unclecode/crawl4ai · error · AttributeError

Setting '{name}' is deprecated. {message}

Error message

Setting '{name}' is deprecated. {message}

What it means

Raised by JsonElementExtractionStrategy.generate_schema when the deprecated provider or api_token keyword arguments are passed (non-None). The method checks each name in _GENERATE_SCHEMA_UNWANTED_PROPS against locals() and raises AttributeError with migration guidance to llm_config.

Source

Thrown at crawl4ai/extraction_strategy.py:1818

                validation retries) are added to it in-place.
            **kwargs: Additional args passed to LLM processor.

        Returns:
            dict: Generated schema following the JsonElementExtractionStrategy format.

        Raises:
            ValueError: If neither html nor url is provided.
        """
        from .utils import aperform_completion_with_backoff, preprocess_html_for_schema

        # Validate inputs
        if html is None and (url is None or (isinstance(url, list) and len(url) == 0)):
            raise ValueError("Either 'html' or 'url' must be provided")

        # Check deprecated parameters
        for name, message in JsonElementExtractionStrategy._GENERATE_SCHEMA_UNWANTED_PROPS.items():
            if locals()[name] is not None:
                raise AttributeError(f"Setting '{name}' is deprecated. {message}")

        if llm_config is None:
            llm_config = create_llm_config()

        # Save original HTML(s) before preprocessing (for validation against real HTML)
        original_htmls = []

        # Fetch HTML from URL(s) if provided
        if url is not None:
            from .async_webcrawler import AsyncWebCrawler
            from .async_configs import BrowserConfig, CrawlerRunConfig, CacheMode

            browser_config = BrowserConfig(
                headless=True,
                text_mode=True,
                light_mode=True,
            )
            crawler_config = CrawlerRunConfig(cache_mode=CacheMode.BYPASS)

View on GitHub (pinned to 7e80152142)

Solutions

1. Replace provider/api_token kwargs with llm_config=LLMConfig(provider='...', api_token='...')
2. Search your codebase for generate_schema( calls using provider= or api_token= and migrate them all

Example fix

// before
schema = await JsonElementExtractionStrategy.generate_schema(
    url=u, provider="openai/gpt-4o", api_token=tok)  # AttributeError

// after
from crawl4ai import LLMConfig
schema = await JsonElementExtractionStrategy.generate_schema(
    url=u, llm_config=LLMConfig(provider="openai/gpt-4o", api_token=tok))

import inspect

banned = {"provider", "api_token"}
sig = inspect.signature(JsonElementExtractionStrategy.generate_schema)
kwargs = {k: v for k, v in my_kwargs.items() if k not in banned and k in sig.parameters}
schema = await JsonElementExtractionStrategy.generate_schema(**kwargs)

try:
    schema = await JsonElementExtractionStrategy.generate_schema(html=h, **kwargs)
except AttributeError as e:
    if "deprecated" in str(e):
        kwargs.pop("provider", None); kwargs.pop("api_token", None)
        kwargs["llm_config"] = llm_config
        schema = await JsonElementExtractionStrategy.generate_schema(html=h, **kwargs)
    else:
        raise

When it happens

Trigger: Calling generate_schema(html=..., provider='openai/gpt-4o') or generate_schema(url=..., api_token='sk-...'). Only these two names are checked; passing them as None is tolerated.

Common situations: Old scripts from before the LLMConfig refactor that configured the schema generator per-call; copy-pasted snippets from outdated tutorials.

Related errors

• Setting '{name}' is deprecated. {self._UNWANTED_PROPS[name]}
• LLM returned an empty response
• Failed to parse schema JSON: {str(e)}
• Failed to generate schema: {str(e)}
• Failed to generate schema: no attempts succeeded

AI-assisted analysis of unclecode/crawl4ai@7e80152142 (2026-08-14). Data as JSON: /api/errors/db562db7415a2912. Report an issue: GitHub.