scripts/import_markdown.py aktualisiert
Some checks failed
Deploy mindnet to llm-node / deploy (push) Failing after 1s
Some checks failed
Deploy mindnet to llm-node / deploy (push) Failing after 1s
This commit is contained in:
parent
057c12c2d9
commit
f7b1995b08
|
|
@ -1,17 +1,41 @@
|
||||||
#!/usr/bin/env python3
|
#!/usr/bin/env python3
|
||||||
|
# -*- coding: utf-8 -*-
|
||||||
|
"""
|
||||||
|
Name: scripts/import_markdown.py
|
||||||
|
Version: v2.1.1 (2025-09-05)
|
||||||
|
Kurzbeschreibung:
|
||||||
|
Importiert Obsidian-Markdown-Notes in Qdrant (Notes/Chunks/Edges).
|
||||||
|
Leitet Wikilink-Edges (references/backlink/references_at) direkt beim Import
|
||||||
|
aus Volltext + echten Chunk-Texten ab. Keine Abhängigkeit zu edges.py (Legacy).
|
||||||
|
|
||||||
|
Aufruf (aus Projekt-Root, im venv):
|
||||||
|
python3 -m scripts.import_markdown --vault ./vault [--apply] [--note-id NOTE_ID] [--embed-note]
|
||||||
|
|
||||||
|
Parameter:
|
||||||
|
--vault Pfad zum Vault (z. B. ./vault)
|
||||||
|
--apply Führt Upserts in Qdrant aus (ohne Flag = Dry-Run mit JSON-Summaries)
|
||||||
|
--note-id Bearbeite nur eine konkrete Note-ID
|
||||||
|
--embed-note Optional: Note-Vektor (Volltext) zusätzlich einbetten
|
||||||
|
|
||||||
|
Umgebungsvariablen (optional):
|
||||||
|
QDRANT_URL, QDRANT_API_KEY, COLLECTION_PREFIX, VECTOR_DIM (Default 384)
|
||||||
|
|
||||||
|
Exitcodes:
|
||||||
|
0 = OK, 2 = keine Markdown-Dateien gefunden
|
||||||
|
|
||||||
|
Wichtige Hinweise:
|
||||||
|
- Verwendet ausschließlich app.core.derive_edges für Wikilink-Kanten.
|
||||||
|
- Chunk-Texte werden an derive_wikilink_edges übergeben; nur so entstehen references_at.
|
||||||
|
- Edges-IDs sind deterministisch (UUIDv5 in qdrant_points), idempotent.
|
||||||
|
- Edges-Collection bleibt 1D-Dummy-Vektor (DOT), reine Payload-Nutzung.
|
||||||
|
|
||||||
|
Changelog:
|
||||||
|
v2.1.1: Entfernt Abhängigkeit auf app.core.edges; direkte Nutzung von derive_wikilink_edges;
|
||||||
|
sichert Übergabe der Chunk-Texte → references_at werden erzeugt.
|
||||||
|
v2.1.0: Note-Index über gesamten Vault; erste Fassung mit direkter Edge-Ableitung.
|
||||||
|
v2.0.x: Import Notes/Chunks/Embeddings, Edges via Backfill (separat).
|
||||||
|
"""
|
||||||
from __future__ import annotations
|
from __future__ import annotations
|
||||||
"""
|
|
||||||
WP-03 – STEP 1
|
|
||||||
Ziel dieses Patches: Edges (references/backlink/references_at) entstehen direkt beim Import
|
|
||||||
auf Basis der **echten Chunks** und eines **Vorab-Note-Index**. Damit verschwinden
|
|
||||||
"dangling" references_at-Quellen aus dem Backfill-Ansatz.
