context, executemany):
    conn.info.setdefault('query_start_time', []).append(time.time())

@event.listens_for(engine, "after_cursor_execute")
def after_cursor(conn, cursor, statement, params, context, executemany):
    total = time.time() - conn.info['query_start_time'].pop(-1)
    if total > 0.1:  # Log slow queries
        print(f"SLOW QUERY ({total:.2f}s): {statement}")

# Interview Power Move: Implement caching layer
from functools import lru_cache

class CachedEmployeeRepository(EmployeeRepository):
    @lru_cache(maxsize=100)
    def get_by_id(self, employee_id):
        return super().get_by_id(employee_id)
    
    def invalidate_cache(self, employee_id):
        self.get_by_id.cache_clear()

# Reduces database hits by 70% in read-heavy applications

# Pro Tip: Schema versioning in CI/CD pipelines
# Sample .gitlab-ci.yml snippet
deploy_db:
  stage: deploy
  script:
    - alembic upgrade head
    - pytest tests/db_tests.py  # Verify schema compatibility
  only:
    - main

# Real-World Case Study: E-commerce inventory system
class Product(Base):
    __tablename__ = 'products'
    id = Column(Integer, primary_key=True)
    sku = Column(String(20), unique=True)
    stock = Column(Integer, default=0)
    
    # Atomic stock update (prevents race conditions)
    def decrement_stock(self, quantity, session):
        result = session.query(Product).filter(
            Product.id == self.id,
            Product.stock >= quantity
        ).update({"stock": Product.stock - quantity})
        if not result:
            raise ValueError("Insufficient stock")

# Usage during checkout
product.decrement_stock(2, session)

# Basic Merging - The absolute foundation
from PyPDF2 import PdfMerger

merger = PdfMerger()
pdf_files = ["report1.pdf", "report2.pdf", "summary.pdf"]

for file in pdf_files:
    merger.append(file)

merger.write("combined_report.pdf")
merger.close()

# Merge Specific Pages - Precision control
merger = PdfMerger()
merger.append("full_document.pdf", pages=(0, 3))  # First 3 pages
merger.append("appendix.pdf", pages=(2, 5))       # Pages 3-5 (0-indexed)
merger.write("custom_merge.pdf")

# Insert Pages at Position - Structured document assembly
merger = PdfMerger()
merger.append("cover.pdf")
merger.merge(1, "content.pdf")  # Insert at index 1
merger.merge(2, "charts.pdf", pages=(4, 6))  # Insert specific pages
merger.write("structured_report.pdf")

# Handling Encrypted PDFs - Production reality
merger = PdfMerger()
merger.append("secure_doc.pdf", password="secret123")
merger.write("decrypted_merge.pdf")

# Bookmarks for Navigation - Professional touch
merger = PdfMerger()
merger.append("chapter1.pdf", outline_item="Introduction")
merger.append("chapter2.pdf", outline_item="Methodology")
merger.append("chapter3.pdf", outline_item="Results")
merger.write("bookmarked_report.pdf")

# Memory Optimization - Critical for large files
from PyPDF2 import PdfReader

merger = PdfMerger()
for file in ["large1.pdf", "large2.pdf"]:
    reader = PdfReader(file)
    merger.append(reader)
    del reader  # Immediate memory cleanup
merger.write("optimized_merge.pdf")

# Batch Processing - Real-world automation
import os
from PyPDF2 import PdfMerger

def merge_pdfs_in_folder(folder, output="combined.pdf"):
    merger = PdfMerger()
    for file in sorted(os.listdir(folder)):
        if file.endswith(".pdf"):
            merger.append(f"{folder}/{file}")
    merger.write(output)
    merger.close()

merge_pdfs_in_folder("quarterly_reports", "Q3_results.pdf")

# Error Handling - Production-grade code
from PyPDF2 import PdfMerger, PdfReadError

def safe_merge(inputs, output):
    merger = PdfMerger()
    try:
        for file in inputs:
            try:
                merger.append(file)
            except PdfReadError:
                print(f"Skipping corrupted: {file}")
    finally:
        merger.write(output)
        merger.close()

safe_merge(["valid.pdf", "corrupted.pdf", "valid2.pdf"], "partial_merge.pdf")

# Metadata Preservation - Legal/compliance requirement
merger = PdfMerger()
merger.append("source.pdf")

# Copy metadata from first document
meta = merger.metadata
merger.add_metadata({
    **meta,
    "/Producer": "Python Automation v3.0",
    "/CustomField": "CONFIDENTIAL"
})
merger.write("metadata_enhanced.pdf")

