landvex: IOL v3 + Identify API + Neo4j-graf + byggpipeline

- FastAPI Identify API (port 8081) - Neo4j propertygraf (64 noder, 80 kanter) - Ingest-pipeline (Python) - Byggscript + integrationstester - Docker Compose setup
2026-07-05 05:55:48 +00:00
parent 8dcca96185
commit 7d40cf95bf
61 changed files with 22330 additions and 0 deletions
@@ -0,0 +1,62 @@
+#!/usr/bin/env python3
+"""
+Exportera master-data till CSV för Neo4j-import.
+"""
+import json
+import csv
+from pathlib import Path
+
+DATA = Path(__file__).parent.parent / "10-data"
+DB = Path(__file__).parent.parent / "db"
+
+def main():
+    poster = {}
+    for fn in ("lvx-klassposter-master-v3.json", "lvx-produktmodeller-vag2.json",
+               "lvx-produktmodeller-vag3.json", "lvx-produktmodeller-auto.json"):
+        fp = DATA / fn
+        if not fp.exists(): continue
+        with open(fp, encoding="utf-8") as f:
+            for p in json.load(f).get("poster", []):
+                poster[p["lvx_id"]] = p
+
+    # Noder
+    with open(DB / "noder.csv", "w", newline="", encoding="utf-8") as f:
+        w = csv.DictWriter(f, fieldnames=["lvx_id","posttyp","namn_sv","namn_en","doman","tier","verifieringsniva","konfidens"])
+        w.writeheader()
+        for p in poster.values():
+            prov = p.get("proveniens", {})
+            w.writerow({
+                "lvx_id": p["lvx_id"],
+                "posttyp": p.get("posttyp", ""),
+                "namn_sv": p.get("namn", {}).get("sv", ""),
+                "namn_en": p.get("namn", {}).get("en", ""),
+                "doman": p.get("doman", ""),
+                "tier": p.get("tier", ""),
+                "verifieringsniva": prov.get("verifieringsniva", "obekraftad"),
+                "konfidens": prov.get("konfidens", 0),
+            })
+
+    # Kanter
+    with open(DATA / "lvx-graf-kanter-v3.json", encoding="utf-8") as f:
+        kanter = json.load(f)["kanter"]
+
+    with open(DB / "kanter.csv", "w", newline="", encoding="utf-8") as f:
+        w = csv.DictWriter(f, fieldnames=["subjekt","predikat","objekt","niva","beskrivning"])
+        w.writeheader()
+        for k in kanter:
+            if k["objekt_typ"] != "lvx":
+                continue  # hoppa över textnoder i grafen
+            w.writerow({
+                "subjekt": k["subjekt"],
+                "predikat": k["predikat"],
+                "objekt": k["objekt"],
+                "niva": k.get("niva", ""),
+                "beskrivning": k.get("beskrivning", ""),
+            })
+
+    print(f"Exporterat: {len(poster)} noder, {len(kanter)} kanter")
+    print(f"  → {DB}/noder.csv")
+    print(f"  → {DB}/kanter.csv")
+
+if __name__ == "__main__":
+    main()