|
|
||||||
|
|
||||||
Ausführung (aus Projekt-Root, im venv):
|
|
||||||
(.venv) python3 -m scripts.import_markdown --vault ./vault --apply
|
|
||||||
(.venv) python3 -m scripts.validate_edges --prefix mindnet --details
|
|
||||||
|
|
||||||
Abhängigkeiten: qdrant-client, python-dotenv
|
|
||||||
"""
|
|
||||||
import argparse
|
import argparse
|
||||||
import os
|
import os
|
||||||
import glob
|
import glob
|
||||||
|
|
@ -21,7 +45,7 @@ from typing import List, Dict
|
||||||
|
|
||||||
from dotenv import load_dotenv
|
from dotenv import load_dotenv
|
||||||
|
|
||||||
# Core-Module
|
# Core-Module (gemäß Projektstruktur)
|
||||||
from app.core.parser import (
|
from app.core.parser import (
|
||||||
read_markdown,
|
read_markdown,
|
||||||
normalize_frontmatter,
|
normalize_frontmatter,
|
||||||
|
|
@ -39,14 +63,10 @@ from app.core.qdrant_points import (
|
||||||
points_for_edges,
|
points_for_edges,
|
||||||
upsert_batch,
|
upsert_batch,
|
||||||
)
|
)
|
||||||
# Neue direkte Edge-Ableitung beim Import
|
# Nur Wikilink-Ableitung (keine Kollision mit edges.py):
|
||||||
from app.core.derive_edges import build_note_index, derive_wikilink_edges
|
from app.core.derive_edges import build_note_index, derive_wikilink_edges
|
||||||
|
|
||||||
|
|
||||||
# -----------------------------------------------------------------------------
|
|
||||||
# Utilities
|
|
||||||
# -----------------------------------------------------------------------------
|
|
||||||
|
|
||||||
def iter_md(root: str, exclude=("/.obsidian/", "/_backup_frontmatter/", "/_imported/")) -> List[str]:
|
def iter_md(root: str, exclude=("/.obsidian/", "/_backup_frontmatter/", "/_imported/")) -> List[str]:
|
||||||
files = [p for p in glob.glob(os.path.join(root, "**", "*.md"), recursive=True)]
|
files = [p for p in glob.glob(os.path.join(root, "**", "*.md"), recursive=True)]
|
||||||
out: List[str] = []
|
out: List[str] = []
|
||||||
|
|
@ -71,10 +91,6 @@ def minimal_note_index_payload(abs_path: str, vault_root: str) -> Dict:
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
# -----------------------------------------------------------------------------
|
|
||||||
# Main
|
|
||||||
# -----------------------------------------------------------------------------
|
|
||||||
|
|
||||||
def main():
|
def main():
|
||||||
load_dotenv()
|
load_dotenv()
|
||||||
|
|
||||||
|
|
@ -85,7 +101,7 @@ def main():
|
||||||
ap.add_argument("--embed-note", action="store_true", help="Auch Note-Volltext einbetten (optional)")
|
ap.add_argument("--embed-note", action="store_true", help="Auch Note-Volltext einbetten (optional)")
|
||||||
args = ap.parse_args()
|
args = ap.parse_args()
|
||||||
|
|
||||||
# Qdrant
|
# Qdrant-Konfiguration (+ Collections sicherstellen)
|
||||||
cfg = QdrantConfig(
|
cfg = QdrantConfig(
|
||||||
url=os.getenv("QDRANT_URL", "http://127.0.0.1:6333"),
|
url=os.getenv("QDRANT_URL", "http://127.0.0.1:6333"),
|
||||||
api_key=os.getenv("QDRANT_API_KEY") or None,
|
api_key=os.getenv("QDRANT_API_KEY") or None,
|
||||||
|
|
@ -101,28 +117,23 @@ def main():
|
||||||
print("Keine Markdown-Dateien gefunden.", file=sys.stderr)
|
print("Keine Markdown-Dateien gefunden.", file=sys.stderr)
|
||||||
sys.exit(2)
|
sys.exit(2)
|
||||||
|
|
||||||
# ------------------------------------------------------------------
|
# (1) Vorab-Lauf: Note-Index (robuste Auflösung via ID/Titel-Slug/File-Slug)
|
||||||
# (1) Vorab-Lauf: Note-Index für robuste Zielauflösung
|
|
||||||
# ------------------------------------------------------------------
|
|
||||||
index_payloads: List[Dict] = []
|
index_payloads: List[Dict] = []
|
||||||
for path in files:
|
for path in files:
|
||||||
try:
|
try:
|
||||||
pl = minimal_note_index_payload(path, root)
|
pl = minimal_note_index_payload(path, root)
|
||||||
if not pl.get("note_id"):
|
if not pl.get("note_id"):
|
||||||
# ohne stabile ID lassen wir die Note komplett aus
|
|
||||||
continue
|
continue
|
||||||
if args.note_id and pl["note_id"] != args.note_id:
|
if args.note_id and pl["note_id"] != args.note_id:
|
||||||
continue
|
continue
|
||||||
index_payloads.append(pl)
|
index_payloads.append(pl)
|
||||||
except Exception:
|
except Exception:
|
||||||
# bewusst still: Einzeldefekte sollen den Gesamtimport nicht stoppen
|
# Einzeldefekte nicht fatal
|
||||||
continue
|
continue
|
||||||
|
|
||||||
note_index = build_note_index(index_payloads) # (by_id, by_slug, by_file_slug)
|
note_index = build_note_index(index_payloads)
|
||||||
|
|
||||||
# ------------------------------------------------------------------
|
# (2) Hauptlauf: Import + direkte Wikilink-Edge-Ableitung
|
||||||
# (2) Hauptlauf: pro Note Chunks/Embeddings + Edges aus echten Chunks
|
|
||||||
# ------------------------------------------------------------------
|
|
||||||
total_notes = 0
|
total_notes = 0
|
||||||
for path in files:
|
for path in files:
|
||||||
parsed = read_markdown(path)
|
parsed = read_markdown(path)
|
||||||
|
|
@ -154,31 +165,41 @@ def main():
|
||||||
note_vec = embed_one(parsed.body)
|
note_vec = embed_one(parsed.body)
|
||||||
|
|
||||||
# Edges direkt aus realen Chunks + Note-Index
|
# Edges direkt aus realen Chunks + Note-Index
|
||||||
# Für derive_wikilink_edges darf der Note-Payload einen Volltext enthalten
|
# ACHTUNG: derive_wikilink_edges erwartet Chunks mit Text → wir bauen eine
|
||||||
note_pl_for_edges = dict(note_pl)
