Python

Fix Python SSL: CERTIFICATE_VERIFY_FAILED Error

You’re making an HTTP request in Python and suddenly: s s l . S S L C e r t V e r i f i c a t i o n E r r o r : [ S S L : C E R T I F I C A T E _ V E R I F Y _ F A I L E D ] c e r t i f i c a t e v e r i f y f a i l e d : u n a b l e t o g e t l o c a l i s s u e r c e r t i f i c a t e Or with requests: ...

Why Your Environment Variables Aren't Loading (And How to Fix It)

You’ve set DATABASE_URL in your .env file. You’ve exported it in your shell. But your application still can’t find it. Sound familiar? Environment variable issues are one of the most common — and frustrating — debugging sessions. Here’s a systematic guide to finding and fixing the problem. The Quick Checklist Before diving deep, check these common culprits: Did you restart your application? Environment variables are read at process start Are you in the right shell session? Variables exported in one terminal don’t exist in another Is your .env file in the right directory? Most libraries look in the current working directory Did you spell it correctly? DATABASE_URL vs DATABSE_URL happens more than you’d think Python: The os.environ Mystery Problem: KeyError When Accessing Variables 1 2 import os database_url = os.environ['DATABASE_URL'] # KeyError! Fix 1: Check if the variable exists first ...

Fix 'Address Already in Use' Error (EADDRINUSE) — Find and Kill the Process

You try to start your server and get hit with: E r r o r : l i s t e n E A D D R I N U S E : a d d r e s s a l r e a d y i n u s e : : : 3 0 0 0 Or the Python equivalent: ...

Fix: Python imaplib search() Returns Empty Results

You’re using Python’s imaplib to search emails, but mail.search() keeps returning empty results even though you know matching emails exist. Here’s why it happens and how to fix it. The Problem 1 2 3 4 5 6 7 8 9 import imaplib mail = imaplib.IMAP4_SSL('imap.gmail.com') mail.login('user@gmail.com', 'app-password') mail.select('INBOX') # This returns nothing! status, messages = mail.search(None, 'FROM "john@example.com" SUBJECT "invoice"') print(messages) # [b''] You’ve verified the emails exist. You can see them in Gmail. But search() returns an empty byte string. ...

Why Your Python Subprocess Hangs (And How to Fix the Deadlock)

You wrote what looks like perfectly reasonable Python code: 1 2 3 4 5 6 7 8 9 10 11 12 import subprocess proc = subprocess.Popen( ["some-command"], stdout=subprocess.PIPE, stderr=subprocess.PIPE ) # Read the output output = proc.stdout.read() errors = proc.stderr.read() proc.wait() And it hangs. Forever. No error, no timeout, just… nothing. Welcome to the pipe buffer deadlock, one of Python’s most frustrating subprocess gotchas. Why This Happens The problem is OS pipe buffers. When you create a subprocess with stdout=subprocess.PIPE, the OS creates a pipe with a fixed buffer size (typically 64KB on Linux). ...

Building Reliable LLM-Powered Features in Production

Adding an LLM to your application is easy. Making it reliable enough for production is another story. API timeouts, rate limits, hallucinations, and surprise $500 invoices await the unprepared. Here’s how to build LLM features that actually work. The Basics: Robust API Calls Never call an LLM API without proper error handling: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 import anthropic from tenacity import retry, stop_after_attempt, wait_exponential import time client = anthropic.Anthropic() @retry( stop=stop_after_attempt(3), wait=wait_exponential(multiplier=1, min=2, max=30), reraise=True ) def call_llm(prompt: str, max_tokens: int = 1024) -> str: try: response = client.messages.create( model="claude-sonnet-4-20250514", max_tokens=max_tokens, messages=[{"role": "user", "content": prompt}] ) return response.content[0].text except anthropic.RateLimitError: time.sleep(60) # Back off on rate limits raise except anthropic.APIStatusError as e: if e.status_code >= 500: raise # Retry on server errors raise # Don't retry on client errors (4xx) Timeouts Are Non-Negotiable LLM calls can hang. Always set timeouts: ...

Production-Ready LLM API Integrations: Patterns That Actually Work

You’ve got your OpenAI or Anthropic API key. The hello-world example works. Now you need to put it in production and suddenly everything is different. LLM APIs have characteristics that break standard integration patterns: high latency, unpredictable response times, token-based billing, and outputs that can vary wildly for the same input. Here’s what actually works. The Unique Challenges Traditional API calls return in milliseconds. LLM calls can take 5-30 seconds. Traditional APIs have predictable costs per call. LLM costs depend on input and output length — and you don’t control the output. ...

