Dateien nach "app/core" hochladen

app/core/derive_edges.py

@@ -1,286 +1,423 @@
-# app/core/derive_edges.py
+#!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 """
-Mindnet V2 — Edge derivation
+Modul: app/core/derive_edges.py
+Zweck:
+- Bewahrt bestehende Edgelogik (belongs_to, prev/next, references, backlink)
+- Ergänzt typenbasierte Default-Kanten (edge_defaults aus config/types.yaml)
+- Unterstützt "typed inline relations" ([[rel:KIND | Target]] / [[rel:KIND Target]])
+- Unterstützt Obsidian-Callouts (> [!edge] KIND: [[Target]] [[Target2]] ...)
 
-Features preserved & extended:
+Kompatibilität:
-- Structure edges: belongs_to, next, prev (deterministic order by index/ord/chunk_id)
+- build_edges_for_note(...) Signatur unverändert
-- Explicit references from wikilinks [[Title]]  -> kind="references", rule_id="explicit:wikilink", confidence=1.0
+- rule_id Werte exakt wie zuvor erwartet (ohne Versionssuffix):
-- Inline relations:
+  * structure:belongs_to
-    A) [[rel:<kind> <Target>]]                      -> kind from link, rule_id="inline:rel", confidence=0.95
+  * structure:order
-    B) rel: <kind> [[Target1]] [[Target2]] ...      -> supports multiple wikilinks on a single line
+  * explicit:wikilink
-- Callout relations:
+  * inline:rel
-    > [!edge] <kind>: [[Target1]] [[Target2]]       -> rule_id="callout:edge", confidence=0.9
+  * callout:edge
-- Type-based defaults (from types.yaml):
+  * edge_defaults:<type>:<relation>
-    For each note type's `edge_defaults`, derive additional edges for every explicit/inline/callout target.
+  * derived:backlink
-    rule_id="edge_defaults:<note_type>:<kind>", confidence=0.7
-    Symmetric kinds ("related_to","similar_to") also add reversed edges.
-- De-duplication across all sources by (kind, scope, source_id, target_id, rule_id)
-- Backward-compatible function signature
-
-Return: list[dict] with at least:
-    kind, scope, source_id, target_id, note_id, chunk_id, rule_id, confidence
 """
 
 from __future__ import annotations
 
+import os
 import re
-from typing import Any, Dict, Iterable, List, Optional, Sequence, Tuple
+from typing import Iterable, List, Optional, Tuple, Set, Dict
 
-# Kanten, die symmetrisch interpretiert werden (Rückkante wird erzeugt)
+try:
-SYMMETRIC_KINDS = {"related_to", "similar_to"}
+    import yaml  # optional, nur für types.yaml
+except Exception:  # pragma: no cover
+    yaml = None
 
-# -----------------------------------------------------------------------------
+# --------------------------------------------------------------------------- #
 # Utilities
-# -----------------------------------------------------------------------------
+# --------------------------------------------------------------------------- #
 
-def _get(d: Dict[str, Any], *keys: str, default: Any = None) -> Any:
+def _get(d: dict, *keys, default=None):
     for k in keys:
-        if isinstance(d, dict) and k in d:
+        if k in d and d[k] is not None:
             return d[k]
     return default
 
-def _safe_int(x: Any, default: int = 10**9) -> int:
+def _chunk_text_for_refs(chunk: dict) -> str:
-    try:
+    # bevorzugt 'window' → dann 'text' → 'content' → 'raw'
-        return int(x)
+    return (
-    except Exception:
+        _get(chunk, "window")
-        return default
+        or _get(chunk, "text")
+        or _get(chunk, "content")
+        or _get(chunk, "raw")
+        or ""
+    )
 
-def _sort_chunks(chs: Sequence[Dict[str, Any]]) -> List[Dict[str, Any]]:
+def _dedupe_seq(seq: Iterable[str]) -> List[str]:
-    """Sort chunks primarily by index, then ord, then chunk_id to guarantee stable next/prev."""
+    seen: Set[str] = set()
-    return sorted(chs, key=lambda ch: (
+    out: List[str] = []
-        _get(ch, "index") is None, _safe_int(_get(ch, "index")),
+    for s in seq:
-        _get(ch, "ord")   is None, _safe_int(_get(ch, "ord")),
+        if s not in seen:
-        str(_get(ch, "chunk_id") or _get(ch, "id") or "")
+            seen.add(s)
-    ))
+            out.append(s)
+    return out
 
