cli_tools_by_oleksa
A zero-dependency Python library for robust console input (validation, menus, type conversion) and cross-platform terminal control and styling.
Purpose
This toolkit is designed to eliminate the notorious boilerplate required for reliable interactive console applications. It allows developers to completely bypass the need for writing repetitive input validation loops (while True: try/except) and managing fragile ANSI escape sequences for styling and cursor control.
Target Audience
cli_tools_by_oleksa is the perfect choice for projects where simplicity and minimal dependencies are paramount:
- Students and Educators: Focus on logic, not boilerplate. Write clean code without reinventing complex validation every time.
- Utility & Script Developers: Build robust, interactive tools and prototypes quickly, without the overhead of heavy frameworks like Click or Typer.
- Minimalist Developers: Achieve production-level reliability and polished UI while maintaining a zero-dependency footprint.
Source Code
https://github.com/01eksa/cli_tools_by_oleksa
Core Features
Input & Validation
get_input— The central function for user input. Features a robust pipeline (RegEx → Validator → Converter) and an automatic retry loop (retry=Trueby default).get_password— Securely prompts the user for a password (text is not echoed). It supports the full validation pipeline (RegEx → Validator) and automatic retry logic.- Specialized
get_*Wrappers — Dynamically generated functions (e.g.,get_int,get_email,get_time) that pre-package specific RegEx patterns and converters for immediate use. - Validator Factories — Reusable functions that generate validators (
is_in_range,is_list_of, etc.), allowing users to easily compose complex input validation logic. - Validator Composition — Combine multiple rules using
all_ofandany_offactories for complex, multi-stage data validation. - Predefined Patterns — A set of included RegEx constants (
INT,FLOAT,EMAIL,TIME_24H) available for immediate use in input validation.
Parsing, Splitting, and Converting
extract_match— Extracts regex groups from a string and converts them to target types in a single step.split— Splits strings by delimiter or regex with optional automatic type conversion of elements.safe_int/safe_float— Fault-tolerant converters that returnNoneon failure instead of raising exceptions.
Menus & Selection
menu— Generates a numbered console menu from a dictionary and returns the selected key (int).get_choice— Selects a string from a list of options. It is case-insensitive by default and includes auto-retry logic.yes_no— A standard confirmation prompt returning abool(Truefor yes,Falsefor no).
Terminal Control & Dynamic Display
- Cursor Control — A comprehensive suite of functions (
move_to,move_to_column,move_up/down, etc.) for building dynamic interfaces like spinners or progress bars. - Cross-Platform Reliability — Includes a built-in WinAPI fallback mechanism, ensuring cursor movements and screen clearing work correctly on Windows even if ANSI is not natively supported.
set_title— Sets the console window title.
Styling & Formatting
stylize— Apply multiple styles to text.rgb- Convert RGB (tuple or hex) into ANSI code for this color- Style Helpers — Ready-made, semantic functions (
error,success,warning,info) for applying standard color schemes instantly. - Style Constants — Ready-to-use constants for foregrounds (
RED), backgrounds (BG_BLUE), and attributes (BOLD,UNDERLINE). - Pretty Printing —
print_tableandprint_iterablefor formatting lists and tabular data with custom patterns. print_header— Displays a centered title with decorative separators.progress_bar— Adaptive context manager with automatic title wrapping and screen refresh rate limit.
Utilities & Error Safety
safe_run— A context manager that handles exceptions gracefully and ensures a clean exit onKeyboardInterrupt(Ctrl+C).try_until_ok— A universal wrapper to repeatedly execute unstable functions (e.g., network calls) until success.on_interrupt— Global interrupt control: use it to define a consistent behavior for Ctrl+C across your entire application.
Requires Python 3.10+ and has zero external dependencies.
Installation
pip install --upgrade cli_tools_by_oleksa
What's new in 2.4.0?
- Added new patterns:
HEX_INT,SCIENTIFIC(scientific notation),SHORT_HEX_RGB, andANY_HEX_RGB. - Enhanced numeric patterns:
FLOATandNUMBERnow support "lazy" syntax (e.g.,.5and5.). - Expanded
NUMBERpattern: Added full support for scientific notation (e.g.,1.2e-3). - Improved semantics: Renamed
USERNAMEpattern toIDENTIFIER.USERNAMEandVARremain available as aliases. - Fixed documentation: Corrected the
progress_bar.pyusage example.
What's new in 2.3.0?
- Added tests for core input functions
- Removed
DATE_DMYandDATE_YMDpatterns andget_date_dmyandget_date_ymdfunctions because of unsafety - Added
is_dateandis_timevalidators to safe check if strings match date / time format - Fixed issues with import
get_password
What's new in 2.2.0?
- Added
on_interruptfunction to set defaultKeyboardInterrupthandling for all functions withon_keyboard_interruptparameter. - Added simple progress bar context manager (look for example 3)
- The source code is available on GitHub
What's new in 2.1.1?