LLM API Integration Patterns for Production Applications

Integrating LLMs into production applications is deceptively simple. Call an API, get text back. But building reliable, cost-effective systems requires more thought. Here are patterns that work at scale. The Basic Call Every LLM integration starts here: 1 2 3 4 5 6 7 8 import openai def complete(prompt: str) -> str: response = openai.chat.completions.create( model="gpt-4", messages=[{"role": "user", "content": prompt}] ) return response.choices[0].message.content This works for prototypes. Production needs more. Retry with Exponential Backoff LLM APIs have rate limits and occasional failures: ...

Python Patterns for Command-Line Scripts

Python is the go-to language for automation scripts. Here’s how to write CLI tools that are reliable and user-friendly. Basic Script Structure 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 #!/usr/bin/env python3 """One-line description of what this script does.""" import argparse import sys def main(): parser = argparse.ArgumentParser(description=__doc__) parser.add_argument('input', help='Input file path') parser.add_argument('-o', '--output', help='Output file path') parser.add_argument('-v', '--verbose', action='store_true') args = parser.parse_args() # Your logic here process(args.input, args.output, args.verbose) if __name__ == '__main__': main() Argument Parsing with argparse Positional Arguments 1 2 3 4 parser.add_argument('filename') # Required parser.add_argument('files', nargs='+') # One or more parser.add_argument('files', nargs='*') # Zero or more parser.add_argument('config', nargs='?') # Optional positional Optional Arguments 1 2 3 4 5 parser.add_argument('-v', '--verbose', action='store_true') parser.add_argument('-q', '--quiet', action='store_false', dest='verbose') parser.add_argument('-n', '--count', type=int, default=10) parser.add_argument('-f', '--format', choices=['json', 'csv', 'table']) parser.add_argument('--config', type=argparse.FileType('r')) Subcommands 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 parser = argparse.ArgumentParser() subparsers = parser.add_subparsers(dest='command', required=True) # 'init' command init_parser = subparsers.add_parser('init', help='Initialize project') init_parser.add_argument('--force', action='store_true') # 'run' command run_parser = subparsers.add_parser('run', help='Run the application') run_parser.add_argument('--port', type=int, default=8080) args = parser.parse_args() if args.command == 'init': do_init(args.force) elif args.command == 'run': do_run(args.port) Error Handling 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 import sys def main(): try: result = process() return 0 except FileNotFoundError as e: print(f"Error: File not found: {e.filename}", file=sys.stderr) return 1 except PermissionError: print("Error: Permission denied", file=sys.stderr) return 1 except KeyboardInterrupt: print("\nInterrupted", file=sys.stderr) return 130 except Exception as e: print(f"Error: {e}", file=sys.stderr) return 1 if __name__ == '__main__': sys.exit(main()) Logging 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 import logging def setup_logging(verbose=False): level = logging.DEBUG if verbose else logging.INFO logging.basicConfig( level=level, format='%(asctime)s - %(levelname)s - %(message)s', datefmt='%Y-%m-%d %H:%M:%S' ) def main(): args = parse_args() setup_logging(args.verbose) logging.info("Starting process") logging.debug("Detailed info here") logging.warning("Something might be wrong") logging.error("Something went wrong") Log to File and Console 1 2 3 4 5 6 7 8 9 10 11 def setup_logging(verbose=False, log_file=None): handlers = [logging.StreamHandler()] if log_file: handlers.append(logging.FileHandler(log_file)) logging.basicConfig( level=logging.DEBUG if verbose else logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s', handlers=handlers ) Progress Indicators Simple Progress 1 2 3 4 5 6 7 8 import sys def process_items(items): total = len(items) for i, item in enumerate(items, 1): process(item) print(f"\rProcessing: {i}/{total}", end='', flush=True) print() # Newline at end With tqdm 1 2 3 4 5 6 7 8 9 10 from tqdm import tqdm for item in tqdm(items, desc="Processing"): process(item) # Or wrap any iterable with tqdm(total=100) as pbar: for i in range(100): do_work() pbar.update(1) Reading Input From File or Stdin 1 2 3 4 5 6 7 8 9 10 import sys def read_input(filepath=None): if filepath: with open(filepath) as f: return f.read() elif not sys.stdin.isatty(): return sys.stdin.read() else: raise ValueError("No input provided") Line by Line 1 2 3 4 5 import fileinput # Reads from files in args or stdin for line in fileinput.input(): process(line.strip()) Output Formatting JSON Output 1 2 3 4 5 6 7 import json def output_json(data, pretty=False): if pretty: print(json.dumps(data, indent=2, default=str)) else: print(json.dumps(data, default=str)) Table Output 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 def print_table(headers, rows): # Calculate column widths widths = [len(h) for h in