Preserve wikilink text in Obsidian indexing and fix duplicate sync

Two fixes:
- Convert wikilinks to display text instead of stripping them entirely.
  [[Noah]] becomes "Noah", [[target|display]] becomes "display". This
  was causing names and references in wikilinks to be invisible to search.
- Switch _get_obsidian_indexed_files to async engine to avoid stale reads
  from the separate sync engine, which caused files to be re-indexed
  every cycle.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

blueprints/rag/logic.py

@@ -10,6 +10,7 @@ from langchain_openai import OpenAIEmbeddings
 from langchain_postgres import PGVector
 from langchain_text_splitters import RecursiveCharacterTextSplitter
 from sqlalchemy import create_engine, text
+from sqlalchemy.ext.asyncio import create_async_engine
 
 from .fetchers import PaperlessNGXService
 from utils.obsidian_service import ObsidianService
@@ -63,6 +64,13 @@ def _get_engine():
     return _get_engine._engine
 
 
+def _get_async_engine():
+    """Get an async SQLAlchemy engine for direct queries."""
+    if not hasattr(_get_async_engine, "_engine"):
+        _get_async_engine._engine = create_async_engine(_pgvector_url)
+    return _get_async_engine._engine
+
+
 text_splitter = RecursiveCharacterTextSplitter(
     chunk_size=1000,  # chunk size (characters)
     chunk_overlap=200,  # chunk overlap (characters)
@@ -233,14 +241,14 @@ async def index_obsidian_documents():
     return {"indexed": len(documents)}
 
 
-def _get_obsidian_indexed_files() -> dict[str, float]:
+async def _get_obsidian_indexed_files() -> dict[str, float]:
     """Return {filepath: indexed_at} for all obsidian chunks in pgvector."""
     collection_id = _get_collection_id()
     if not collection_id:
         return {}
-    engine = _get_engine()
+    engine = _get_async_engine()
-    with engine.connect() as conn:
+    async with engine.connect() as conn:
-        result = conn.execute(
+        result = await conn.execute(
             text(
                 "SELECT DISTINCT cmetadata->>'filepath' AS filepath, "
                 "MAX((cmetadata->>'indexed_at')::float) AS indexed_at "
@@ -263,7 +271,7 @@ async def sync_obsidian_documents() -> dict[str, int]:
         Dict with counts of added, updated, and deleted files.
     """
     obsidian_service = ObsidianService()
-    indexed_files = _get_obsidian_indexed_files()
+    indexed_files = await _get_obsidian_indexed_files()
 
     # Build map of current vault files -> mtime
     vault_files: dict[str, float] = {}

utils/obsidian_service.py

@@ -106,8 +106,12 @@ class ObsidianService:
         embeds = [e.split(":")[0].strip() if ":" in e else e.strip() for e in embeds]
 
         # Clean body content
-        # Remove wikilinks [[...]] and embeds [[!...]]
+        # Remove embeds ![[...]]
-        cleaned_content = re.sub(r"\[\[.*?\]\]", "", body_content)
+        cleaned_content = re.sub(r"!\[\[.*?\]\]", "", body_content)
+        # Convert wikilinks to display text: [[target|display]] → display, [[target]] → target
+        cleaned_content = re.sub(
+            r"\[\[([^\]|]+\|)?([^\]]+)\]\]", r"\2", cleaned_content
+        )
         cleaned_content = re.sub(r"\n{3,}", "\n\n", cleaned_content).strip()
 
         return {