-def _make_edge(kind: str,
+def _edge(kind: str, scope: str, source_id: str, target_id: str, note_id: str, extra: Optional[dict] = None) -> dict:
-               source_id: str,
+    pl = {
-               target_id: str,
-               note_id: str,
-               chunk_id: str,
-               rule_id: str,
-               confidence: float,
-               scope: str = "chunk") -> Dict[str, Any]:
-    return {
         "kind": kind,
-        "scope": scope,
+        "relation": kind,   # Alias (v2)
+        "scope": scope,     # "chunk" | "note"
         "source_id": source_id,
         "target_id": target_id,
-        "note_id": note_id,
+        "note_id": note_id, # Träger-Note der Kante
-        "chunk_id": chunk_id,
-        "rule_id": rule_id,
-        "confidence": float(confidence),
     }
+    if extra:
+        pl.update(extra)
+    return pl
 
-def _dedupe_key(e: Dict[str, Any]) -> Tuple[Any, ...]:
+def _mk_edge_id(kind: str, s: str, t: str, scope: str, rule_id: Optional[str] = None) -> str:
-    return (
+    base = f"{kind}:{s}->{t}#{scope}"
-        e.get("kind"),
+    if rule_id:
-        e.get("scope"),
+        base += f"|{rule_id}"
-        e.get("source_id"),
+    try:
-        e.get("target_id"),
+        import hashlib
-        e.get("rule_id"),
+        return hashlib.blake2s(base.encode("utf-8"), digest_size=12).hexdigest()
-    )
+    except Exception:  # pragma: no cover
+        return base
 
-# -----------------------------------------------------------------------------
+# --------------------------------------------------------------------------- #
-# Parsing helpers
+# Typen-Registry (types.yaml)
-# -----------------------------------------------------------------------------
+# --------------------------------------------------------------------------- #
 
-RE_WIKILINK = re.compile(r"\[\[([^\]]+?)\]\]")
+def _env(n: str, default: Optional[str] = None) -> str:
-RE_INLINE_BRACKET = re.compile(
+    v = os.getenv(n)
-    r"\[\[\s*rel\s*:\s*(?P<kind>[a-z_][a-z0-9_]*)\s+(?P<target>[^\]]+?)\s*\]\]",
+    return v if v is not None else (default or "")
-    flags=re.IGNORECASE,
-)
-RE_INLINE_PREFIX = re.compile(
-    r"(?m)\brel\s*:\s*(?P<kind>[a-z_][a-z0-9_]*)\s+(?P<links>(?:\[\[[^\]]+\]\]\s*)+)",
-    flags=re.IGNORECASE,
-)
-RE_CALLOUT = re.compile(
-    r"(?m)^\s*>\s*\[!edge\]\s*(?P<kind>[a-z_][a-z0-9_]*)\s*:\s*(?P<body>.+)$",
-    flags=re.IGNORECASE,
-)
 
-def _extract_wikilinks(text: str) -> List[str]:
+def _load_types_registry() -> dict:
-    """All [[Title]] except those starting with 'rel:' (reserved for inline relations)."""
+    """Lädt die YAML-Registry aus MINDNET_TYPES_FILE oder ./config/types.yaml"""
-    targets: List[str] = []
+    p = _env("MINDNET_TYPES_FILE", "./config/types.yaml")
-    for m in RE_WIKILINK.finditer(text or ""):
+    if not os.path.isfile(p) or yaml is None:
-        label = m.group(1).strip()
+        return {}
-        if label.lower().startswith("rel:"):
+    try:
-            # handled by explicit inline relation parser
+        with open(p, "r", encoding="utf-8") as f:
-            continue
+            data = yaml.safe_load(f) or {}
-        if label:
+            return data
-            targets.append(label)
+    except Exception:
-    return targets
+        return {}
 
