Skip to content

errors.md

Latest commit

 

History

History
126 lines (96 loc) · 4.98 KB

errors.md

File metadata and controls

126 lines (96 loc) · 4.98 KB

⬅ Back to Table of Contents ⬅ Back to Math

Errors and Exceptions

This page documents the error and exception model in Pythonic, including the full error hierarchy, common error types, and idiomatic error handling.

Error Hierarchy

All Pythonic exceptions inherit from pythonic::PythonicError, allowing you to catch all library errors with a single catch block, while still supporting granular error handling.

Exception Type Description Example Usage
PythonicError Base class for all Pythonic errors catch (const PythonicError& e)
PythonicWarning Base for warnings (not errors)
PythonicTypeError Wrong type operation catch (const PythonicTypeError& e)
PythonicValueError Right type, wrong value
PythonicIndexError Sequence index out of bounds
PythonicKeyError Mapping key not found
PythonicOverflowError Numeric overflow
PythonicZeroDivisionError Division or modulo by zero
PythonicFileError File operation failed
PythonicAttributeError Attribute reference/assignment failed
PythonicGraphError Graph-specific operation failure
PythonicIterationError Iterator exhausted or used incorrectly
PythonicStopIteration Signals iterator exhaustion (like Python)
PythonicRuntimeError General runtime error
PythonicNotImplementedError Feature not yet implemented

Common Error Types

Type Errors

Raised when an operation receives a value of the wrong type.

  • Example: "hello" * "world", Int("not a number")

Value Errors

Raised when a value is inappropriate (right type, wrong content).

  • Example: range(0, 10, 0), sqrt(-1)

Index Errors

Raised when a sequence index is out of range.

  • Example: list[100] when len(list) == 5

Key Errors

Raised when a mapping key is not found.

  • Example: dict["nonexistent_key"]

Arithmetic Errors

  • PythonicOverflowError: Raised when numeric operation overflows and promotion is exhausted.
  • PythonicZeroDivisionError: Raised on division or modulo by zero.

File/IO Errors

Raised when a file operation fails (not found, cannot open, not open).

Attribute Errors

Raised when an attribute reference or assignment fails.

  • Example: Calling .upper() on an int

Graph Errors

Raised for graph-specific operation failures (invalid node, edge not found, cycle detected).

Iterator Errors

  • PythonicIterationError: Raised when an iterator is exhausted or used incorrectly.
  • PythonicStopIteration: Signals iterator exhaustion (like Python's StopIteration).

Runtime & Not Implemented Errors

  • PythonicRuntimeError: General runtime errors.
  • PythonicNotImplementedError: Feature not yet implemented.

Idiomatic Error Handling

You can catch specific errors or all Pythonic errors:

try {
    // Some operation that may fail
    var result = Int("not a number");
} catch (const pythonic::PythonicTypeError& e) {
    print("Type error:", e.what());
} catch (const pythonic::PythonicError& e) {
    print("Any pythonic error:", e.what());
}

Example: Arithmetic Error

try {
    var x = INT_MAX;
    var y = 1;
    print(add(x, y)); // May throw PythonicOverflowError
} catch (const pythonic::PythonicOverflowError& e) {
    print("Overflow:", e.what());
}

Example: Index Error

try {
    var lst = list(1,2,3);
    print(lst[10]); // Throws PythonicIndexError
} catch (const pythonic::PythonicIndexError& e) {
    print("Index error:", e.what());
}

Notes

  • All error messages are prefixed with pythonic: for clarity.
  • Most errors support optional source location tracking for debugging (in debug builds).
  • Use the PYTHONIC_THROW macro to throw with source location info.
  • Warnings (PythonicWarning) can be caught separately from errors.
  • The error hierarchy allows both broad and fine-grained exception handling.

Next Check