mirror of
https://github.com/agentscope-ai/ReMe.git
synced 2026-09-21 00:22:45 +00:00
514bf35050
Some checks failed
Pre-commit / run (ubuntu-latest) (push) Has been cancelled
docs: add ReMe2 architecture design documentation - Add comprehensive design document (reme2.md) detailing the three-layer architecture (L1/L2/L3) for the vault system - Document new protocols for folder notes and memory management - Specify interface contracts for memory_* and vault_* tools - Outline implementation phases from current state to target refactor: fix typo in personal retriever class - Correct spelling error: 'retri eved_nodes' -> 'retrieved_nodes' in PersonalRetriever.result assignment chore: update gitignore with vault-related patterns - Add '/vault' to ignore vault directory - Add '/reme-plugin' to ignore plugin files - Add '/reme2/vault' to ignore new vault implementation ```
28 lines
1.1 KiB
Python
28 lines
1.1 KiB
Python
"""ParsedFile — Parser's single output, carrying everything one file pass produces.
|
|
|
|
Inherits `FileMetadata` (file / path / st_mtime / metadata) and adds the
|
|
parsed contents:
|
|
|
|
chunks: list[FileChunk] — semantic blocks (text + position + hash);
|
|
embeddings ARE attached here — the parser
|
|
owns the embedding step (with hash-diff
|
|
cache via `existing_chunks` parameter).
|
|
edges: list[FileEdge] — wikilink graph edges (regex / frontmatter /
|
|
LLM provenance preserved per-edge).
|
|
|
|
The Watcher passes a `ParsedFile` into `file_store.upsert_parsed(...)`,
|
|
which dispatches the three sub-payloads (meta / edges / chunks) to the
|
|
appropriate persistence pipelines while keeping the upsert atomic at
|
|
the store layer.
|
|
"""
|
|
|
|
from pydantic import Field
|
|
|
|
from .file_chunk import FileChunk
|
|
from .file_edge import FileEdge
|
|
from .file_metadata import FileMetadata
|
|
|
|
|
|
class ParsedFile(FileMetadata):
|
|
chunks: list[FileChunk] = Field(default_factory=list)
|
|
edges: list[FileEdge] = Field(default_factory=list)
|