-def _extract_inline_relations(text: str) -> List[Tuple[str, str]]:
+def _get_types_map(reg: dict) -> dict:
-    """Find inline relations from:
+    if isinstance(reg, dict) and isinstance(reg.get("types"), dict):
-       A) [[rel:kind Target]]
+        return reg["types"]
-       B) rel: kind [[Target1]] [[Target2]] (supports multiple targets on one line)
+    return reg if isinstance(reg, dict) else {}
+
+def _edge_defaults_for(note_type: Optional[str], reg: dict) -> List[str]:
     """
-    out: List[Tuple[str, str]] = []
+    Liefert die edge_defaults-Liste für den gegebenen Notiztyp.
+    Fallback-Reihenfolge:
+      1) reg['types'][note_type]['edge_defaults']
+      2) reg['defaults']['edge_defaults']  (oder 'default'/'global')
+      3) []
+    """
+    types_map = _get_types_map(reg)
+    if note_type and isinstance(types_map, dict):
+        t = types_map.get(note_type)
+        if isinstance(t, dict) and isinstance(t.get("edge_defaults"), list):
+            return [str(x) for x in t["edge_defaults"] if isinstance(x, str)]
+    for key in ("defaults", "default", "global"):
+        v = reg.get(key)
+        if isinstance(v, dict) and isinstance(v.get("edge_defaults"), list):
+            return [str(x) for x in v["edge_defaults"] if isinstance(x, str)]
+    return []
 
-    # A) Bracket form
+# --------------------------------------------------------------------------- #
-    for m in RE_INLINE_BRACKET.finditer(text or ""):
+# Parser für Links / Relationen
-        kind = m.group("kind").strip().lower()
+# --------------------------------------------------------------------------- #
-        target = m.group("target").strip()
-        if kind and target:
-            out.append((kind, target))
 
-    # B) Prefix form with multiple wikilinks
+# Normale Wikilinks (Fallback)
-    for m in RE_INLINE_PREFIX.finditer(text or ""):
+_WIKILINK_RE = re.compile(r"\[\[(?:[^\|\]]+\|)?([a-zA-Z0-9_\-#:. ]+)\]\]")
-        kind = m.group("kind").strip().lower()
-        links = m.group("links")
-        for lm in RE_WIKILINK.finditer(links or ""):
-            target = lm.group(1).strip()
-            if target:
-                out.append((kind, target))
 
-    return out
+# Getypte Inline-Relationen:
+#   [[rel:depends_on | Target]]
+#   [[rel:related_to Target]]
+_REL_PIPE  = re.compile(r"\[\[\s*rel:(?P<kind>[a-z_]+)\s*\|\s*(?P<target>[^\]]+?)\s*\]\]", re.IGNORECASE)
+_REL_SPACE = re.compile(r"\[\[\s*rel:(?P<kind>[a-z_]+)\s+(?P<target>[^\]]+?)\s*\]\]",   re.IGNORECASE)
 
