Aludel.DocumentConverter (aludel v0.1.18)

Copy Markdown View Source

Converts documents between formats for LLM consumption.

Currently supports:

  • PDF → PNG (first page only, 150 DPI)

Usage

PDF-to-image conversion is used for:

  • OpenAI: Chat API only supports images (PDFs work in Assistants API)
  • Ollama: Vision models require image formats

Native PDF support:

  • Anthropic Claude 4.5+: Supports PDFs via document API

Requirements

Requires ImageMagick v7+ to be installed:

  • macOS: brew install imagemagick
  • Ubuntu/Debian: apt-get install imagemagick
  • Docker: Install in runtime image

Configuration

The conversion adapter can be configured in config files:

config :aludel, :document_converter,
  adapter: Aludel.Interfaces.DocumentConverter.Adapters.Imagemagick

For testing, use a stub adapter.

Summary

Types

convert_result()
document()

Functions

pdf_to_image(doc)

Converts a PDF document to PNG format.

Types

convert_result()

@type convert_result() :: {:ok, document()} | {:error, term()}

document()

@type document() :: %{data: binary(), content_type: String.t()}

Functions

pdf_to_image(doc)

@spec pdf_to_image(document()) :: convert_result()

Converts a PDF document to PNG format.

Only converts the first page at 150 DPI for optimal text readability. Creates temporary files for conversion and cleans them up afterwards.

Examples

iex> pdf_doc = %{data: pdf_binary, content_type: "application/pdf"}
iex> {:ok, png_doc} = DocumentConverter.pdf_to_image(pdf_doc)
iex> png_doc.content_type
"image/png"

iex> image_doc = %{data: png_binary, content_type: "image/png"}
iex> {:ok, ^image_doc} = DocumentConverter.pdf_to_image(image_doc)
:ok