admin-api-lib 3.2.0__py3-none-any.whl → 3.3.0__py3-none-any.whl

admin_api_lib-3.3.0.dist-info/METADATA

@@ -0,0 +1,129 @@
+Metadata-Version: 2.3
+Name: admin-api-lib
+Version: 3.3.0
+Summary: The admin backend is responsible for the document management. This includes deletion, upload and returning the source document.
+License: Apache-2.0
+Author: STACKIT GmbH & Co. KG
+Author-email: data-ai@stackit.cloud
+Maintainer: Andreas Klos
+Maintainer-email: andreas.klos@stackit.cloud
+Requires-Python: >=3.13,<4.0
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.13
+Requires-Dist: boto3 (>=1.38.10,<2.0.0)
+Requires-Dist: dependency-injector (>=4.46.0,<5.0.0)
+Requires-Dist: fastapi (>=0.118.0,<0.119.0)
+Requires-Dist: langchain-experimental (>=0.3.4,<0.4.0)
+Requires-Dist: langfuse (==3.6.1)
+Requires-Dist: nltk (>=3.9.2,<4.0.0)
+Requires-Dist: python-dateutil (>=2.9.0.post0,<3.0.0)
+Requires-Dist: python-multipart (>=0.0.20,<0.0.21)
+Requires-Dist: pyyaml (>=6.0.2,<7.0.0)
+Requires-Dist: rag-core-lib (>=3.3.0,<4.0.0)
+Requires-Dist: redis (>=6.0.0,<7.0.0)
+Requires-Dist: starlette (>=0.47.2,<0.49.0)
+Requires-Dist: tenacity (==9.1.2)
+Requires-Dist: tqdm (>=4.67.1,<5.0.0)
+Requires-Dist: uvicorn (>=0.37.0,<0.38.0)
+Project-URL: Homepage, https://pypi.org/project/admin-api-lib
+Project-URL: Repository, https://github.com/stackitcloud/rag-template
+Description-Content-Type: text/markdown
+
+# admin-api-lib
+
+Document lifecycle orchestration for the STACKIT RAG template. This library exposes a FastAPI-compatible admin surface that receives raw user content, coordinates extraction, summarisation, chunking, and storage, and finally hands normalized information pieces to the core RAG API.
+
+It powers the [`services/admin-backend`](https://github.com/stackitcloud/rag-template/tree/main/services/admin-backend) deployment and is the primary integration point for operators managing their document corpus.
+
+## Responsibilities
+
+1. **Ingestion** – Accept files or external sources from the admin UI or API clients.
+2. **Extraction** – Call `extractor-api-lib` to obtain normalized information pieces.
+3. **Enhancement** – Summarize and enrich content using LLMs and tracing hooks from `rag-core-lib`.
+4. **Chunking** – Split content via recursive or semantic strategies before vectorization.
+5. **Persistence** – Store raw assets in S3-compatible storage and push processed chunks to `rag-core-api`.
+6. **Status tracking** – Keep track of upload progress and expose document status endpoints backed by KeyDB/Redis.
+
+## Feature highlights
+
+- Ready-to-wire dependency-injector container with sensible defaults for S3 storage, KeyDB status tracking, and background tasks.
+- Pluggable chunkers (`recursive` vs `semantic`) and summariser implementations with shared retry/backoff controls.
+- Rich Pydantic request/response models covering uploads, non-file sources, and document status queries.
+- Thin endpoint implementations that can be swapped or extended while keeping the public API stable.
+- Structured tracing (Langfuse) and logging that mirror the behaviour of the chat backend.
+
+## Installation
+
+```bash
+pip install admin-api-lib
+```
+
+Requires Python 3.13 and `rag-core-lib`.
+
+## Module tour
+
+- `dependency_container.py` – Configures and wires dependency-injection providers. Override registrations here to customise behaviour.
+- `api_endpoints/` & `impl/api_endpoints/` – Endpoints + abstractions for file uploads, source uploads, deletions, document status, and reference retrieval.
+- `apis/` – Admin API abstractions and implementations.
+- `chunker/` & `impl/chunker/` – Abstractions + default text/semantic chunkers and chunker type selection class.
+- `extractor_api_client/` & `rag_backend_client/` – Generated OpenAPI clients to talk to the extractor and rag core API services.
+- `file_services/` & `impl/file_services/` – Abstract and default S3 interface.
+- `summarizer/` & `impl/summarizer/` – Interfaces and LangChain-based summariser that leverage shared retry logic.
+- `information_enhancer/` & `impl/information_enhancer/` – Abstractions + page and summary enhancer. Enhancers are centralized with general enhancer.
+- `impl/key_db/` – KeyDB/Redis client implementation for document status tracking.
+- `impl/mapper/` – Mapper between extractor documents and langchain documents.
+- `impl/settings/` – Configuration settings for dependency injection container components.
+- `prompt_templates/` – Default summarisation prompt shipped with the template.
+- `utils/` – Utility functions and classes.
+
+## Endpoints provided
+
+- `POST /upload_file` – Uploads user selected files
+- `POST /upload_source` - Uploads user selected sources
+- `DELETE /documents/{identification}` – Deletes a document from the system.
+- `GET /document_reference/{identification}` – Retrieves a document reference.
+- `GET /all_documents_status` – Retrieves the status of all documents.
+
+Refer to [`libs/README.md`](../README.md#2-admin-api-lib) for in-depth API documentation.
+
+## Configuration overview
+
+All settings are powered by `pydantic-settings`, so you can use environment variables or instantiate classes manually:
+
+- `S3_*` (`S3_ACCESS_KEY_ID`, `S3_SECRET_ACCESS_KEY`, `S3_ENDPOINT`, `S3_BUCKET`) – configure storage for raw uploads.
+- `DOCUMENT_EXTRACTOR_HOST` – base URL of the extractor service.
+- `RAG_API_HOST` – base URL of the rag-core API.
+- `CHUNKER_CLASS_TYPE_CHUNKER_TYPE` – choose `recursive` (default) or `semantic` chunking.
+- `CHUNKER_*` (`CHUNKER_MAX_SIZE`, `CHUNKER_OVERLAP`, `CHUNKER_BREAKPOINT_THRESHOLD_TYPE`, …) – fine-tune chunking behaviour.
+- `SUMMARIZER_MAXIMUM_INPUT_SIZE`, `SUMMARIZER_MAXIMUM_CONCURRENCY`, `SUMMARIZER_MAX_RETRIES`, etc. – tune summariser limits and retry behaviour.
+- `SOURCE_UPLOADER_TIMEOUT` – adjust how long non-file source ingestions wait before timing out.
+- `USECASE_KEYVALUE_HOST` / `USECASE_KEYVALUE_PORT` – configure the KeyDB/Redis instance that persists document status.
+
+The Helm chart forwards these values through `adminBackend.envs.*`, keeping deployments declarative. Local development can rely on `.env` as described in the repository root README.
+
+## Typical usage
+
+```python
+from admin_api_lib.main import app as perfect_admin_app
+```
+
+The admin frontend (`services/frontend` → Admin app) and automation scripts talk to these endpoints to manage the corpus. Downstream, `rag-core-api` receives the processed information pieces and stores them in the vector database.
+
+## Extending the library
+
+1. Implement a new interface (e.g., `Chunker`, `Summarizer`, `FileService`).
+2. Register it in `dependency_container.py` or override via dependency-injector in your service.
+3. Update or add API endpoints if you expose new capabilities.
+4. Cover the new behaviour with pytest-based unit tests under `libs/admin-api-lib/tests`.
+
+Because components depend on interfaces defined here, downstream services can swap behavior without modifying the public API surface.
+
+## Contributing
+
+Ensure new endpoints or adapters remain thin and defer to [`rag-core-lib`](../rag-core-lib/) for shared logic. Run `poetry run pytest` and the configured linters before opening a PR. For further instructions see the [Contributing Guide](https://github.com/stackitcloud/rag-template/blob/main/CONTRIBUTING.md).
+
+## License
+
+Licensed under the project license. See the root [`LICENSE`](https://github.com/stackitcloud/rag-template/blob/main/LICENSE) file.
+

