Cool python libraries

This repositroy demonstrates some of very useful python libraries

Inspired by this video

1) Argh

As the docs says:

Argh is a smart wrapper for argparse. Argparse is a very powerful tool; Argh just makes it easy to use.

Argh installation

pip install argh

Argh usage

Code snippets for this module can be found in directory "argh".

For example, a calculator created using argh:

import argh



def calc(expr):
    print(f'Evaluating expression {expr}')
    try:
        return eval(expr)
    except:
        return 'Given algebraic expression is invalid'


if __name__ == '__main__':
    argh.dispatch_command(calc)

cd argh

python calc.py -h 
usage: calc.py [-h] expr positional arguments: expr -
optional arguments: -h, --help show this help message and exit 

python calc.py
5+5 Evaluating expression 5+5 10

python calc.py 2+2*2 Evaluating expression
2+2*2 6

2) Tqdm

As the docs says:

Tqdm instantly makes your loops show a smart progress meter - just wrap any iterable with tqdm(iterable), and you’re done!

Tqdm installation

pip install tqdm

Tqdm usage

Code snippets for this module can be found in directory "tqdm".

For example, basic tqdm usage:

from tqdm import tqdm
from time import sleep

a = 0

for i in tqdm(range(100)):
    a += i
    sleep(0.1)

print(a)

cd tqdm

python quickstart.py
100%|█████████████████████████████████████████████████████████████| 100/100 [00:10<00:00,  9.86it/s]
4950

3) Typing

Typing forces Python to be strongly-typed language (in combination with mypy). This module provides with a variety of types for type hints: Any, Callable, Union, Dict, List, Iterable and so on.

Docs

Typing installation

No need to install

Typing usage

Code snippets for this module can be found in directory "typing".

For example, functions that accept some types:

def factorial(n: Union[int, float]) -> int:

    # Because None is ignored (accepted)
    if n is None:
        return 0

    n = int(n)

    if n < n:
        return 0
    if n == 0:
        return 1

    return n * factorial(n - 1)


print(factorial(3.5))

'''
python -m mypy quickstart.py

Outputs an error:
error: Argument 1 to "factorial" has incompatible type "str"; expected "int"
Found 1 error in 1 file (checked 1 source file)

'''


def map_int_list(fn: Callable, l: List[int]) -> List[int]:
    return [fn(e) for e in l]


print(map_int_list(factorial, [0, 1, 2, 3, 4]))


def map_int_dict(fn: Callable, d: Dict[Any, int]) -> Dict[Any, int]:
    return {key: fn(value) for key, value in d.items()}


print(map_int_dict(factorial, {'a': 1, 'b': 5, 'c': 4, 'j': 9}))

4) Pydantic

Data validation and settings management using Python type hinting.

Pydantic is a huge help in parsing JSON with such features feild validation, alias for field names and fields exclusion.

Docs

Pydantic installation

pip install pydantic

Pydantic usage

Code snippets for this module can be found in directory "pydantic".

For example, creating models and parsing incoming json:

from pydantic import BaseModel, ValidationError


class City(BaseModel):
    id: int
    name: str
    population: int


incoming_json = '''
{
    "id": 228,
    "name": "Kokshetau",
    "population": "123492"
}
'''

city = City.parse_raw(incoming_json)
print(city)
print(city.id, city.name, city.population)
#> id=228 name='Kokshetau' population=123492
#> 228 Kokshetau 123492

5) Loguru

Very simple logging library.

Loguru makes logging a fulfilling process. It handles all the hardships and hindrances of logging configuration. Moreover, loguru works out of the box just fine.

Docs

Loguru installation

pip install loguru

Loguru usage

Code snippets for this module can be found in directory "loguru".

For example, debugging program scripts and handling exceptions:

from loguru import logger


logger.debug('No need for configuration, handlers etc. Ready to go.')
logger.info('Starting the application.')
logger.warning('Too many connections. Restarting the application.')
logger.critical('DDOS Attack. Shutting down the application.')
logger.error('Cannot start the application.')


try:
    5 / 0
except:
    logger.exception('Execution failed')


@logger.catch
def do_crazy_stuff():
    return 'Hello W' + 0 + 'rld'

do_crazy_stuff()