docs/how_to/extraction_long_text/

[giscus[bot]]

docs/how_to/extraction_long_text/

When working with files, like PDFs, you're likely to encounter text that exceeds your language model's context window. To process this text, consider these strategies:

https://python.langchain.com/docs/how_to/extraction_long_text/

[alexminza]

Pydantic schema method from this example has issues, described on GitHub #24225

[alexminza]

How to specify enum values for a Pydantic schema field?

The following pseudo-code leads to TypeError: list indices must be integers or slices, not str with ChatGoogleGenerativeAI:

class TestAttribute(BaseModel):
    type: Literal["attribute1", "attribute2", "attribute3", "attribute4"]

class TestExtract(BaseModel):
    test_attributes: list[TestAttribute]

[sanjeed5]

Great tutorial. Thanks!

There could be a combined strategy where brute force is applied, but if required, relevant context from other areas are added. For example in PDFs where some section asks to refer to another section.

Also, other splitting methods could also be referenced in the tutorial.