-def _extract_callout_relations(text: str) -> List[Tuple[str, str]]:
+def _extract_typed_relations(text: str) -> Tuple[List[Tuple[str,str]], str]:
-    """Find callout > [!edge] kind: [[Target1]] [[Target2]] | Target, Target2"""
+    """
-    out: List[Tuple[str, str]] = []
+    Gibt Liste (kind, target) zurück und den Text mit entfernten getypten Relation-Links,
-    for m in RE_CALLOUT.finditer(text or ""):
+    damit die generische Wikilink-Erkennung sie nicht doppelt zählt.
-        kind = (m.group("kind") or "").strip().lower()
+    """
-        body = m.group("body") or ""
+    pairs: List[Tuple[str,str]] = []
-        # Prefer wikilinks; fallback: split by comma
+    def _collect(m):
-        wl = [mm.group(1).strip() for mm in RE_WIKILINK.finditer(body)]
+        k = (m.group("kind") or "").strip().lower()
-        if wl:
+        t = (m.group("target") or "").strip()
-            out.extend((kind, t) for t in wl if t)
+        if k and t:
+            pairs.append((k, t))
+        return ""  # Link entfernen
+    text = _REL_PIPE.sub(_collect, text)
+    text = _REL_SPACE.sub(_collect, text)
+    return pairs, text
+
+# Obsidian Callout Parser
+_CALLOUT_START = re.compile(r"^\s*>\s*\[!edge\]\s*(.*)$", re.IGNORECASE)
+_REL_LINE      = re.compile(r"^(?P<kind>[a-z_]+)\s*:\s*(?P<targets>.+?)\s*$", re.IGNORECASE)
+_WIKILINKS_IN_LINE = re.compile(r"\[\[([^\]]+)\]\]")
+
+def _extract_callout_relations(text: str) -> Tuple[List[Tuple[str,str]], str]:
+    """
+    Findet [!edge]-Callouts und extrahiert (kind, target). Entfernt den gesamten
+    Callout-Block aus dem Text (damit Wikilinks daraus nicht zusätzlich als
+    "references" gezählt werden).
+    """
+    if not text:
+        return [], text
+
+    lines = text.splitlines()
+    out_pairs: List[Tuple[str,str]] = []
+    keep_lines: List[str] = []
+    i = 0
+
+    while i < len(lines):
+        m = _CALLOUT_START.match(lines[i])
+        if not m:
+            keep_lines.append(lines[i])
+            i += 1
+            continue
+
+        block_lines: List[str] = []
+        first_rest = m.group(1) or ""
+        if first_rest.strip():
+            block_lines.append(first_rest)
+
+        i += 1
+        while i < len(lines) and lines[i].lstrip().startswith('>'):
+            block_lines.append(lines[i].lstrip()[1:].lstrip())
+            i += 1
+
+        for bl in block_lines:
+            mrel = _REL_LINE.match(bl)
+            if not mrel:
+                continue
+            kind = (mrel.group("kind") or "").strip().lower()
+            targets = mrel.group("targets") or ""
+            found = _WIKILINKS_IN_LINE.findall(targets)
+            if found:
+                for t in found:
+                    t = t.strip()
+                    if t:
+                        out_pairs.append((kind, t))
             else:
-            for raw in body.split(","):
+                for raw in re.split(r"[,;]", targets):
                     t = raw.strip()
                     if t:
-                    out.append((kind, t))
+                        out_pairs.append((kind, t))
-    return out
 
-# -----------------------------------------------------------------------------
+        # Callout wird NICHT in keep_lines übernommen
-# Edge builder
-# -----------------------------------------------------------------------------
-
-def _edge_defaults_for_type(types_cfg: Optional[Dict[str, Any]], note_type: Optional[str]) -> List[str]:
-    """Read defaults from types.yaml structure: {'types': {<type>: {'edge_defaults': [..]}}}"""
-    if not types_cfg or not note_type:
-        return []
-    t = types_cfg.get("types", {}).get(note_type, {})
-    vals = t.get("edge_defaults", []) or []
-    return [str(v) for v in vals if isinstance(v, (str,))]
-
-def _append_with_dedupe(edges: List[Dict[str, Any]], new_edges: Iterable[Dict[str, Any]]) -> None:
-    seen = { _dedupe_key(e) for e in edges }
-    for e in new_edges:
-        k = _dedupe_key(e)
-        if k in seen:
         continue
-        edges.append(e)
-        seen.add(k)
 
