error-translator-cli-v2 1.1.3

A CLI tool that explains Python errors in simple human language.

Error Translator

Translate raw Python tracebacks into clear explanations and actionable fixes. Built for local-first developer workflows: CLI, import-hook auto mode, Python API, and FastAPI.

✨ Key Features

• CLI-first workflow: run a script, paste an error, or pipe logs directly into explain-error.
• Auto mode: import error_translator.auto installs a custom sys.excepthook for unhandled crashes.
• HTTP API: expose the same engine through FastAPI for service integrations.
• Rules engine: regex-based matching from rules.json for deterministic, offline translations.
• AST analysis hooks: error-type-specific handlers (via registry) can add deeper typo/context insight.

Installation

Install from PyPI

pip install error-translator-cli-v2

Check if the CLI is working:

explain-error --version

Local development setup

git clone https://github.com/gourabanandad/error-translator.git
cd error-translator
python -m venv .venv
# Windows
.venv\Scripts\activate
# macOS/Linux
source .venv/bin/activate

pip install -r requirements.txt
pip install -e .
pytest

Usage

1) CLI usage

Run a Python script and translate unhandled errors:

explain-error run script.py

Translate a raw error string directly:

explain-error "NameError: name 'usr_count' is not defined"

Translate piped traceback logs:

cat error.log | explain-error

2) Python auto-import mode

import error_translator.auto

# Any unhandled exception in this process is intercepted
# and rendered as a translated, colorized explanation.

def main():
    total = "Users: " + 42

main()

3) FastAPI server usage

Start server:

uvicorn error_translator.server:app --host 127.0.0.1 --port 8000 --reload

Health check:

curl http://127.0.0.1:8000/

Translate via API:

curl -X POST http://127.0.0.1:8000/translate \
  -H "Content-Type: application/json" \
  -d "{\"traceback_setting\":\"Traceback (most recent call last):\\n  File 'app.py', line 14, in <module>\\n    total = 'Users: ' + 42\\nTypeError: can only concatenate str (not 'int') to str\"}"

Real-World Input → Output Examples

Example A: Type mismatch

Input traceback:

Traceback (most recent call last):
  File "app.py", line 14, in <module>
    total = "Users: " + 42
TypeError: can only concatenate str (not "int") to str

Translated output (CLI render):

Error Detected:
TypeError: can only concatenate str (not "int") to str

Location: app.py (Line 14)

Explanation:
You are trying to add a string to an int, which Python cannot do.

Suggested Fix:
Convert the int to a string first using str() before concatenating.

Example B: Missing variable typo

Input error line:

NameError: name 'usr_count' is not defined

Translated output:

Explanation:
You tried to use a variable or function named 'usr_count', but Python doesn't recognize it.

Suggested Fix:
Check if 'usr_count' is spelled correctly, or ensure you defined/imported it before using it.

Architecture

Error Translator uses one core engine with multiple entry points.

1. Input ingestion: traceback text from CLI, API, or auto-hook.
2. Rule matching: core.py checks the last error line against compiled regex rules from rules.json.
3. Context extraction: file path, line number, and source line are extracted when available.
4. Optional AST insight: for selected error types, handlers from ast_handlers.py can add extra hints.
5. Structured output: returns a stable dictionary (explanation, fix, file, line, code, matched_error, optional ast_insight).

How It Works Internally

• load_rules() reads rules.json once and caches it.
• compiled_rules() compiles regex patterns once for fast repeated matching.
• translate_error():
  • normalizes traceback input,
  • finds the terminal error line,
  • applies first matching rule,
  • formats placeholders ({0}, {1}, ...),
  • enriches with file/line/code context,
  • dispatches to AST_REGISTRY by error type when applicable.
• If no rule matches, a safe fallback explanation/fix from default is returned.

Project Structure

error-translator/
├─ error_translator/
│  ├─ core.py           # Translation pipeline, rule loading, context extraction
│  ├─ rules.json        # Regex patterns + explanations + fixes
│  ├─ ast_handlers.py   # AST handler registry and insight plugins
│  ├─ cli.py            # `explain-error` command and colorized terminal renderer
│  ├─ auto.py           # Auto mode via sys.excepthook override
│  └─ server.py         # FastAPI app and /translate endpoint
├─ tests/
│  └─ test_core.py      # Core translation and regex extraction tests
└─ docs/
   ├─ ARCHITECTURE.md
   └─ CONTRIBUTING.md

How To Add New Rules

Add a new object to error_translator/rules.json under rules:

{
  "pattern": "ValueError: invalid literal for int\\(\\) with base 10: '(.*)'",
  "explanation": "You tried to convert '{0}' to an integer, but it is not a valid whole number.",
  "fix": "Ensure '{0}' contains only digits, or parse it as float if decimals are expected."
}

Guidelines:

• Keep regex patterns specific to avoid false positives.
• Use capture groups for dynamic values and reference them with {0}, {1}, etc.
• Keep explanations beginner-friendly and fixes immediately actionable.
• Add/extend tests in tests/test_core.py for each new rule.

Contributing

Contributions are welcome, from first-time contributors to maintainers.

Typical workflow:

1. Fork and clone the repository.
2. Create a feature branch.
3. Add or improve rules/tests/docs.
4. Run pytest.
5. Open a PR with a sample traceback and expected translation.

Please see:

• docs/CONTRIBUTING.md
• docs/ARCHITECTURE.md

Maintainer

Built and maintained by Gourabananda Datta and contributors.