67 lines
1.9 KiB
Python
67 lines
1.9 KiB
Python
#!/usr/bin/env python3
|
|
"""
|
|
scripts/import_markdown.py
|
|
Refactored CLI-Wrapper für den IngestionService.
|
|
"""
|
|
import argparse
|
|
import os
|
|
import json
|
|
import sys
|
|
from dotenv import load_dotenv
|
|
from app.core.ingestion import IngestionService
|
|
|
|
def iter_md(root: str):
|
|
out = []
|
|
for dp, _, fns in os.walk(root):
|
|
for fn in fns:
|
|
if fn.endswith(".md") and "/.obsidian/" not in dp:
|
|
out.append(os.path.join(dp, fn).replace("\\", "/"))
|
|
return sorted(out)
|
|
|
|
def main():
|
|
load_dotenv()
|
|
|
|
# FIX: Default Prefix aus Environment holen, sonst Fallback auf "mindnet"
|
|
default_prefix = os.getenv("COLLECTION_PREFIX", "mindnet")
|
|
|
|
ap = argparse.ArgumentParser()
|
|
ap.add_argument("--vault", required=True)
|
|
ap.add_argument("--apply", action="store_true")
|
|
ap.add_argument("--purge-before-upsert", action="store_true")
|
|
ap.add_argument("--force-replace", action="store_true")
|
|
|
|
# Hier nutzen wir jetzt die Variable
|
|
ap.add_argument("--prefix", default=default_prefix)
|
|
|
|
args = ap.parse_args()
|
|
|
|
print(f"Init IngestionService (Prefix: {args.prefix})...")
|
|
service = IngestionService(collection_prefix=args.prefix)
|
|
|
|
files = iter_md(os.path.abspath(args.vault))
|
|
print(f"Found {len(files)} files in vault.")
|
|
|
|
processed = 0
|
|
errors = 0
|
|
|
|
for f in files:
|
|
res = service.process_file(
|
|
file_path=f,
|
|
vault_root=os.path.abspath(args.vault),
|
|
apply=args.apply,
|
|
force_replace=args.force_replace,
|
|
purge_before=args.purge_before_upsert
|
|
)
|
|
|
|
if res.get("status") not in ["skipped", "unchanged"]:
|
|
print(json.dumps(res, ensure_ascii=False))
|
|
processed += 1
|
|
|
|
if res.get("error"):
|
|
print(json.dumps(res, ensure_ascii=False), file=sys.stderr)
|
|
errors += 1
|
|
|
|
print(f"Done. Processed/Changed: {processed}. Errors: {errors}")
|
|
|
|
if __name__ == "__main__":
|
|
main() |