JiriOndrusek opened a new pull request, #3016:
URL: https://github.com/apache/camel-kamelets/pull/3016
Fixes #3015.
Requires Camel 4.23.0 with https://github.com/apache/camel/pull/26190
Four Kamelets forming the ingestion half of a RAG setup, backed by the new
`camel-langchain4j-ingest` component (CAMEL-24621, apache/camel#26190, Camel
4.23):
- **langchain4j-ingest-sink** — splits, batch-embeds and stores documents
in a LangChain4j `EmbeddingStore`, each segment stamped with the pipeline name
and document id;
optional dedup by id and a document size cap
- **tika-extract-text-action** / **docling-convert-action** — generic
document→text/markdown parse steps (useful beyond ingestion) that capture the
document id *before*
the parse, so a crafted document cannot forge its own identity
- **langchain4j-ingest-file-source** — a directory source with
knowledge-base semantics: files are read, not consumed; edits are re-delivered;
parser actions receive the
raw bytes
Citrus tests cover each Kamelet, the source → action → sink compositions
(including a binary PDF) and the file source's register semantics.
⚠️ **The integration-tests check is red on purpose**: the tests need
Camel 4.23+ and fail on the catalog's current 4.22 — the red build is the gate
that keeps this PR from
merging into a release line whose Camel cannot run these Kamelets. It
turns green with the catalog's move to Camel 4.23. Verified locally with
`-Dcamel.version=4.23.0-SNAPSHOT` (6/6 passing).
_PR prepared by Claude Code on behalf of @JiriOndrusek._
🤖 Generated with [Claude Code](https://claude.com/claude-code)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
