feat: NV-ingest image extraction and authentication support (#7795)

* nv-ingest auth workaround

* nv-ingest local dependency

* workaround for RestClient

* use forked nv-ingest-client with auth/url improvements

* update to nv-ingest-client 20250422

* image extraction

* fix splitter options

* improving defaults

* review feedback fixes

* [autofix.ci] apply automated fixes

* ruff fixes

* require base url

* [autofix.ci] apply automated fixes

* add check for empty segment

* Remove html from list of supported

* Use hardcoded list of supported types

* prayer to lint gods

---------

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Jordan Frazier <jordan.frazier@datastax.com>
Co-authored-by: Jordan Frazier <122494242+jordanrfrazier@users.noreply.github.com>
This commit is contained in:
Jeffrey Carpenter 2025-04-28 21:38:57 -07:00 • committed by GitHub
commit 4610bc3eb4
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 4506 additions and 4379 deletions

View file

@ -208,12 +208,8 @@ clickhouse-connect = [
]
nv-ingest = [
# nv-ingest-client 2025.2.7.dev0 does not correctly install its
# dependencies, so we need to install some manually.
"nv-ingest-client==2025.2.7.dev0",
"python-pptx==0.6.23",
"pymilvus[bulk_writer,model]==2.5.0",
"llama-index-embeddings-nvidia==0.1.5",
"nv-ingest-api==2025.4.22.dev20250422",
"nv-ingest-client==2025.4.22.dev20250422",
]
postgresql = [
@ -320,4 +316,4 @@ ignore_missing_imports = true
[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"
build-backend = "hatchling.build"