-def _structure_edges(note_id: str, title: str, chunks: Sequence[Dict[str, Any]]) -> List[Dict[str, Any]]:
+    remainder = "\n".join(keep_lines)
-    edges: List[Dict[str, Any]] = []
+    return out_pairs, remainder
-    ordered = _sort_chunks(chunks)
+
-    # belongs_to
+def _extract_wikilinks(text: str) -> List[str]:
-    for ch in ordered:
+    ids: List[str] = []
+    for m in _WIKILINK_RE.finditer(text or ""):
+        ids.append(m.group(1).strip())
+    return ids
+
+# --------------------------------------------------------------------------- #
+# Hauptfunktion
+# --------------------------------------------------------------------------- #
+
+def build_edges_for_note(
+    note_id: str,
+    chunks: List[dict],
+    note_level_references: Optional[List[str]] = None,
+    include_note_scope_refs: bool = False,
+) -> List[dict]:
+    """
+    Erzeugt Kanten für eine Note.
+
+    - belongs_to:   für jeden Chunk   (chunk -> note)
+    - next / prev:  zwischen aufeinanderfolgenden Chunks
+    - references:   pro Chunk aus window/text (via Wikilinks)
+    - typed inline relations: [[rel:KIND | Target]] oder [[rel:KIND Target]]
+    - Obsidian Callouts: > [!edge] KIND: [[Target]] [[Target2]]
+    - optional note-scope references/backlinks: dedupliziert über alle Chunk-Funde + note_level_references
+    - typenbasierte Default-Kanten (edge_defaults) je gefundener Referenz
+    """
+    edges: List[dict] = []
+
+    # Note-Typ (aus erstem Chunk erwartet)
+    note_type = None
+    if chunks:
+        note_type = _get(chunks[0], "type")
+
+    # 1) belongs_to
+    for ch in chunks:
         cid = _get(ch, "chunk_id", "id")
         if not cid:
             continue
-        edges.append(_make_edge("belongs_to", cid, note_id, note_id, cid, "structure:belongs_to", 1.0))
+        edges.append(_edge("belongs_to", "chunk", cid, note_id, note_id, {
-    # next/prev
+            "chunk_id": cid,
-    for i in range(len(ordered) - 1):
+            "edge_id": _mk_edge_id("belongs_to", cid, note_id, "chunk", "structure:belongs_to"),
-        a, b = ordered[i], ordered[i + 1]
+            "provenance": "rule",
-        a_id = _get(a, "chunk_id", "id"); b_id = _get(b, "chunk_id", "id")
+            "rule_id": "structure:belongs_to",
+            "confidence": 1.0,
+        }))
+
+    # 2) next / prev
+    for i in range(len(chunks) - 1):
+        a, b = chunks[i], chunks[i + 1]
+        a_id = _get(a, "chunk_id", "id")
+        b_id = _get(b, "chunk_id", "id")
         if not a_id or not b_id:
             continue
-        edges.append(_make_edge("next", a_id, b_id, note_id, a_id, "structure:next", 1.0))
+        edges.append(_edge("next", "chunk", a_id, b_id, note_id, {
-        edges.append(_make_edge("prev", b_id, a_id, note_id, b_id, "structure:prev", 1.0))
+            "chunk_id": a_id,
-    return edges
+            "edge_id": _mk_edge_id("next", a_id, b_id, "chunk", "structure:order"),
+            "provenance": "rule",
+            "rule_id": "structure:order",
+            "confidence": 0.95,
+        }))
+        edges.append(_edge("prev", "chunk", b_id, a_id, note_id, {
+            "chunk_id": b_id,
+            "edge_id": _mk_edge_id("prev", b_id, a_id, "chunk", "structure:order"),
+            "provenance": "rule",
+            "rule_id": "structure:order",
+            "confidence": 0.95,
+        }))
 
-def _explicit_and_inline_edges(note_id: str,
+    # 3) references + typed inline + callouts + defaults (chunk-scope)
-                               title: str,
+    reg = _load_types_registry()
-                               note_type: Optional[str],
+    defaults = _edge_defaults_for(note_type, reg)
-                               chunks: Sequence[Dict[str, Any]],
+    refs_all: List[str] = []
-                               types_cfg: Optional[Dict[str, Any]]) -> Tuple[List[Dict[str, Any]], List[str]]:
-    """Return (edges, referenced_targets)"""
-    edges: List[Dict[str, Any]] = []
-    referenced: List[str] = []
 