# Encryption of Output - Security interview question
merger = PdfMerger()
merger.append("sensitive_data.pdf")

merger.encrypt(
    user_pwd="view_only",
    owner_pwd="full_access",
    use_128bit=True
)
merger.write("encrypted_report.pdf")

# Page Rotation - Fix orientation issues
merger = PdfMerger()
merger.append("landscape_charts.pdf", pages=(0, 2), import_outline=False)
merger.merge(0, "portrait_text.pdf")  # Rotate during merge
merger.write("standardized_orientation.pdf")

# Watermarking During Merge - Branding automation
from PyPDF2 import PdfWriter, PdfReader

def add_watermark(input_pdf, watermark_pdf, output_pdf):
    watermark = PdfReader(watermark_pdf).pages[0]
    output = PdfWriter()
    
    with open(input_pdf, "rb") as f:
        reader = PdfReader(f)
        for page in reader.pages:
            page.merge_page(watermark)
            output.add_page(page)
    
    with open(output_pdf, "wb") as f:
        output.write(f)

# Apply during merge process
add_watermark("report.pdf", "watermark.pdf", "branded.pdf")

# Async Merging - Modern Python requirement
import asyncio
from PyPDF2 import PdfMerger

async def async_merge(files, output):
    merger = PdfMerger()
    for file in files:
        await asyncio.to_thread(merger.append, file)
    merger.write(output)

# Usage in async application
asyncio.run(async_merge(["doc1.pdf", "doc2.pdf"], "async_merge.pdf"))

# CLI Tool Implementation - Interview favorite
import sys
from PyPDF2 import PdfMerger

def main():
    if len(sys.argv) < 3:
        print("Usage: pdfmerge output.pdf input1.pdf input2.pdf ...")
        sys.exit(1)
    
    merger = PdfMerger()
    for pdf in sys.argv[2:]:
        merger.append(pdf)
    merger.write(sys.argv[1])

if __name__ == "__main__":
    main()
# Run via: python pdfmerge.py final.pdf *.pdf

# Performance Benchmarking - Optimization proof
import time
from PyPDF2 import PdfMerger

start = time.time()
merger = PdfMerger()
for _ in range(50):
    merger.append("sample.pdf")
merger.write("50x_merge.pdf")
print(f"Time: {time.time()-start:.2f}s")  # Baseline for optimization

# Memory-Mapped Processing - Handle 1GB+ files
import mmap
from PyPDF2 import PdfMerger

def memmap_merge(large_files, output):
    merger = PdfMerger()
    for file in large_files:
        with open(file, "rb") as f:
            mmapped = mmap.mmap(f.fileno(), 0, access=mmap.ACCESS_READ)
            merger.append(mmapped)
    merger.write(output)

memmap_merge(["huge1.pdf", "huge2.pdf"], "giant_merge.pdf")

# PDF/A Compliance - Archival standards
merger = PdfMerger()
merger.append("archive_source.pdf")

# Convert to PDF/A-1b standard
merger.add_metadata({
    "/GTS_PDFXVersion": "PDF/A-1b",
    "/GTS_PDFXConformance": "B"
})
merger.write("compliant_archive.pdf")

# Split and Re-Merge Workflow - Advanced manipulation
from PyPDF2 import PdfReader, PdfWriter

def split_and_merge(source, chunk_size=10):
    reader = PdfReader(source)
    chunks = [reader.pages[i:i+chunk_size] for i in range(0, len(reader.pages), chunk_size)]
    
    for i, chunk in enumerate(chunks):
        writer = PdfWriter()
        for page in chunk:
            writer.add_page(page)
        with open(f"chunk_{i}.pdf", "wb") as f:
            writer.write(f)
    
    # Now merge chunks with new order
    merger = PdfMerger()
    for i in reversed(range(len(chunks))):
        merger.append(f"chunk_{i}.pdf")
    merger.write("reversed_document.pdf")

split_and_merge("master.pdf")

