apache/seatunnel · error · ImportError

openai package required for AI_PROVIDER=openai. Install it…

Error message

openai package required for AI_PROVIDER=openai. Install it: pip install openai

What it means

Raised by OpenAIProvider.__init__ when the optional openai Python package is absent while AI_PROVIDER=openai is configured. The provider lazily imports the dependency, so the config selection is the input at fault, not missing core functionality.

Solutions

Run: pip install openai in the environment running the CLI
Install via the CLI's LLM extra if available
Verify with 'python -c "import openai"' using the same interpreter

Example fix

// before
export AI_PROVIDER=openai
seatunnel-cli ...   # ImportError
// after
pip install openai
export AI_PROVIDER=openai

Defensive patterns

Strategy: try-catch

Validate before calling

python -c "import openai" || pip install openai

Try / catch

try:
    provider = OpenAIProvider()
except ImportError as e:
    print(f'Missing LLM dependency: {e}. Run: pip install openai')
    sys.exit(2)

Prevention

Install the openai package in the CLI's interpreter
Pin openai version compatible with the CLI
Run an import smoke test in CI
Install the matching extra when switching AI_PROVIDER

When it happens

Trigger: AI_PROVIDER=openai selects OpenAIProvider and 'import openai' fails in __init__ because the package is not installed in the environment.

Common situations: Fresh venv/container without the openai package; switched AI_PROVIDER from another provider; very old CLI where openai was not a dependency.

Understand the failure class

Background: "X is not installed. Please install it with pip install Y": missing optional dependency errors — ImportError/ValueError raised when a library's optional extra was never installed — this error's family across 22 libraries.

Related errors

AI-assisted analysis of apache/seatunnel@cf67b549a7 (2026-09-10). Data as JSON: /api/errors/9b7fe49b12215758. Report an issue: GitHub.

Appendix: source

Thrown at seatunnel-cli/seatunnel_cli/llm_provider.py:705

            stop_reason = "tool_use"

        return {
            "output": {"message": {"role": "assistant", "content": content}},
            "stopReason": stop_reason,
        }



# ─── OpenAI Provider ───

class OpenAIProvider(LLMProvider):
    """OpenAI Chat Completions API provider."""

    def __init__(self):
        try:
            import openai
        except ImportError:
            raise ImportError(
                "openai package required for AI_PROVIDER=openai. "
                "Install it: pip install openai"
            )

        api_key = os.environ.get("OPENAI_API_KEY")
        base_url = os.environ.get("OPENAI_BASE_URL")  # Supports compatible APIs
        if not api_key:
            raise ValueError("OPENAI_API_KEY environment variable is required for AI_PROVIDER=openai")

        client_kwargs = {"api_key": api_key}
        if base_url:
            client_kwargs["base_url"] = base_url
        self._client = openai.OpenAI(**client_kwargs)
        self._model_id = os.environ.get("OPENAI_MODEL", "gpt-4o")
        self._fast_model_id = os.environ.get("OPENAI_SMALL_FAST_MODEL", "gpt-4o-mini")
        self._echo_reasoning_content = _env_bool("OPENAI_ECHO_REASONING_CONTENT", True)

    @property

View on GitHub (pinned to cf67b549a7)