feat: add Docling document converter#3066
Conversation
Signed-off-by: Panos Vagenas <pva@zurich.ibm.com>
Signed-off-by: Panos Vagenas <pva@zurich.ibm.com>
|
|
julian-risch
left a comment
There was a problem hiding this comment.
Thank you so much @vagenas for the great and super quick collaboration!
I added backward compatibility for the old import paths and a test for that. In addition, I slightly changed how serialization works. There was a small issue. I ensure backward compatibility for deserialization too.
I will transfer the open GitHub issues from the docling repository and will comment on the open PRs there too. Feel free to archive the docling repo and point to haystack-core-integrations.
Part of #3097
Proposed Changes:
Adding Docling document converter.
How did you test it?
Notes for the reviewer
from docling_haystack.converter import DoclingConverter.from haystack_integrations.components.converters.docling import DoclingConverter=> Presumably the PyPI package will become a shim to accordingly redirect the exposed packages for backwards compatibility?
Checklist
fix:,feat:,build:,chore:,ci:,docs:,style:,refactor:,perf:,test:.