|
# minimalistische Liste {"chunk_id","text"} in gleicher Reihenfolge.
|
||||||
note_pl_for_edges.setdefault("fulltext", parsed.body)
|
note_pl_for_edges = {
|
||||||
edges = derive_wikilink_edges(note_pl_for_edges, chunk_pls, note_index)
|
"note_id": fm["id"],
|
||||||
|
"title": fm.get("title"),
|
||||||
|
"path": note_pl["path"],
|
||||||
|
"fulltext": parsed.body,
|
||||||
|
}
|
||||||
|
chunks_for_links = []
|
||||||
|
for i, pl in enumerate(chunk_pls):
|
||||||
|
cid = pl.get("chunk_id") or pl.get("id")
|
||||||
|
txt = chunks[i].text if i < len(chunks) else ""
|
||||||
|
chunks_for_links.append({"chunk_id": cid, "text": txt})
|
||||||
|
|
||||||
# Dry-Run-Log
|
edges = derive_wikilink_edges(note_pl_for_edges, chunks_for_links, note_index)
|
||||||
summary = {
|
|
||||||
|
# Dry-Run-Summary
|
||||||
|
print(json.dumps({
|
||||||
"note_id": fm["id"],
|
"note_id": fm["id"],
|
||||||
"title": fm["title"],
|
"title": fm["title"],
|
||||||
"chunks": len(chunk_pls),
|
"chunks": len(chunk_pls),
|
||||||
"edges": len(edges),
|
"edges": len(edges),
|
||||||
"path": note_pl["path"],
|
"path": note_pl["path"],
|
||||||
}
|
}, ensure_ascii=False))
|
||||||
print(json.dumps(summary, ensure_ascii=False))
|
|
||||||
|
|
||||||
if args.apply:
|
if args.apply:
|
||||||
# Notes upsert (idempotent; UUIDv5)
|
# Notes upsert (idempotent; UUIDv5)
|
||||||
notes_col, note_pts = points_for_note(cfg.prefix, note_pl, note_vec, cfg.dim)
|
notes_col, note_pts = points_for_note(cfg.prefix, note_pl, note_vec, cfg.dim)
|
||||||
upsert_batch(client, notes_col, note_pts)
|
upsert_batch(client, notes_col, note_pts)
|
||||||
|
|
||||||
# Chunks upsert (idempotent; chunk_id stabilisiert notfalls qdrant_points)
|
# Chunks upsert (idempotent)
|
||||||
chunks_col, chunk_pts = points_for_chunks(cfg.prefix, chunk_pls, vectors)
|
chunks_col, chunk_pts = points_for_chunks(cfg.prefix, chunk_pls, vectors)
|
||||||
upsert_batch(client, chunks_col, chunk_pts)
|
upsert_batch(client, chunks_col, chunk_pts)
|
||||||
|
|
||||||
# Edges upsert (idempotent; edge_id deterministisch aus kind/src/tgt/seq)
|
# Edges upsert (idempotent; deterministisches edge_id)
|
||||||
edges_col, edge_pts = points_for_edges(cfg.prefix, edges)
|
edges_col, edge_pts = points_for_edges(cfg.prefix, edges)
|
||||||
upsert_batch(client, edges_col, edge_pts)
|
upsert_batch(client, edges_col, edge_pts)
|
||||||
|
|
||||||
|
|
|
||||||
Loading…
Reference in New Issue
Block a user