1 parent
3fe8463a03
commit
53058d1504
2 files changed
+78
-35
No files matched your search
+48
-14
@@ -76,9 +76,10 @@ def find_edges_for_note(note_id: str, prefix: str = "mindnet"):
|
||||
else:
|
||||
print()
|
||||
|
||||
# 4. Eingehende Kanten - Titel#Abschnitt Varianten
|
||||
print("🔍 EINGEHENDE KANTEN (target_id beginnt mit 'Titel#'):")
|
||||
# 4. Eingehende Kanten - Titel#Abschnitt Varianten mit Fuzzy-Matching
|
||||
print("🔍 EINGEHENDE KANTEN (target_id beginnt mit 'Titel#' + Fuzzy-Matching):")
|
||||
if note_title:
|
||||
import re
|
||||
# Lade alle Kanten und filtere clientseitig (wie Case D in ui_graph_service.py)
|
||||
all_filter = rest.Filter(
|
||||
must=[rest.FieldCondition(key="kind", match=rest.MatchExcept(**{"except": ["prev", "next", "belongs_to"]}))]
|
||||
@@ -86,17 +87,47 @@ def find_edges_for_note(note_id: str, prefix: str = "mindnet"):
|
||||
all_edges, _ = client.scroll(edges_col, scroll_filter=all_filter, limit=10000, with_payload=True)
|
||||
print(f" Gesamt geladen: {len(all_edges)} Kanten aus der Datenbank")
|
||||
|
||||
# Clientseitige Filterung
|
||||
matched = []
|
||||
# Normalisierungs-Funktion (wie in ui_graph_service.py)
|
||||
def normalize_title(t):
|
||||
if not t:
|
||||
return ""
|
||||
t = re.sub(r'\s*\([^)]*\)', '', t)
|
||||
t = re.sub(r'\s*\d{4}[\s–\-]*\d{0,4}', '', t)
|
||||
t = re.sub(r'^(Mein|Meine)\s+', '', t, flags=re.IGNORECASE)
|
||||
t = re.sub(r'\s+', ' ', t).strip()
|
||||
return t.lower()
|
||||
|
||||
note_title_norm = normalize_title(note_title)
|
||||
print(f" Normalisierter Note-Titel: '{note_title_norm}'")
|
||||
|
||||
# Clientseitige Filterung: Exakte Matches
|
||||
matched_exact = []
|
||||
for e in all_edges:
|
||||
tgt_id = e.payload.get("target_id", "")
|
||||
if tgt_id and (tgt_id == note_title or tgt_id.startswith(note_title + "#")):
|
||||
matched.append(e)
|
||||
matched_exact.append(e)
|
||||
|
||||
print(f" Gefunden: {len(matched)} Kanten (mit Titel#Abschnitt Varianten)")
|
||||
# Clientseitige Filterung: Fuzzy-Matches
|
||||
matched_fuzzy = []
|
||||
for e in all_edges:
|
||||
tgt_id = e.payload.get("target_id", "")
|
||||
if not tgt_id or e in matched_exact:
|
||||
continue
|
||||
tgt_base = tgt_id.split("#")[0].strip()
|
||||
tgt_norm = normalize_title(tgt_base)
|
||||
if tgt_norm and note_title_norm and len(note_title_norm) > 5:
|
||||
if (tgt_norm == note_title_norm or
|
||||
tgt_norm.startswith(note_title_norm) or
|
||||
note_title_norm.startswith(tgt_norm)):
|
||||
matched_fuzzy.append((e, tgt_norm))
|
||||
|
||||
matched = matched_exact + [e for e, _ in matched_fuzzy]
|
||||
print(f" Gefunden: {len(matched_exact)} exakte Matches, {len(matched_fuzzy)} Fuzzy-Matches")
|
||||
print(f" Gesamt: {len(matched)} Kanten")
|
||||
for i, e in enumerate(matched[:10], 1):
|
||||
pl = e.payload
|
||||
print(f" {i}. {pl.get('kind')}: {pl.get('source_id')} -> {pl.get('target_id')}")
|
||||
match_type = "EXAKT" if e in matched_exact else "FUZZY"
|
||||
print(f" {i}. [{match_type}] {pl.get('kind')}: {pl.get('source_id')} -> {pl.get('target_id')}")
|
||||
if len(matched) > 10:
|
||||
print(f" ... und {len(matched) - 10} weitere\n")
|
||||
else:
|
||||
@@ -105,14 +136,17 @@ def find_edges_for_note(note_id: str, prefix: str = "mindnet"):
|
||||
# Zeige auch einige Beispiele von target_ids, die NICHT matchen
|
||||
print(" 🔍 DEBUG: Beispiel target_ids die NICHT matchen (erste 10):")
|
||||
non_matched = []
|
||||
for e in all_edges[:100]: # Nur erste 100 prüfen
|
||||
for e in all_edges[:200]: # Erste 200 prüfen
|
||||
tgt_id = e.payload.get("target_id", "")
|
||||
if tgt_id and tgt_id != note_title and not tgt_id.startswith(note_title + "#"):
|
||||
non_matched.append(tgt_id)
|
||||
if len(non_matched) >= 10:
|
||||
break
|
||||
for i, tgt in enumerate(non_matched, 1):
|
||||
print(f" {i}. '{tgt}'")
|
||||
if not tgt_id or e in matched:
|
||||
continue
|
||||
tgt_base = tgt_id.split("#")[0].strip()
|
||||
tgt_norm = normalize_title(tgt_base)
|
||||
non_matched.append((tgt_id, tgt_norm))
|
||||
if len(non_matched) >= 10:
|
||||
break
|
||||
for i, (tgt, tgt_norm) in enumerate(non_matched, 1):
|
||||
print(f" {i}. '{tgt}' (normalisiert: '{tgt_norm}')")
|
||||
print()
|
||||
|
||||
# 5. Zeige Beispiel target_ids aus der Datenbank
|
||||
|
||||
Reference in new issue
Block a user