271 lines
10 KiB
Python
271 lines
10 KiB
Python
#!/usr/bin/env python3
|
|
# -*- coding: utf-8 -*-
|
|
"""
|
|
Script: import_markdown.py — Markdown → Qdrant (Notes, Chunks, Edges)
|
|
Version: 3.2.0
|
|
Datum: 2025-09-09
|
|
|
|
Kurzbeschreibung
|
|
----------------
|
|
Liest Markdown-Dateien aus einem Vault ein und schreibt Notes, Chunks und Edges
|
|
idempotent nach Qdrant. Change-Detection basiert standardmäßig auf dem **Body-Hash**.
|
|
Neu: Hash-Modus und Normalisierung sind auch per **CLI** steuerbar.
|
|
|
|
ENV / Qdrant
|
|
------------
|
|
- QDRANT_URL (oder QDRANT_HOST/QDRANT_PORT)
|
|
- QDRANT_API_KEY (optional)
|
|
- COLLECTION_PREFIX (Default: mindnet)
|
|
- VECTOR_DIM (Default: 384)
|
|
- MINDNET_HASH_MODE: "body" (Default) | "frontmatter" | "body+frontmatter"
|
|
- MINDNET_HASH_NORMALIZE: "canonical" (Default) | "none"
|
|
|
|
CLI (übersteuert ENV)
|
|
---------------------
|
|
--hash-mode body|frontmatter|body+frontmatter
|
|
--hash-normalize canonical|none
|
|
|
|
Aufruf
|
|
------
|
|
python3 -m scripts.import_markdown --vault ./vault
|
|
python3 -m scripts.import_markdown --vault ./vault --apply
|
|
python3 -m scripts.import_markdown --vault ./vault --apply --purge-before-upsert
|
|
python3 -m scripts.import_markdown --vault ./vault --note-id 20250821-foo --apply
|
|
python3 -m scripts.import_markdown --vault ./vault --apply --embed-note
|
|
# Feingranulare Erkennung (jede Kleinigkeit im Body zählt):
|
|
python3 -m scripts.import_markdown --vault ./vault --hash-normalize none
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import argparse
|
|
import json
|
|
import os
|
|
import sys
|
|
from typing import Dict, List, Optional, Tuple
|
|
|
|
from dotenv import load_dotenv
|
|
from qdrant_client.http import models as rest
|
|
|
|
from app.core.parser import (
|
|
read_markdown,
|
|
normalize_frontmatter,
|
|
validate_required_frontmatter,
|
|
extract_wikilinks,
|
|
)
|
|
from app.core.note_payload import make_note_payload
|
|
from app.core.chunker import assemble_chunks
|
|
from app.core.chunk_payload import make_chunk_payloads
|
|
from app.core.qdrant import QdrantConfig, get_client, ensure_collections
|
|
from app.core.qdrant_points import (
|
|
points_for_chunks,
|
|
points_for_note,
|
|
points_for_edges,
|
|
upsert_batch,
|
|
)
|
|
|
|
try:
|
|
from app.core.embed import embed_texts, embed_one # optional
|
|
except Exception:
|
|
embed_texts = None
|
|
embed_one = None
|
|
|
|
|
|
# -----------------------------------------------------------------------------
|
|
# Hilfsfunktionen
|
|
# -----------------------------------------------------------------------------
|
|
|
|
def iter_md(root: str) -> List[str]:
|
|
out: List[str] = []
|
|
for dirpath, _, filenames in os.walk(root):
|
|
for fn in filenames:
|
|
if not fn.lower().endswith(".md"):
|
|
continue
|
|
p = os.path.join(dirpath, fn)
|
|
pn = p.replace("\\", "/")
|
|
if any(ex in pn for ex in ["/.obsidian/", "/_backup_frontmatter/", "/_imported/"]):
|
|
continue
|
|
out.append(p)
|
|
return sorted(out)
|
|
|
|
def collections(prefix: str) -> Tuple[str, str, str]:
|
|
return f"{prefix}_notes", f"{prefix}_chunks", f"{prefix}_edges"
|
|
|
|
def fetch_existing_note_hash(client, prefix: str, note_id: str) -> Optional[str]:
|
|
notes_col, _, _ = collections(prefix)
|
|
f = rest.Filter(must=[rest.FieldCondition(
|
|
key="note_id",
|
|
match=rest.MatchValue(value=note_id),
|
|
)])
|
|