- Added
on_keyboard_interruptparameter to customizeKeyboardInterrupthandling toget_input,try_until_ok,safe_run. Also, by default, a newline character is now printed before callingsys.exit(0). - Fixed an issue with style support on Linux
What's new in 2.1?
- Functions for combining multiple validators:
any_ofandall_ofandnot_in_listvalidator get_inputand all child functions now haveretry=Trueby default
Examples
1. Basic Functions
from cli_tools import print_header, get_input, get_int, get_number
from cli_tools.validators import is_in_range
print_header('Basic CLI Functions')
name = get_input(
'What is your name? ',
# DEMO: converter can modify input before return.
# NOTE: For simple string ops like this, get_input().strip().capitalize() is cleaner.
converter=lambda s: s.strip().capitalize(),
# returns 'Anonymous' if the user presses Enter.
default='Anonymous'
)
# get_int ensures integer format.
age = get_int(
'How old are you? ',
# validator checks if the number is within the range [10, 100].
validator=is_in_range(10, 100),
# message to show if input is incorrect
if_invalid='Age must be an integer between 10 and 100.',
)
# get_number passes int and float numbers (returns always float).
num = get_number(
'What is your favourite number? ',
if_invalid='Please, enter a number.'
)
print(f"Nice to meet you, {name}! You are {age} years old and your favorite number is {num}. I like it!")
2. Menu, Choices and Format Output
from cli_tools import print_header, get_choice, menu, yes_no, print_iterable, print_table
# Defines the main menu options for the menu() function.
# Keys (int) are the expected user input; values (str) are the displayed option text.
main_menu = {
1: 'New game',
2: 'Best players',
3: 'Show results',
0: 'Quit',
}
# Simple list of players for print_iterable demonstration.
players = [
'You',
'Not you',
'Who?'
]
# Dictionary of results for print_table demonstration.
# The items (key-value pairs) will be unpacked into table rows.
results = {
'Easy': 2308,
'Medium': 1841,
'Hard': 1550,
}
def play():
"""Starts the game flow, handling difficulty selection."""
modes = ['Easy', 'Medium', 'Hard']
# get_choice() validates user input against a list of options (modes).
# show=True displays the options automatically before prompting.
mode = get_choice(
options = modes,
prompt = 'Choose difficulty: ',
if_invalid = 'Please, enter a valid mode name.',
show = True
).lower()
print(f'Some game logic for {mode} mode...')
def main():
"""Main loop for the CLI application, controlling flow via the menu."""
# print_header() adds visual separation and a centered title to the console.
print_header('Choices And Format Demo', char='=', space=3)
while True:
# menu() handles display, validates input against the dictionary keys, and returns the selected key (int).
match menu(main_menu, 'What would you like to do? '):
case 1:
play()
case 2:
# print_iterable() outputs a one-dimensional list with custom formatting.
print_iterable(
players,
start='\nBest players:\n',
item_pattern='- {}'
)
case 3:
# print_table() iterates over results.items() and unpacks each pair into the row_pattern.
print_table(
results.items(),
# Pattern uses format specifiers for alignment:
# {:<6} for left-aligned string, {:>8} for right-aligned integer
row_pattern = '{:<6}{:>8}',
start = '\nMode\tResults\n'
)
case 0:
# yes_no() prompts for confirmation and returns a boolean (True for 'yes', False for 'no').
if yes_no('Exit? [y/N]: '):
break
print('Bye!')
if __name__ == '__main__':
main()
3. Progress Bar
from time import sleep
from cli_tools import progress_bar
with progress_bar(1000, 'Processing...', length=20) as bar:
for i in range(500):
sleep(0.01) # do your job here
bar.update(steps=2)
print('Done')
4. Project Example: Simple Calculator
import re
from cli_tools import (print_header, get_input,
extract_match) # extract list of all matches, optionally convert every match
from cli_tools.patterns import NUMBER # Ready-made pattern: accepts both int and float
# Pattern <number> <operator> <number>, spaces ignored
simple_expr_pattern = re.compile(fr' *({NUMBER.pattern}) *([+\-*/^]) *({NUMBER.pattern}) *')
# Accepts either numbers or operators. Converts numbers to float
converter = lambda x: float(x) if x not in '+-*/^' else x
print_header('| Simple calculator |', '—')
print('Supports simple expressions in format <number> <operator> <number>. Press Ctrl+C to exit.')
while True:
# Guaranteed to pass only input that matches the pattern
expr = get_input('> ', pattern=simple_expr_pattern, if_invalid='Wrong format!')
# Unpack the input, immediately converting the numbers
left, operator, right = extract_match(expr, simple_expr_pattern, converter=converter)
match operator:
case '+':
print(left+right)
case '-':
print(left-right)
case '*':
print(left*right)
case '/':
if right == 0:
print('Division by zero is not allowed!')
continue
print(left/right)
case '^':
print(left**right)
case _:
print('Wrong operator!')