headers] for row in rows: for i, cell in enumerate(row): widths[i] = max(widths[i], len(str(cell))) # Print header header_line = ' | '.join(h.ljust(widths[i]) for i, h in enumerate(headers)) print(header_line) print('-' * len(header_line)) # Print rows for row in rows: print(' | '.join(str(cell).ljust(widths[i]) for i, cell in enumerate(row))) With tabulate 1 2 3 4 5 6 7 from tabulate import tabulate data = [ ['Alice', 30, 'Engineer'], ['Bob', 25, 'Designer'], ] print(tabulate(data, headers=['Name', 'Age', 'Role'], tablefmt='grid')) Configuration Files YAML Config 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 import yaml from pathlib import Path def load_config(config_path=None): paths = [ config_path, Path.home() / '.myapp.yaml', Path('/etc/myapp/config.yaml'), ] for path in paths: if path and Path(path).exists(): with open(path) as f: return yaml.safe_load(f) return {} # Defaults Environment Variables 1 2 3 4 5 6 7 8 import os def get_config(): return { 'api_key': os.environ.get('API_KEY'), 'debug': os.environ.get('DEBUG', '').lower() in ('true', '1', 'yes'), 'timeout': int(os.environ.get('TIMEOUT', '30')), } Running External Commands 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 import subprocess def run_command(cmd, check=True): """Run command and return output.""" result = subprocess.run( cmd, shell=isinstance(cmd, str), capture_output=True, text=True, check=check ) return result.stdout.strip() # Usage output = run_command(['git', 'status', '--short']) output = run_command('ls -la | head -5') With Timeout 1 2 3 4 5 6 7 8 9 try: result = subprocess.run( ['slow-command'], timeout=30, capture_output=True, text=True ) except subprocess.TimeoutExpired: print("Command timed out") Temporary Files 1 2 3 4 5 6 7 8 9 10 11 12 13 import tempfile from pathlib import Path # Temporary file with tempfile.NamedTemporaryFile(mode='w', suffix='.json', delete=False) as f: f.write('{"data": "value"}') temp_path = f.name # Temporary directory with tempfile.TemporaryDirectory() as tmpdir: work_file = Path(tmpdir) / 'work.txt' work_file.write_text('working...') # Directory deleted when context exits Path Handling 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 from pathlib import Path def process_files(directory): base = Path(directory) # Find files for path in base.glob('**/*.py'): print(f"Processing: {path}") # Path operations print(f" Name: {path.name}") print(f" Stem: {path.stem}") print(f" Suffix: {path.suffix}") print(f" Parent: {path.parent}") # Read/write content = path.read_text() path.with_suffix('.bak').write_text(content) Complete Example 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 #!/usr/bin/env python3 """Process log files and output statistics.""" import argparse import json import logging import sys from collections import Counter from pathlib import Path def setup_logging(verbose): logging.basicConfig( level=logging.DEBUG if verbose else logging.INFO, format='%(levelname)s: %(message)s' ) def parse_args(): parser = argparse.ArgumentParser( description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter ) parser.add_argument( 'logfiles', nargs='+', type=Path, help='Log files to process' ) parser.add_argument( '-o', '--output', type=argparse.FileType('w'), default=sys.stdout, help='Output file (default: stdout)' ) parser.add_argument( '-f', '--format', choices=['json', 'text'], default='text', help='Output format' ) parser.add_argument( '-v', '--verbose', action='store_true', help='Enable verbose output' ) return parser.parse_args() def analyze_logs(logfiles): stats = Counter() for logfile in logfiles: logging.info(f"Processing {logfile}") if not logfile.exists(): logging.warning(f"File not found: {logfile}") continue for line in logfile.read_text().splitlines(): if 'ERROR' in line: stats['errors'] += 1 elif 'WARNING' in line: stats['warnings'] += 1 stats['total'] += 1 return dict(stats) def output_results(stats, output, fmt): if fmt == 'json': json.dump(stats, output, indent=2) output.write('\n') else: for key, value in stats.items(): output.write(f"{key}: {value}\n") def main(): args = parse_args() setup_logging(args.verbose) try: stats = analyze_logs(args.logfiles) output_results(stats, args.output, args.format) return 0 except Exception as e: logging.error(f"Failed: {e}") return 1 if __name__ == '__main__': sys.exit(main()) Usage: ...

Getting Structured Data from LLMs: JSON Mode and Beyond

The biggest challenge with LLMs in production isn’t getting good responses—it’s getting parseable responses. When you need JSON for your pipeline, “Here’s the data you requested:” followed by markdown-wrapped output breaks everything. Here’s how to reliably extract structured data. The Problem 1 2 3 4 5 6 7 8 response = client.chat.completions.create( model="gpt-4", messages=[{"role": "user", "content": "Extract the person's name and age from: 'John Smith is 34 years old'"}] ) print(response.choices[0].message.content) # "The person's name is John Smith and their age is 34." # ... not what we needed You wanted {"name": "John Smith", "age": 34}. You got prose. ...