points, _ = client.scroll(
|
|
collection_name=notes_col,
|
|
scroll_filter=f,
|
|
with_payload=True,
|
|
with_vectors=False,
|
|
limit=1,
|
|
)
|
|
if not points:
|
|
return None
|
|
pl = points[0].payload or {}
|
|
return pl.get("hash_fulltext")
|
|
|
|
def purge_note_artifacts(client, prefix: str, note_id: str) -> None:
|
|
_, chunks_col, edges_col = collections(prefix)
|
|
f_chunks = rest.Filter(must=[rest.FieldCondition(key="note_id", match=rest.MatchValue(value=note_id))])
|
|
client.delete(collection_name=chunks_col, points_selector=f_chunks, wait=True)
|
|
should = [
|
|
rest.FieldCondition(key="source_id", match=rest.MatchText(text=f"{note_id}#")),
|
|
rest.FieldCondition(key="target_id", match=rest.MatchText(text=f"{note_id}#")),
|
|
rest.FieldCondition(key="source_id", match=rest.MatchValue(value=note_id)),
|
|
rest.FieldCondition(key="target_id", match=rest.MatchValue(value=note_id)),
|
|
]
|
|
f_edges = rest.Filter(should=should)
|
|
client.delete(collection_name=edges_col, points_selector=f_edges, wait=True)
|
|
|
|
def _normalize_rel_path(abs_path: str, vault_root: str) -> str:
|
|
try:
|
|
rel = os.path.relpath(abs_path, vault_root)
|
|
except Exception:
|
|
rel = abs_path
|
|
return rel.replace("\\", "/").lstrip("/")
|
|
|
|
|
|
# -----------------------------------------------------------------------------
|
|
# Main
|
|
# -----------------------------------------------------------------------------
|
|
|
|
def main() -> None:
|
|
load_dotenv()
|
|
ap = argparse.ArgumentParser()
|
|
ap.add_argument("--vault", required=True, help="Pfad zum Obsidian-Vault (Root-Ordner)")
|
|
ap.add_argument("--apply", action="store_true", help="Schreibt in Qdrant; ohne Flag nur Dry-Run")
|
|
ap.add_argument("--purge-before-upsert", action="store_true",
|
|
help="Vor Upsert Chunks & Edges der GEÄNDERTEN Note löschen")
|
|
ap.add_argument("--note-id", help="Nur eine bestimmte Note-ID verarbeiten")
|
|
ap.add_argument("--embed-note", action="store_true", help="Optional: Note-Volltext einbetten")
|
|
ap.add_argument("--force-replace", action="store_true",
|
|
help="Änderungserkennung ignorieren und immer upserten (+ optional Purge)")
|
|
# NEU: Hash-Steuerung per CLI
|
|
ap.add_argument("--hash-mode", choices=["body", "frontmatter", "body+frontmatter"], default=None)
|
|
ap.add_argument("--hash-normalize", choices=["canonical", "none"], default=None)
|
|
args = ap.parse_args()
|
|
|
|
cfg = QdrantConfig.from_env()
|
|
client = get_client(cfg)
|
|
ensure_collections(client, cfg.prefix, cfg.dim)
|
|
notes_col, chunks_col, edges_col = collections(cfg.prefix)
|
|
|
|
root = os.path.abspath(args.vault)
|
|
files = iter_md(root)
|
|
if not files:
|
|
print("Keine Markdown-Dateien gefunden.", file=sys.stderr)
|
|
sys.exit(2)
|
|
|
|
processed = 0
|
|
for path in files:
|
|
parsed = read_markdown(path)
|
|
fm = normalize_frontmatter(parsed.frontmatter)
|
|
|
|
try:
|
|
validate_required_frontmatter(fm)
|
|
except Exception as e:
|
|
print(json.dumps({"path": path, "error": f"Frontmatter invalid: {e}"}))
|
|
continue
|
|
|
|
if args.note_id and fm.get("id") != args.note_id:
|
|
continue
|
|
|
|
processed += 1
|
|
|
|
# Note-Payload (mit expliziten Hash-Parametern)
|
|
note_pl = make_note_payload(parsed, vault_root=root,
|
|
hash_mode=args.hash_mode, hash_normalize=args.hash_normalize)
|
|
if "fulltext" not in (note_pl or {}):
|
|
note_pl["fulltext"] = parsed.body or ""
|
|
if note_pl.get("path"):
