29 lines
717 B
Elixir
29 lines
717 B
Elixir
|
```elixir title="Elixir"
|
||
|
|
alias Kreuzberg.ExtractionConfig
|
||
|
|
|
||
|
|
# Configure OCR-specific settings
|
||
|
|
# Includes backend selection, language settings, and preprocessing options
|
||
|
|
config = %ExtractionConfig{
|
||
|
|
ocr: %{
|
||
|
|
"enabled" => true,
|
||
|
|
"backend" => "tesseract",
|
||
|
|
"language" => "eng",
|
||
|
|
"preprocessing" => true
|
||
|
|
},
|
||
|
|
chunking: %{
|
||
|
|
"max_characters" => 2000,
|
||
|
|
"overlap" => 200
|
||
|
|
},
|
||
|
|
use_cache: true,
|
||
|
|
force_ocr: false
|
||
|
|
}
|
||
|
|
|
||
|
|
{:ok, result} = Kreuzberg.extract_file("scanned_document.pdf", nil, config)
|
||
|
|
|
||
|
|
IO.puts("OCR Configuration Applied:")
|
||
|
|
IO.puts("Backend: tesseract")
|
||
|
|
IO.puts("Language: eng")
|
||
|
|
IO.puts("Content extracted: #{byte_size(result.content)} bytes")
|
||
|
|
IO.puts("Metadata: #{inspect(result.metadata)}")
|
||
|
|
```
|