safe-builtins 0.2.0

Python builtins wrapped with rust-like Result

safe-builtins

Python's builtins wrapped inside rust-like Result type

[!WARNING] This library doesn't depend on original python Result library, instead it depends on my fork which is up-to-date with the one maintained by @montasaurus, as per this README is written, it's still labeled as experimental and unstable, so use this library at your own risk!

Installation

Install the package from PyPi using uv:

uv add safe-builtins

or install it directly from master branch on github for the latest update

uv add git+https://github.com/kkkfasy/safe-builtins

Summary

This library provides safe wrappers for Python's built-in functions using the Result type from the montasaurus/result library. Instead of raising exceptions, these functions return a Result that either contains the successful value (Ok) or the caught exception (Err).

This approach eliminates the need for extensive try/except blocks and makes error handling more explicit and functional in style.

API

Creating Safe Builtins

Each Python builtin function has a corresponding safe_ prefixed version:

from safe_builtins import safe_int, safe_open, safe_len

# Instead of raising ValueError, returns Err(ValueError(...))
result = safe_int("not a number")
# Returns Ok(42)
result = safe_int("42")

# Instead of raising FileNotFoundError, returns Err(FileNotFoundError(...))
result = safe_open("non_existent_file.txt")
# Returns Ok(file_object)
result = safe_open("existing_file.txt")

# Instead of raising TypeError, returns Err(TypeError(...))
result = safe_len(123)  # integers don't have length
# Returns Ok(3)
result = safe_len([1, 2, 3])

Checking Results

You can check if a result is successful or contains an error:

from safe_builtins import safe_int
from result import Ok, Err

result = safe_int("42")

# Using isinstance checks
if isinstance(result, Ok):
    value = result.unwrap()
    print(f"Got value: {value}")
elif isinstance(result, Err):
    error = result.unwrap_err()
    print(f"Got error: {error}")

# Using is_ok/is_err methods
if result.is_ok():
    value = result.unwrap()
    print(f"Got value: {value}")
elif result.is_err():
    error = result.unwrap_err()
    print(f"Got error: {error}")

Accessing Values

from safe_builtins import safe_int
from result import Ok

result = safe_int("42")

# Get the value or raise an exception if it's an error
value = result.unwrap()

# Get the value or a default if it's an error
value = result.unwrap_or(0)

# Get the value or compute a default if it's an error
value = result.unwrap_or_else(lambda err: 0)

# Convert to optional (None if error)
optional_value = result.ok()

Chaining Operations

from safe_builtins import safe_int, safe_open
from result import Ok

# Chain operations that might fail
result = (
    safe_int("42")
    .map(lambda x: x * 2)  # Only executed if the result is Ok
    .map_err(lambda err: f"Error processing number: {err}")  # Only executed if the result is Err
)

# Process file contents safely
result = (
    safe_open("data.txt")
    .map(lambda f: f.read())  # Read file contents
    .map(lambda contents: contents.splitlines())  # Split into lines
    .map(lambda lines: len(lines))  # Count lines
)

Available Safe Builtins

The following Python builtins are available as safe versions:

• safe_abs
• safe_aiter
• safe_all
• safe_any
• safe_anext
• safe_ascii
• safe_bin
• safe_breakpoint
• safe_bytearray
• safe_bytes
• safe_chr
• safe_compile
• safe_complex
• safe_delattr
• safe_dict
• safe_dir
• safe_divmod
• safe_enumerate
• safe_eval
• safe_exec
• safe_filter
• safe_float
• safe_format
• safe_frozenset
• safe_getattr
• safe_hash
• safe_hex
• safe_id
• safe_input
• safe_int
• safe_isinstance
• safe_issubclass
• safe_iter
• safe_len
• safe_list
• safe_map
• safe_max
• safe_memoryview
• safe_min
• safe_next
• safe_object
• safe_oct
• safe_open
• safe_ord
• safe_pow
• safe_range
• safe_reversed
• safe_round
• safe_set
• safe_setattr
• safe_slice
• safe_sorted
• safe_staticmethod
• safe_str
• safe_sum
• safe_super
• safe_tuple
• safe_type
• safe_vars
• safe_zip
• safe_import
• safe_bool

Related Projects

• result - The underlying Result type implementation
• returns - Another functional approach to error handling in Python

License

MIT