{admin_api_lib-3.2.0.dist-info → admin_api_lib-3.3.0.dist-info}/RECORD

@@ -101,6 +101,6 @@ admin_api_lib/summarizer/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZ
 admin_api_lib/summarizer/summarizer.py,sha256=D0rkW0iZSys-68LcO1-PIkE0Faf2Grg-_9wu75Rc1OY,966
 admin_api_lib/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 admin_api_lib/utils/utils.py,sha256=eaNQ_NzUEp4hwhCU9EEsUXvbRH_ekVariF7tTsO9Sco,834
-admin_api_lib-3.2.0.dist-info/METADATA,sha256=6lj_15ckxlPhHWzUZOqYtJQqOyzVWhoK82sjRJHeCBE,1049
-admin_api_lib-3.2.0.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-admin_api_lib-3.2.0.dist-info/RECORD,,
+admin_api_lib-3.3.0.dist-info/METADATA,sha256=PuhPx4uTP5vKqivM5zJ6UBVZcu32agtHh0pHaWIjh6U,7422
+admin_api_lib-3.3.0.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+admin_api_lib-3.3.0.dist-info/RECORD,,

admin_api_lib-3.2.0.dist-info/METADATA

@@ -1,24 +0,0 @@
-Metadata-Version: 2.3
-Name: admin-api-lib
-Version: 3.2.0
-Summary: The admin backend is responsible for the document management. This includes deletion, upload and returning the source document.
-Author: STACKIT Data and AI Consulting
-Author-email: data-ai-consulting@stackit.cloud
-Requires-Python: >=3.13,<4.0
-Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.13
-Requires-Dist: boto3 (>=1.38.10,<2.0.0)
-Requires-Dist: dependency-injector (>=4.46.0,<5.0.0)
-Requires-Dist: fastapi (>=0.118.0,<0.119.0)
-Requires-Dist: langchain-experimental (>=0.3.4,<0.4.0)
-Requires-Dist: langfuse (==3.6.1)
-Requires-Dist: nltk (>=3.9.2,<4.0.0)
-Requires-Dist: python-dateutil (>=2.9.0.post0,<3.0.0)
-Requires-Dist: python-multipart (>=0.0.20,<0.0.21)
-Requires-Dist: pyyaml (>=6.0.2,<7.0.0)
-Requires-Dist: rag-core-lib (==3.2.0)
-Requires-Dist: redis (>=6.0.0,<7.0.0)
-Requires-Dist: starlette (>=0.47.2,<0.49.0)
-Requires-Dist: tenacity (==9.1.2)
-Requires-Dist: tqdm (>=4.67.1,<5.0.0)
-Requires-Dist: uvicorn (>=0.37.0,<0.38.0)