# Cloud Integration - Production pipeline example
from google.cloud import storage
from PyPDF2 import PdfMerger

def merge_from_gcs(bucket_name, prefix, output_path):
    storage_client = storage.Client()
    bucket = storage_client.bucket(bucket_name)
    blobs = bucket.list_blobs(prefix=prefix)
    
    merger = PdfMerger()
    for blob in blobs:
        if blob.name.endswith(".pdf"):
            temp_path = f"/tmp/{blob.name.split('/')[-1]}"
            blob.download_to_filename(temp_path)
            merger.append(temp_path)
    
    merger.write(output_path)
    merger.close()

merge_from_gcs("client-reports", "Q3/", "/tmp/merged.pdf")

# Dockerized Microservice - Deployment pattern
# Dockerfile snippet:
# FROM python:3.10-slim
# RUN pip install pypdf
# COPY merge_service.py /app/
# CMD ["python", "/app/merge_service.py"]

# merge_service.py
from http.server import HTTPServer, BaseHTTPRequestHandler
from PyPDF2 import PdfMerger
import json

class MergeHandler(BaseHTTPRequestHandler):
    def do_POST(self):
        content_len = int(self.headers.get('Content-Length'))
        body = json.loads(self.rfile.read(content_len))
        
        merger = PdfMerger()
        for url in body['inputs']:
            # Download from URLs (simplified)
            merger.append(download_pdf(url))
        merger.write("/output/merged.pdf")
        
        self.send_response(200)
        self.end_headers()

HTTPServer(('', 8000), MergeHandler).serve_forever()

# Interview Power Move: Parallel Merging
from concurrent.futures import ThreadPoolExecutor
from PyPDF2 import PdfMerger

def parallel_merge(pdf_list, output, max_workers=4):
    chunks = [pdf_list[i::max_workers] for i in range(max_workers)]
    temp_files = []
    
    def merge_chunk(chunk, idx):
        temp = f"temp_{idx}.pdf"
        merger = PdfMerger()
        for pdf in chunk:
            merger.append(pdf)
        merger.write(temp)
        return temp
    
    with ThreadPoolExecutor() as executor:
        temp_files = list(executor.map(merge_chunk, chunks, range(max_workers)))
    
    # Final merge of chunks
    final_merger = PdfMerger()
    for temp in temp_files:
        final_merger.append(temp)
    final_merger.write(output)

parallel_merge(["doc1.pdf", "doc2.pdf", ...], "parallel_merge.pdf")

# Pro Tip: Validate PDFs before merging
from PyPDF2 import PdfReader

def is_valid_pdf(path):
    try:
        with open(path, "rb") as f:
            reader = PdfReader(f)
            return len(reader.pages) > 0
    except:
        return False

valid_pdfs = [f for f in pdf_files if is_valid_pdf(f)]
merger.append(valid_pdfs)  # Only merge valid files

# Real-World Case Study: Invoice Processing Pipeline
import glob
from PyPDF2 import PdfMerger

def process_monthly_invoices():
    # 1. Download invoices from SFTP
    download_invoices("sftp://vendor.com/invoices/*.pdf")
    
    # 2. Validate and sort
    invoices = sorted(
        [f for f in glob.glob("invoices/*.pdf") if is_valid_pdf(f)],
        key=lambda x: extract_invoice_date(x)
    )
    
    # 3. Merge with cover page
    merger = PdfMerger()
    merger.append("cover_template.pdf")
    for inv in invoices:
        merger.append(inv, outline_item=get_client_name(inv))
    
    # 4. Add metadata and encrypt
    merger.add_metadata({"/InvoiceCount": str(len(invoices))})
    merger.encrypt(owner_pwd="finance_team_2023")
    merger.write(f"Q3_Invoices_{datetime.now().strftime('%Y%m')}.pdf")
    
    # 5. Upload to secure storage
    upload_to_s3("secure-bucket/processed/", "Q3_Invoices.pdf")

process_monthly_invoices()

from num2words import num2words

# Example 1: Basic integer
number1 = 123
words1 = num2words(number1)
print(f"'{number1}' in words: {words1}")

# Example 2: Larger integer
number2 = 543210
words2 = num2words(number2, lang='en') # Explicitly set language
print(f"'{number2}' in words: {words2}")

# Example 3: Decimal number
number3 = 100.75
words3 = num2words(number3)
print(f"'{number3}' in words: {words3}")

# Example 4: Negative number
number4 = -45
words4 = num2words(number4)
print(f"'{number4}' in words: {words4}")

# Example 5: Number for an ordinal form
number5 = 3
words5 = num2words(number5, to='ordinal')
print(f"Ordinal '{number5}' in words: {words5}")