|
|
note_pl["path"] = _normalize_rel_path(
|
|
os.path.join(root, note_pl["path"]) if not os.path.isabs(note_pl["path"]) else note_pl["path"], root
|
|
)
|
|
else:
|
|
note_pl["path"] = _normalize_rel_path(parsed.path, root)
|
|
|
|
note_id = note_pl["note_id"]
|
|
|
|
# Change-Detection
|
|
new_hash = note_pl.get("hash_fulltext")
|
|
old_hash = None if args.force_replace else fetch_existing_note_hash(client, cfg.prefix, note_id)
|
|
changed = args.force_replace or (old_hash != new_hash)
|
|
|
|
# Chunks + Embeddings (Nullvektor-Fallback)
|
|
chunks = assemble_chunks(fm["id"], parsed.body, fm.get("type", "concept"))
|
|
chunk_pls = make_chunk_payloads(fm, note_pl["path"], chunks)
|
|
if embed_texts:
|
|
vecs = embed_texts([getattr(c, "text", "") for c in chunks])
|
|
else:
|
|
vecs = [[0.0] * cfg.dim for _ in chunks]
|
|
|
|
# Edges (leichtgewichtig direkt hier ableiten)
|
|
# belongs_to / prev/next aus Chunk-Nachbarschaften + Wikilinks als references
|
|
edges: List[Dict] = []
|
|
for ch in chunk_pls:
|
|
cid = ch["id"]
|
|
edges.append({"src_id": cid, "dst_id": note_id, "edge_type": "belongs_to", "scope": "chunk"})
|
|
nb = ch.get("neighbors") or {}
|
|
if nb.get("prev"):
|
|
edges.append({"src_id": nb["prev"], "dst_id": cid, "edge_type": "next", "scope": "chunk"})
|
|
edges.append({"src_id": cid, "dst_id": nb["prev"], "edge_type": "prev", "scope": "chunk"})
|
|
if nb.get("next"):
|
|
edges.append({"src_id": cid, "dst_id": nb["next"], "edge_type": "next", "scope": "chunk"})
|
|
edges.append({"src_id": nb["next"], "dst_id": cid, "edge_type": "prev", "scope": "chunk"})
|
|
for ref in (ch.get("references") or []):
|
|
tid = ref.get("target_id")
|
|
if tid:
|
|
edges.append({"src_id": cid, "dst_id": tid, "edge_type": "references", "scope": "chunk"})
|
|
for tid in (note_pl.get("references") or []):
|
|
edges.append({"src_id": note_id, "dst_id": tid, "edge_type": "references", "scope": "note"})
|
|
edges.append({"src_id": tid, "dst_id": note_id, "edge_type": "backlink", "scope": "note"})
|
|
# Dedupe
|
|
_uniq = {}
|
|
for e in edges:
|
|
_uniq[(e["src_id"], e["dst_id"], e["edge_type"], e.get("scope", ""))] = e
|
|
edges = list(_uniq.values())
|
|
|
|
# Zusammenfassung
|
|
summary = {
|
|
"note_id": note_id,
|
|
"title": fm.get("title"),
|
|
"chunks": len(chunk_pls),
|
|
"edges": len(edges),
|
|
"changed": changed,
|
|
"decision": ("apply" if args.apply and changed else
|
|
"apply-skip-unchanged" if args.apply and not changed else
|
|
"dry-run"),
|
|
"path": note_pl["path"],
|
|
"hash_mode": args.hash_mode or os.environ.get("MINDNET_HASH_MODE", "body"),
|
|
"hash_normalize": args.hash_normalize or os.environ.get("MINDNET_HASH_NORMALIZE", "canonical"),
|
|
"hash_old": old_hash,
|
|
"hash_new": new_hash,
|
|
}
|
|
print(json.dumps(summary, ensure_ascii=False))
|
|
|
|
if not args.apply:
|
|
continue
|
|
|
|
if changed and args.purge_before_upsert:
|
|
purge_note_artifacts(client, cfg.prefix, note_id)
|
|
|
|
notes_name, note_pts = points_for_note(cfg.prefix, note_pl, None, cfg.dim)
|
|
upsert_batch(client, notes_name, note_pts)
|
|
|
|
chunks_name, chunk_pts = points_for_chunks(cfg.prefix, chunk_pls, vecs)
|
|
upsert_batch(client, chunks_name, chunk_pts)
|
|
|
|
edges_name, edge_pts = points_for_edges(cfg.prefix, edges)
|
|
upsert_batch(client, edges_name, edge_pts)
|
|
|
|
print(f"Done. Processed notes: {processed}")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|