Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update docling and docling-parse dependencies to use docling-parse >= v3 #436

Open
courtneypacheco opened this issue Dec 10, 2024 · 1 comment

Comments

@courtneypacheco
Copy link
Contributor

Background / Context

docling-parse v3.0.0 was released on 9 December 2024 and it contained a large number of breaking changes, namely to its syntax. These syntax changes broke SDG (and therefore some InstructLab logic in general). See relevant issue here: instructlab/instructlab#2765

Furthermore, it appears that the main docling package was updated ~7 hours afterward to use the new docling-parse v3.0.0 release. We can see that the v2.10.0 docling release makes these changes effective.

Current Workaround

To temporarily workaround this issue, I created #434 to force docling-parse>=2.0.0,<3.0.0, and to force docling<=2.8.3. In other words, this change will temporarily ensure we don't use docling-parse's v3's breaking changes either through directly importing docling-parse or indirectly through docling.

Goals

We eventually want to migrate to the newer v3 syntax so that we can utilize newer versions and newer functionality, as more is added.

@dmartinol
Copy link

One more point in favor of upgrading the docling dependency.
We want to integrate the docling-haystack package in instructlab to leverage its Haystack components within the RAG pipelines. However, a dependency conflict is currently preventing us from proceeding with the integration:

    instructlab-sdg 0.6.2 depends on docling<=2.8.3 and >=2.4.2
    docling-haystack 0.1.1 depends on docling<3.0.0 and >=2.9.0

requirements.txt

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants