@o-lang/semantic-doc-search 1.0.31 → 1.0.33

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@o-lang/semantic-doc-search",
-  "version": "1.0.31",
+  "version": "1.0.33",
   "description": "O-lang Semantic Document Search Resolver with hybrid search, embeddings, rerank, and streaming.",
   "main": "src/index.js",
   "type": "commonjs",

package/src/resolver.js

@@ -1,5 +1,5 @@
 const VectorRouter = require("./adapters/vectorRouter");
-const embedder = require("./embeddings/local"); // ✅ singleton embedder
+const embedder = require("./embeddings/local"); // singleton embedder
 const { extractQuery } = require("./utils/extractQuery");
 const { formatResults } = require("./utils/formatResults");
 const fs = require("fs");
@@ -28,6 +28,15 @@ function hashContent(str) {
   return crypto.createHash("sha256").update(str).digest("hex");
 }
 
+/**
+ * Clean text for embedding (defensive)
+ */
+function sanitizeTextForEmbedding(text) {
+  if (typeof text !== "string") return "";
+  // Remove wrapping quotes and extra whitespace
+  return text.replace(/^["']|["']$/g, "").trim();
+}
+
 /**
  * Semantic Doc Search Resolver
  */
@@ -35,7 +44,9 @@ async function resolver(action, context = {}) {
   if (typeof action !== "string") return;
   if (!action.toLowerCase().startsWith("ask doc-search")) return;
 
-  const query = extractQuery(action);
+  let query = extractQuery(action);
+  query = sanitizeTextForEmbedding(query);
+  if (!query) return { text: "(Empty query)", meta: { matches: 0 } };
 
   // Vector backend
   const vectorStore = VectorRouter.create(context);
@@ -62,17 +73,14 @@ async function resolver(action, context = {}) {
     for (const doc of context.documents) {
       const chunks = doc.chunks || [doc.content];
       for (let i = 0; i < chunks.length; i++) {
-        const text = chunks[i];
+        const text = sanitizeTextForEmbedding(chunks[i]);
+        if (!text) continue;
+
         const hash = hashContent(text);
         if (cache[hash]) continue; // Skip already ingested
 
         const vector = await embedder.embed(text);
-
-        // ✅ Defensive: skip invalid vectors
-        if (!Array.isArray(vector) || vector.length !== embedder.getDimension() || vector.every(v => v === 0)) {
-          console.warn(`⚠️ Skipping invalid vector for doc ${doc.id}:${i}`);
-          continue;
-        }
+        if (!vector || vector.every(v => v === 0)) continue;
 
         await vectorStore.upsert({
           id: `${doc.id}:${i}`,
@@ -89,12 +97,12 @@ async function resolver(action, context = {}) {
 
   // Embed query & search
   const queryVector = await embedder.embed(query);
-
-  // ✅ Defensive: validate query vector
-  if (!Array.isArray(queryVector) || queryVector.length !== embedder.getDimension()) {
-    throw new Error("Query embedding invalid or not a proper array");
+  if (!queryVector || queryVector.every(v => v === 0)) {
+    console.warn("⚠️ Query embedding invalid");
+    return { text: "(Query could not be embedded)", meta: { matches: 0 } };
   }
 
+  // Top-K + similarity threshold
   const results = await vectorStore.query({
     vector: queryVector,
     topK: context.topK || 5,