@lancedb/lancedb 0.20.1-beta.1 → 0.21.0

package/dist/arrow.js

@@ -682,8 +682,14 @@ async function applyEmbeddingsFromMetadata(table, schema) {
         if (sourceColumn === undefined) {
             throw new Error(`Cannot apply embedding function because the source column '${functionEntry.sourceColumn}' was not present in the data`);
         }
+        // Check if destination column exists and handle accordingly
         if (columns[destColumn] !== undefined) {
-            throw new Error(`Attempt to apply embeddings to table failed because column ${destColumn} already existed`);
+            const existingColumn = columns[destColumn];
+            // If the column exists but is all null, we can fill it with embeddings
+            if (existingColumn.nullCount !== existingColumn.length) {
+                // Column has non-null values, skip embedding application
+                continue;
+            }
         }
         if (table.batches.length > 1) {
             throw new Error("Internal error: `makeArrowTable` unexpectedly created a table with more than one batch");
@@ -765,8 +771,18 @@ async function applyEmbeddings(table, embeddings, schema) {
         }
     }
     else {
+        // Check if destination column exists and handle accordingly
         if (Object.prototype.hasOwnProperty.call(newColumns, destColumn)) {
-            throw new Error(`Attempt to apply embeddings to table failed because column ${destColumn} already existed`);
+            const existingColumn = newColumns[destColumn];
+            // If the column exists but is all null, we can fill it with embeddings
+            if (existingColumn.nullCount !== existingColumn.length) {
+                // Column has non-null values, skip embedding application and return table as-is
+                let newTable = new apache_arrow_1.Table(newColumns);
+                if (schema != null) {
+                    newTable = alignTable(newTable, schema);
+                }
+                return new apache_arrow_1.Table(new apache_arrow_1.Schema(newTable.schema.fields, schemaMetadata), newTable.batches);
+            }
         }
         if (table.batches.length > 1) {
             throw new Error("Internal error: `makeArrowTable` unexpectedly created a table with more than one batch");

package/dist/native.d.ts

@@ -381,7 +381,7 @@ export class VectorQuery {
   analyzePlan(): Promise<string>
 }
 export class JsFullTextQuery {
-  static matchQuery(query: string, column: string, boost: number, fuzziness: number | undefined | null, maxExpansions: number, operator: string): JsFullTextQuery
+  static matchQuery(query: string, column: string, boost: number, fuzziness: number | undefined | null, maxExpansions: number, operator: string, prefixLength: number): JsFullTextQuery
   static phraseQuery(query: string, column: string, slop: number): JsFullTextQuery
   static boostQuery(positive: JsFullTextQuery, negative: JsFullTextQuery, negativeBoost?: number | undefined | null): JsFullTextQuery
   static multiMatchQuery(query: string, columns: Array<string>, boosts: Array<number> | undefined | null, operator: string): JsFullTextQuery

package/dist/query.d.ts

@@ -445,10 +445,12 @@ export declare enum Operator {
  *
  * - `Must`: The term must be present in the document.
  * - `Should`: The term should contribute to the document score, but is not required.
+ * - `MustNot`: The term must not be present in the document.
  */
 export declare enum Occur {
+    Should = "SHOULD",
     Must = "MUST",
-    Should = "SHOULD"
+    MustNot = "MUST_NOT"
 }
 /**
  * Represents a full-text query interface.
@@ -481,12 +483,14 @@ export declare class MatchQuery implements FullTextQuery {
      *   - `fuzziness`: The fuzziness level for the query (default is 0).
      *   - `maxExpansions`: The maximum number of terms to consider for fuzzy matching (default is 50).
      *   - `operator`: The logical operator to use for combining terms in the query (default is "OR").
+     *   - `prefixLength`: The number of beginning characters being unchanged for fuzzy matching.
      */
     constructor(query: string, column: string, options?: {
         boost?: number;
         fuzziness?: number;
         maxExpansions?: number;
         operator?: Operator;
+        prefixLength?: number;
     });
     queryType(): FullTextQueryType;
 }

package/dist/query.js

@@ -708,11 +708,13 @@ var Operator;
  *
  * - `Must`: The term must be present in the document.
  * - `Should`: The term should contribute to the document score, but is not required.
+ * - `MustNot`: The term must not be present in the document.
  */
 var Occur;
 (function (Occur) {
-    Occur["Must"] = "MUST";
     Occur["Should"] = "SHOULD";
+    Occur["Must"] = "MUST";
+    Occur["MustNot"] = "MUST_NOT";
 })(Occur || (exports.Occur = Occur = {}));
 // biome-ignore lint/suspicious/noExplicitAny: we want any here
 function instanceOfFullTextQuery(obj) {
@@ -731,13 +733,14 @@ class MatchQuery {
      *   - `fuzziness`: The fuzziness level for the query (default is 0).
      *   - `maxExpansions`: The maximum number of terms to consider for fuzzy matching (default is 50).
      *   - `operator`: The logical operator to use for combining terms in the query (default is "OR").
+     *   - `prefixLength`: The number of beginning characters being unchanged for fuzzy matching.
      */
     constructor(query, column, options) {
         let fuzziness = options?.fuzziness;
         if (fuzziness === undefined) {
             fuzziness = 0;
         }
-        this.inner = native_1.JsFullTextQuery.matchQuery(query, column, options?.boost ?? 1.0, fuzziness, options?.maxExpansions ?? 50, options?.operator ?? Operator.Or);
+        this.inner = native_1.JsFullTextQuery.matchQuery(query, column, options?.boost ?? 1.0, fuzziness, options?.maxExpansions ?? 50, options?.operator ?? Operator.Or, options?.prefixLength ?? 0);
     }
     queryType() {
         return FullTextQueryType.Match;

package/package.json

@@ -11,7 +11,7 @@
     "ann"
   ],
   "private": false,
-  "version": "0.20.1-beta.1",
+  "version": "0.21.0",
   "main": "dist/index.js",
   "exports": {
     ".": "./dist/index.js",
@@ -100,14 +100,14 @@
     "reflect-metadata": "^0.2.2"
   },
   "optionalDependencies": {
-    "@lancedb/lancedb-darwin-x64": "0.20.1-beta.1",
-    "@lancedb/lancedb-darwin-arm64": "0.20.1-beta.1",
-    "@lancedb/lancedb-linux-x64-gnu": "0.20.1-beta.1",
-    "@lancedb/lancedb-linux-arm64-gnu": "0.20.1-beta.1",
-    "@lancedb/lancedb-linux-x64-musl": "0.20.1-beta.1",
-    "@lancedb/lancedb-linux-arm64-musl": "0.20.1-beta.1",
-    "@lancedb/lancedb-win32-x64-msvc": "0.20.1-beta.1",
-    "@lancedb/lancedb-win32-arm64-msvc": "0.20.1-beta.1"
+    "@lancedb/lancedb-darwin-x64": "0.21.0",
+    "@lancedb/lancedb-darwin-arm64": "0.21.0",
+    "@lancedb/lancedb-linux-x64-gnu": "0.21.0",
+    "@lancedb/lancedb-linux-arm64-gnu": "0.21.0",
+    "@lancedb/lancedb-linux-x64-musl": "0.21.0",
+    "@lancedb/lancedb-linux-arm64-musl": "0.21.0",
+    "@lancedb/lancedb-win32-x64-msvc": "0.21.0",
+    "@lancedb/lancedb-win32-arm64-msvc": "0.21.0"
   },
   "peerDependencies": {
     "apache-arrow": ">=15.0.0 <=18.1.0"