deepset-ai/haystack · error · TypeError

MockTextEmbedder expects a string as an input. In case you w

Error message

MockTextEmbedder expects a string as an input. In case you want to embed a list of Documents, please use the MockDocumentEmbedder.

What it means

MockTextEmbedder.run accepts only a single string. Passing anything else (a list, a Document, None) raises this TypeError, which directs you to MockDocumentEmbedder for Document-list inputs. The check keeps the mock's interface identical to the real OpenAITextEmbedder.

Source

Thrown at haystack/components/embedders/mock_text_embedder.py:140

        if self.embedding is not None:
            return list(self.embedding)
        return _deterministic_embedding(text, self.dimension)

    @component.output_types(embedding=list[float], meta=dict[str, Any])
    def run(self, text: str) -> dict[str, Any]:
        """
        Return a deterministic embedding for the input text without calling any API.

        :param text: The text to embed.
        :returns: A dictionary with the following keys:
            - `embedding`: The embedding of the input text.
            - `meta`: Metadata about the (mock) model.
        :raises TypeError: If `text` is not a string.
        """
        self.warm_up()

        if not isinstance(text, str):
            raise TypeError(
                "MockTextEmbedder expects a string as an input. "
                "In case you want to embed a list of Documents, please use the MockDocumentEmbedder."
            )

        text_to_embed = self.prefix + text + self.suffix
        meta: dict[str, Any] = {"model": self.model, "usage": _estimate_usage([text_to_embed])}
        meta.update(self.meta)
        return {"embedding": self._embed(text_to_embed), "meta": meta}

    @component.output_types(embedding=list[float], meta=dict[str, Any])
    async def run_async(self, text: str) -> dict[str, Any]:
        """
        Asynchronously return a deterministic embedding for the input text without calling any API.

        :param text: The text to embed.
        :returns: A dictionary with the following keys:
            - `embedding`: The embedding of the input text.
            - `meta`: Metadata about the (mock) model.

View on GitHub (pinned to e318778c9b)

Solutions

1. Pass a single string: `embedder.run("text to embed")`
2. Use MockDocumentEmbedder with a list of Documents if you need document-level embedding
3. Unwrap Documents first: `run(document.content)` when you have a single Document

Example fix

// before
result = embedder.run([Document(content="hello")])
// after
result = embedder.run("hello")

assert isinstance(text, str), f"MockTextEmbedder.run expects str, got {type(text)}"
result = embedder.run(text)

def is_embeddable_text(value) -> bool:
    return isinstance(value, str)

try:
    result = embedder.run(text)
except TypeError:
    if isinstance(text, list) and text and isinstance(text[0], Document):
        result = document_embedder.run(text)
    else:
        raise

When it happens

Trigger: `MockTextEmbedder().run(["text1", "text2"])`, `.run(Document(content="x"))`, or `.run(None)` — any non-str `text` argument.

Common situations: Wiring a DocumentEmbedder-style input into a TextEmbedder in a pipeline; batching code that passes a list where one string is expected; tests copied from the document embedder.

Related errors

• MockDocumentEmbedder expects a list of Documents as input.In
• 'response_fn' must return a string or ChatMessage, got {type
• Unsupported source type {type(source)}
• meta must be either None, a dictionary or a list of dictiona
• Pass either 'embedding' or 'embedding_fn', not both.

AI-assisted analysis of deepset-ai/haystack@e318778c9b (2026-08-30). Data as JSON: /api/errors/51dc856614af3812. Report an issue: GitHub.