-    defaults_kinds = set(_edge_defaults_for_type(types_cfg, note_type))
+    for ch in chunks:
-
-    for ch in _sort_chunks(chunks):
-        cid = _get(ch, "chunk_id", "id")
-        text = _get(ch, "text", "window", default="") or ""
-
-        # --- Explicit wikilinks -> references
-        for target in _extract_wikilinks(text):
-            referenced.append(target)
-            edges.append(_make_edge("references", cid, target, note_id, cid, "explicit:wikilink", 1.0))
-
-        # --- Inline relations (both forms)
-        for kind, target in _extract_inline_relations(text):
-            referenced.append(target)
-            # forward
-            edges.append(_make_edge(kind, cid, target, note_id, cid, "inline:rel", 0.95))
-            # symmetric reverse
-            if kind in SYMMETRIC_KINDS:
-                edges.append(_make_edge(kind, target, cid, note_id, cid, "inline:rel", 0.95))
-
-        # --- Callout relations
-        for kind, target in _extract_callout_relations(text):
-            referenced.append(target)
-            edges.append(_make_edge(kind, cid, target, note_id, cid, "callout:edge", 0.90))
-            if kind in SYMMETRIC_KINDS:
-                edges.append(_make_edge(kind, target, cid, note_id, cid, "callout:edge", 0.90))
-
-    # --- Type-based defaults: apply to the union of referenced targets
-    if defaults_kinds and referenced:
-        uniq_targets = list(dict.fromkeys(referenced))  # preserve order, drop dups
-        for ch in _sort_chunks(chunks):
         cid = _get(ch, "chunk_id", "id")
         if not cid:
             continue
-            for kind in defaults_kinds:
+        raw = _chunk_text_for_refs(ch)
-                for target in uniq_targets:
-                    edges.append(_make_edge(kind, cid, target, note_id, cid, f"edge_defaults:{note_type}:{kind}", 0.70))
-                    if kind in SYMMETRIC_KINDS:
-                        edges.append(_make_edge(kind, target, cid, note_id, cid, f"edge_defaults:{note_type}:{kind}", 0.70))
 
-    return edges, referenced
+        # 3a) typed inline relations
+        typed, remainder = _extract_typed_relations(raw)
+        for kind, target in typed:
+            kind = kind.strip().lower()
+            if not kind or not target:
+                continue
+            edges.append(_edge(kind, "chunk", cid, target, note_id, {
+                "chunk_id": cid,
+                "edge_id": _mk_edge_id(kind, cid, target, "chunk", "inline:rel"),
+                "provenance": "explicit",
+                "rule_id": "inline:rel",
+                "confidence": 0.95,
+            }))
+            if kind in {"related_to", "similar_to"}:
+                edges.append(_edge(kind, "chunk", target, cid, note_id, {
+                    "chunk_id": cid,
+                    "edge_id": _mk_edge_id(kind, target, cid, "chunk", "inline:rel"),
+                    "provenance": "explicit",
+                    "rule_id": "inline:rel",
+                    "confidence": 0.95,
+                }))
 
-# -----------------------------------------------------------------------------
+        # 3b) callouts
-# Public API (backward-compatible)
+        call_pairs, remainder2 = _extract_callout_relations(remainder)
-# -----------------------------------------------------------------------------
+        for kind, target in call_pairs:
+            k = (kind or "").strip().lower()
+            if not k or not target:
+                continue
+            edges.append(_edge(k, "chunk", cid, target, note_id, {
+                "chunk_id": cid,
+                "edge_id": _mk_edge_id(k, cid, target, "chunk", "callout:edge"),
+                "provenance": "explicit",
+                "rule_id": "callout:edge",
+                "confidence": 0.95,
+            }))
+            if k in {"related_to", "similar_to"}:
+                edges.append(_edge(k, "chunk", target, cid, note_id, {
+                    "chunk_id": cid,
+                    "edge_id": _mk_edge_id(k, target, cid, "chunk", "callout:edge"),
+                    "provenance": "explicit",
+                    "rule_id": "callout:edge",
+                    "confidence": 0.95,
+                }))
 
-def build_edges_for_note(*args, **kwargs) -> List[Dict[str, Any]]:
+        # 3c) generische Wikilinks → references (+ defaults je Ref)
-    """
+        refs = _extract_wikilinks(remainder2)
-    Backward-compatible entry point.
+        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"),
+                "provenance": "explicit",
+                "rule_id": "explicit:wikilink",
+                "confidence": 1.0,
+            }))
+            for rel in defaults:
+                if rel == "references":
+                    continue
+                edges.append(_edge(rel, "chunk", cid, r, note_id, {
+                    "chunk_id": cid,
+                    "edge_id": _mk_edge_id(rel, cid, r, "chunk", f"edge_defaults:{note_type}:{rel}"),
+                    "provenance": "rule",
+                    "rule_id": f"edge_defaults:{note_type}:{rel}",
+                    "confidence": 0.7,
+                }))
+                if rel in {"related_to", "similar_to"}:
+                    edges.append(_edge(rel, "chunk", r, cid, note_id, {
+                        "chunk_id": cid,
+                        "edge_id": _mk_edge_id(rel, r, cid, "chunk", f"edge_defaults:{note_type}:{rel}"),
+                        "provenance": "rule",
+                        "rule_id": f"edge_defaults:{note_type}:{rel}",
+                        "confidence": 0.7,
+                    }))
+        refs_all.extend(refs)
 