5. Styles
import cli_tools.styles as s
from cli_tools.styles import (
rgb, # Converts full-range RGB (HEX or tuple) into an ANSI color code.
stylize # Applies one or more style codes (color, background, format) to text.
)
# These functions apply a default foreground color and return the styled string.
error_msg = s.error('Error message')
warning_msg = s.warning('Warning message')
success_msg = s.success('Success message')
info_msg = s.info('Info message')
print(error_msg)
print(warning_msg)
print(success_msg)
print(info_msg)
# Define custom color using an RGB tuple (24-bit color).
tuple_yellow = (240, 240, 100)
# Apply the tuple color as foreground (text).
print(
stylize('Yellow text', rgb(tuple_yellow))
)
# Apply the tuple color as background (using is_bg=True) and combine with BLACK foreground (text).
print(
stylize('Yellow background', rgb(tuple_yellow, is_bg=True), s.BLACK)
)
# Define custom colors using the rgb function with HEX codes.
# The first color is foreground (text), the second specifies the background (bg#).
gray = rgb('#dddddd')
darkblue_bg = rgb('bg#0a0a88')
# wrap() applies multiple styles (foreground, background, and formatting constants).
print(
stylize('Styled message', gray, darkblue_bg, s.ITALIC, s.UNDERLINE)
)
6. Terminal
from time import sleep
from cli_tools import terminal as t
t.clear_screen()
t.home_cursor()
t.set_title('Title')
print('Hello!')
print('This text will disappear in a second.')
sleep(1)
t.move_up(1)
t.clear_line()
print('Simple progress bar example: ')
pattern = '[{:<10}]'
for i in range(1, 11):
t.clear_line()
t.move_to_column(1)
print(pattern.format('='*i), end='', flush=True)
sleep(0.2)
print()
print('Simple spinner example: ', end='')
for i in range(4):
for status in ['|', '/', '—', '\\']:
print(status, end='', flush=True)
sleep(0.1)
t.move_backward(1)
print('Done')
7. Error Handling
import random, time
from cli_tools.exceptions import (CLIError, # Base class for all raised errors
APIError, # Error that raised when you pass invalid data to a function.
ValidationError, # Data not validated
ConversionError) # The transferred converter caused an error
from cli_tools import safe_run, try_until_ok, print_header
print_header('Safe Execution Demo')
with safe_run(debug=False, exit_on_error=False):
print("Press Ctrl+C to test graceful exit, or wait for the error...")
for i in range(3, 0, -1):
print(f"Crashing in {i}...")
time.sleep(1)
raise CLIError("Something went wrong inside the app!")
print("App still working.")
###
print_header('Retry Logic Demo')
def unstable_network_request():
"""Simulates a connection that fails 70% of the time."""
if random.random() < 0.7:
raise ConnectionError("Connection timed out")
return "200 OK"
print("Attempting to connect to server...")
status = try_until_ok(
unstable_network_request,
exceptions=ConnectionError,
on_exception="Connection failed. Retrying..."
)
print(f"Success! Server response: {status}")
Roadmap
Next Steps (Technical & Infrastructure):
- Detailed documentation
Next Steps (Feature Development):
- Progress bar and Spinners (High-level API)
- Enhanced Validator Functionality:
- Expand built-in library with more specialized validators
Metadata
Release files for cli-tools-by-oleksa 2.4.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| cli_tools_by_oleksa-2.4.0.tar.gz | 24.1 kB | With provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| cli_tools_by_oleksa-2.4.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 52.0 kB
Release files / cli_tools_by_oleksa-2.4.0.tar.gz
| Download URL | cli_tools_by_oleksa-2.4.0.tar.gz |
|---|---|
| Size | 24.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
e5ca8546348b3debd3d60f5f2e6b13eee7cb9c25c9cb0ae69e44bf999731bb06
|
|
BLAKE2b-256 checksum How to use checksums |
a2aceacfeacd15950303d8d04c0a0229b3ce71d793cd5a5c25944fe223499496
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Mar 9, 2026.
Transparency logRelease files / cli_tools_by_oleksa-2.4.0-py3-none-any.whl
| Download URL | cli_tools_by_oleksa-2.4.0-py3-none-any.whl |
|---|---|
| Size | 27.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
ca5a64ebe1deb3114ab60cf48c4d109aa13f53a085465e5c23fe406d17059e46
|
|
BLAKE2b-256 checksum How to use checksums |
1dd549e19622fd1fbd059366ee068ff2d4a866ecab83fc282451d23cf102d74e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Mar 9, 2026.
Transparency logSecurity for cli-tools-by-oleksa 2.4.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
All 2 files in this release have been uploaded with provenance metadata.
Attestation for cli_tools_by_oleksa-2.4.0.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Mar 9, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.