Initial WP4d

This commit is contained in:
Lars committed 2025-12-29 07:58:20 +01:00
1 parent ea9a54421a
commit feeb7c2d92
3 files changed
+77 -21

No files matched your search

+51 -18
View File
@@ -1,10 +1,11 @@
"""
FILE: app/core/graph/graph_derive_edges.py
DESCRIPTION: Hauptlogik zur Kanten-Aggregation und De-Duplizierung.
AUDIT: Integriert parse_link_target für saubere Graphen-Topologie.
"""
from typing import List, Optional, Dict, Tuple
from .graph_utils import (
_get, _edge, _mk_edge_id, _dedupe_seq,
_get, _edge, _mk_edge_id, _dedupe_seq, parse_link_target,
PROVENANCE_PRIORITY, load_types_registry, get_edge_defaults_for
)
from .graph_extractors import (
@@ -53,47 +54,79 @@ def build_edges_for_note(
# Typed & Candidate Pool (WP-15b Integration)
typed, rem = extract_typed_relations(raw)
for k, t in typed:
edges.append(_edge(k, "chunk", cid, t, note_id, {
for k, raw_t in typed:
t, sec = parse_link_target(raw_t, note_id)
if not t: continue
payload = {
"chunk_id": cid, "edge_id": _mk_edge_id(k, cid, t, "chunk", "inline:rel"),
"provenance": "explicit", "rule_id": "inline:rel", "confidence": PROVENANCE_PRIORITY["inline:rel"]
}))
}
if sec: payload["target_section"] = sec
edges.append(_edge(k, "chunk", cid, t, note_id, payload))
pool = ch.get("candidate_pool") or ch.get("candidate_edges") or []
for cand in pool:
t, k, p = cand.get("to"), cand.get("kind", "related_to"), cand.get("provenance", "semantic_ai")
raw_t, k, p = cand.get("to"), cand.get("kind", "related_to"), cand.get("provenance", "semantic_ai")
t, sec = parse_link_target(raw_t, note_id)
if t:
edges.append(_edge(k, "chunk", cid, t, note_id, {
payload = {
"chunk_id": cid, "edge_id": _mk_edge_id(k, cid, t, "chunk", f"candidate:{p}"),
"provenance": p, "rule_id": f"candidate:{p}", "confidence": PROVENANCE_PRIORITY.get(p, 0.90)
}))
}
if sec: payload["target_section"] = sec
edges.append(_edge(k, "chunk", cid, t, note_id, payload))
# Callouts & Wikilinks
call_pairs, rem2 = extract_callout_relations(rem)
for k, t in call_pairs:
edges.append(_edge(k, "chunk", cid, t, note_id, {
for k, raw_t in call_pairs:
t, sec = parse_link_target(raw_t, note_id)
if not t: continue
payload = {
"chunk_id": cid, "edge_id": _mk_edge_id(k, cid, t, "chunk", "callout:edge"),
"provenance": "explicit", "rule_id": "callout:edge", "confidence": PROVENANCE_PRIORITY["callout:edge"]
}))
}
if sec: payload["target_section"] = sec
edges.append(_edge(k, "chunk", cid, t, note_id, payload))
refs = extract_wikilinks(rem2)
for r in refs:
edges.append(_edge("references", "chunk", cid, r, note_id, {
"chunk_id": cid, "ref_text": r, "edge_id": _mk_edge_id("references", cid, r, "chunk", "explicit:wikilink"),
for raw_r in refs:
r, sec = parse_link_target(raw_r, note_id)
if not r: continue
payload = {
"chunk_id": cid, "ref_text": raw_r, "edge_id": _mk_edge_id("references", cid, r, "chunk", "explicit:wikilink"),
"provenance": "explicit", "rule_id": "explicit:wikilink", "confidence": PROVENANCE_PRIORITY["explicit:wikilink"]
}))
}
if sec: payload["target_section"] = sec
edges.append(_edge("references", "chunk", cid, r, note_id, payload))
for rel in defaults:
if rel != "references":
edges.append(_edge(rel, "chunk", cid, r, note_id, {
def_payload = {
"chunk_id": cid, "edge_id": _mk_edge_id(rel, cid, r, "chunk", f"edge_defaults:{rel}"),
"provenance": "rule", "rule_id": f"edge_defaults:{rel}", "confidence": PROVENANCE_PRIORITY["edge_defaults"]
}))
refs_all.extend(refs)
}
if sec: def_payload["target_section"] = sec
edges.append(_edge(rel, "chunk", cid, r, note_id, def_payload))
# Für Note-Scope Sammlung nutzen wir den Original-String zur Dedup, aber gesäubert
refs_all.extend([parse_link_target(r, note_id)[0] for r in refs])
# 3) Note-Scope & De-Duplizierung
if include_note_scope_refs:
refs_note = _dedupe_seq((refs_all or []) + (note_level_references or []))
# refs_all ist jetzt schon gesäubert (nur Targets)
# note_level_references müssen auch gesäubert werden
cleaned_note_refs = [parse_link_target(r, note_id)[0] for r in (note_level_references or [])]
refs_note = _dedupe_seq((refs_all or []) + cleaned_note_refs)
for r in refs_note:
if not r: continue
edges.append(_edge("references", "note", note_id, r, note_id, {
"edge_id": _mk_edge_id("references", note_id, r, "note", "explicit:note_scope"),
"provenance": "explicit", "confidence": PROVENANCE_PRIORITY["explicit:note_scope"]