-    Supported call styles:
+    # 4) optional note-scope refs/backlinks (+ defaults)
-    1) build_edges_for_note(note_id, title, note_type, chunks, types_cfg)
+    if include_note_scope_refs:
-    2) build_edges_for_note(note_payload=dict, chunks=[...], types_cfg=dict)
+        refs_note = list(refs_all or [])
-       where note_payload contains: note_id/id, title, type
+        if note_level_references:
-    """
+            refs_note.extend([r for r in note_level_references if isinstance(r, str) and r])
-    # Detect signature form
+        refs_note = _dedupe_seq(refs_note)
-    if args and isinstance(args[0], dict) and ("title" in args[0] or "type" in args[0] or "note_id" in args[0] or "id" in args[0]):
+        for r in refs_note:
-        note = args[0]
+            edges.append(_edge("references", "note", note_id, r, note_id, {
-        chunks = args[1] if len(args) > 1 else kwargs.get("chunks", [])
+                "edge_id": _mk_edge_id("references", note_id, r, "note", "explicit:note_scope"),
-        types_cfg = args[2] if len(args) > 2 else kwargs.get("types_cfg")
+                "provenance": "explicit",
-        note_id = _get(note, "note_id", "id")
+                "rule_id": "explicit:note_scope",
-        title = _get(note, "title") or ""
+                "confidence": 1.0,
-        note_type = _get(note, "type")
+            }))
-    else:
+            edges.append(_edge("backlink", "note", r, note_id, note_id, {
-        # Positional legacy form
+                "edge_id": _mk_edge_id("backlink", r, note_id, "note", "derived:backlink"),
-        note_id = args[0] if len(args) > 0 else kwargs.get("note_id")
+                "provenance": "rule",
-        title = args[1] if len(args) > 1 else kwargs.get("title", "")
+                "rule_id": "derived:backlink",
-        note_type = args[2] if len(args) > 2 else kwargs.get("note_type")
+                "confidence": 0.9,
-        chunks = args[3] if len(args) > 3 else kwargs.get("chunks", [])
+            }))
-        types_cfg = args[4] if len(args) > 4 else kwargs.get("types_cfg")
+            for rel in defaults:
+                if rel == "references":
+                    continue
+                edges.append(_edge(rel, "note", note_id, r, note_id, {
+                    "edge_id": _mk_edge_id(rel, note_id, r, "note", f"edge_defaults:{note_type}:{rel}"),
+                    "provenance": "rule",
+                    "rule_id": f"edge_defaults:{note_type}:{rel}",
+                    "confidence": 0.7,
+                }))
+                if rel in {"related_to", "similar_to"}:
+                    edges.append(_edge(rel, "note", r, note_id, note_id, {
+                        "edge_id": _mk_edge_id(rel, r, note_id, "note", f"edge_defaults:{note_type}:{rel}"),
+                        "provenance": "rule",
+                        "rule_id": f"edge_defaults:{note_type}:{rel}",
+                        "confidence": 0.7,
+                    }))
 
-    chunks = list(chunks or [])
+    # 5) De-Dupe (source_id, target_id, relation, rule_id)
-
+    seen: Set[Tuple[str,str,str,str]] = set()
-    # Structure edges
+    out: List[dict] = []
-    edges = _structure_edges(note_id, title, chunks)
+    for e in edges:
-
+        s = str(e.get("source_id") or "")
-    # Explicit / inline / callout / defaults
+        t = str(e.get("target_id") or "")
-    more, _ = _explicit_and_inline_edges(note_id, title, note_type, chunks, types_cfg)
+        rel = str(e.get("relation") or e.get("kind") or "edge")
-    _append_with_dedupe(edges, more)
+        rule = str(e.get("rule_id") or "")
-
+        key = (s, t, rel, rule)
-    return edges
+        if key in seen:
+            continue
+        seen.add(key)
+        out.append(e)
+    return out