| 1 | # vector_db.py DOX |
| 2 | |
| 3 | ## Purpose |
| 4 | |
| 5 | - Own the `vector_db.py` helper module. |
| 6 | - This module wraps FAISS vector storage, retrieval, and comparator behavior. |
| 7 | - Keep this file-level DOX profile synchronized with `vector_db.py` because this directory is intentionally flat. |
| 8 | |
| 9 | ## Ownership |
| 10 | |
| 11 | - `vector_db.py` owns the runtime implementation. |
| 12 | - `vector_db.py.dox.md` owns durable notes about responsibilities, contracts, side effects, and verification for that implementation. |
| 13 | - Classes: |
| 14 | - `MyFaiss` (`FAISS`) |
| 15 | - `get_by_ids(self, ids: Sequence[str]) -> List[Document]` |
| 16 | - `async aget_by_ids(self, ids: Sequence[str]) -> List[Document]` |
| 17 | - `get_all_docs(self) -> dict[str, Document]` |
| 18 | - `VectorDB` (no explicit base class) |
| 19 | - `async search_by_similarity_threshold(self, query: str, limit: int, threshold: float, filter: str=...)` |
| 20 | - `async search_by_metadata(self, filter: str, limit: int=...) -> list[Document]` |
| 21 | - `async insert_documents(self, docs: list[Document])` |
| 22 | - `async delete_documents_by_ids(self, ids: list[str])` |
| 23 | - Top-level functions: |
| 24 | - `format_docs_plain(docs: list[Document]) -> list[str]` |
| 25 | - `cosine_normalizer(val: float) -> float` |
| 26 | - `get_comparator(condition: str)` |
| 27 | |
| 28 | ## Runtime Contracts |
| 29 | |
| 30 | - Helper modules own reusable framework APIs and must preserve public callers unless all callers, tests, and docs are updated together. |
| 31 | - Update this file whenever public functions, classes, persistence behavior, path/security assumptions, side effects, or cross-module contracts change. |
| 32 | - Observed side-effect areas: filesystem deletion. |
| 33 | - Imported dependency areas include: `agent`, `faiss`, `helpers`, `langchain.embeddings`, `langchain.storage`, `langchain_community.docstore.in_memory`, `langchain_community.vectorstores`, `langchain_community.vectorstores.utils`, `langchain_core.documents`, `simpleeval`, `typing`. |
| 34 | |
| 35 | ## Key Concepts |
| 36 | |
| 37 | - Important called helpers/classes observed in the source: `self.get_by_ids`, `agent.get_embedding_model`, `self._get_embeddings`, `faiss.IndexFlatIP`, `MyFaiss`, `get_comparator`, `self.db.get_all_docs`, `InMemoryByteStore`, `CacheBackedEmbeddings.from_bytes_store`, `self.db.asearch`, `comparator`, `guids.generate_id`, `zip`, `self.db.add_documents`, `self.db.aget_by_ids`, `simple_eval`, `self.embeddings.embed_query`, `InMemoryDocstore`, `self.db.adelete`. |
| 38 | - Keep request/response, tool, or helper semantics documented here at the same time as source changes. |
| 39 | |
| 40 | ## Work Guidance |
| 41 | |
| 42 | - Preserve public helper APIs used by core code and plugins unless every caller is updated. |
| 43 | - Keep path, auth, secret, persistence, network, and subprocess behavior explicit and bounded. |
| 44 | - Prefer adding cohesive helper functions here only when behavior is reused across modules. |
| 45 | |
| 46 | ## Verification |
| 47 | |
| 48 | - Run targeted tests for changed helper behavior; run security regressions for auth, filesystem, WebSocket, tunnel, upload, or secret-handling helpers. |
| 49 | - No direct test reference was found by name search; choose the nearest behavioral test or perform a focused smoke check. |
| 50 | |
| 51 | ## Child DOX Index |
| 52 | |
| 53 | No child DOX files. |