Normalize best-match queries

The query is trimmed and a single pair of wrapping quotes is stripped
-- they carry no phrase semantics, since the whole query embeds as one
string. A query that normalizes to nothing (e.g., just quotes) is
treated as no search at all rather than being scored against noise.

chrome/content/zotero/xpcom/collectionTreeRow.js

@@ -718,7 +718,7 @@ Zotero.CollectionTreeRow.prototype.getBestMatchSource = function () {
 	}
 	// An active best-match quick search overrides a selected saved search's
 	// own marker, so the user's typed query wins
-	if (this.searchText
+	if (this.searchText && Zotero.Embeddings.normalizeQuery(this.searchText)
 			&& (this.searchMode || Zotero.Prefs.get('search.quicksearch-mode')) === 'bestMatch') {
 		return false;
 	}
@@ -742,7 +742,7 @@ Zotero.CollectionTreeRow.prototype.isBestMatchSearch = function () {
 	if (this.getBestMatchSource()) {
 		return true;
 	}
-	return !!this.searchText
+	return !!(this.searchText && Zotero.Embeddings.normalizeQuery(this.searchText))
 		&& (this.searchMode || Zotero.Prefs.get('search.quicksearch-mode')) === 'bestMatch';
 };
 

chrome/content/zotero/xpcom/data/search.js

@@ -881,7 +881,8 @@ Zotero.Search.prototype.getBestMatchQuery = function () {
 		else if (condition.condition == 'groupEnd') {
 			depth--;
 		}
-		else if (depth == 0 && condition.condition == 'bestMatch' && condition.value) {
+		else if (depth == 0 && condition.condition == 'bestMatch' && condition.value
+				&& Zotero.Embeddings.normalizeQuery(condition.value)) {
 			return {
 				query: condition.value,
 				topK: parseInt(condition.operator) || false

chrome/content/zotero/xpcom/embeddings.js

@@ -685,6 +685,19 @@ Zotero.Embeddings = new function () {
 	// search triggers (per-row membership cutoffs plus the merged ranking)
 	let _queryCache = null;
 
+	/**
+	 * Normalize a best-match query: trim whitespace and strip a single pair
+	 * of wrapping quotes, which carry no phrase semantics here -- the whole
+	 * query embeds as one string. A query that normalizes to an empty string
+	 * is no query at all, and callers treat it as no active search.
+	 *
+	 * @param {String} text
+	 * @return {String}
+	 */
+	this.normalizeQuery = function (text) {
+		return text.trim().replace(/^"(.*)"$/s, '$1').trim();
+	};
+
 	/**
 	 * Embed a search query string, applying the active model's query prefix.
 	 *
@@ -692,6 +705,13 @@ Zotero.Embeddings = new function () {
 	 * @return {Promise<Float32Array>}
 	 */
 	this.embedQuery = function (text) {
+		text = this.normalizeQuery(text);
+		// Callers treat a query that normalizes to nothing as no search at
+		// all, so it should never get this far -- embedding just the model's
+		// query prefix would rank against noise
+		if (!text) {
+			throw new Error("Empty best-match query");
+		}
 		let modelVersion = this.getModelVersion();
 		if (_queryCache && _queryCache.modelVersion === modelVersion
 				&& _queryCache.text === text) {

test/tests/embeddingsTest.js

@@ -87,6 +87,31 @@ describe("Zotero.Embeddings", function () {
 			}
 		});
 
+		it("should strip a single pair of wrapping quotes", async function () {
+			let embedStub = sinon.stub(Zotero.Embeddings, 'embed').resolves(new Float32Array([1]));
+			let stubs = [
+				sinon.stub(Zotero.Embeddings.Indexing, 'startIndexing').resolves(),
+				sinon.stub(Zotero.Embeddings, 'pruneModels').resolves(),
+				embedStub
+			];
+			Zotero.Prefs.set('embeddings.model', 'bge-small-en-v1.5');
+			try {
+				await Zotero.Embeddings.Indexing.waitForPendingModelSwitch();
+				// Whitespace around the quotes doesn't defeat the stripping
+				await Zotero.Embeddings.embedQuery(' "wrapped query" ');
+				assert.include(embedStub.firstCall.args[0], 'wrapped query');
+				assert.notInclude(embedStub.firstCall.args[0], '"');
+				// A query that normalizes to nothing is a caller bug
+				assert.throws(() => Zotero.Embeddings.embedQuery('""'));
+			}
+			finally {
+				Zotero.Prefs.set('embeddings.model', '');
+				await Zotero.Embeddings.Indexing.waitForPendingModelSwitch();
+				Zotero.Prefs.clear('embeddings.indexingPaused');
+				stubs.forEach(stub => stub.restore());
+			}
+		});
+
 		it("should share one in-flight embed across concurrent calls", async function () {
 			let deferred = Zotero.Promise.defer();
 			let stubs = [

test/tests/searchTest.js

@@ -234,6 +234,14 @@ describe("Zotero.Search", function () {
 			assert.deepEqual(s.getBestMatchQuery(), { query: 'some query', topK: false });
 			assert.sameMembers(await s.search(), [itemA.id, itemB.id]);
 
+			// A query that normalizes to nothing (e.g., just quotes) is no
+			// query at all
+			let sEmpty = new Zotero.Search();
+			sEmpty.libraryID = itemA.libraryID;
+			sEmpty.addCondition('resultLevel', 'item');
+			sEmpty.addCondition('bestMatch', 'contains', '""');
+			assert.isFalse(sEmpty.getBestMatchQuery());
+
 			// With a cutoff, membership is the K most similar
 			let stub = sinon.stub(Zotero.Embeddings, 'scoreItemIDs').callsFake(
 				async (query, itemIDs) => new Map(itemIDs.map(id => [id, id == itemB.id ? 0.9 : 0.5]))