Bump version: 0.14.0 → 0.14.1-beta.0

Fixes https://github.com/lancedb/lancedb/issues/1750

Cargo.toml

@@ -22,13 +22,13 @@ categories = ["database-implementations"]
 [workspace.dependencies]
 lance = { "version" = "=0.18.3", "features" = [
     "dynamodb",
-]}
-lance-index = { "version" = "=0.18.3"}
-lance-linalg = { "version" = "=0.18.3"}
-lance-table = { "version" = "=0.18.3"}
-lance-testing = { "version" = "=0.18.3"}
-lance-datafusion = { "version" = "=0.18.3"}
-lance-encoding = { "version" = "=0.18.3"}
+], git = "https://github.com/lancedb/lance.git", tag = "v0.18.3-beta.2" }
+lance-index = { "version" = "=0.18.3", git = "https://github.com/lancedb/lance.git", tag = "v0.18.3-beta.2" }
+lance-linalg = { "version" = "=0.18.3", git = "https://github.com/lancedb/lance.git", tag = "v0.18.3-beta.2" }
+lance-table = { "version" = "=0.18.3", git = "https://github.com/lancedb/lance.git", tag = "v0.18.3-beta.2" }
+lance-testing = { "version" = "=0.18.3", git = "https://github.com/lancedb/lance.git", tag = "v0.18.3-beta.2" }
+lance-datafusion = { "version" = "=0.18.3", git = "https://github.com/lancedb/lance.git", tag = "v0.18.3-beta.2" }
+lance-encoding = { "version" = "=0.18.3", git = "https://github.com/lancedb/lance.git", tag = "v0.18.3-beta.2" }
 # Note that this one does not include pyarrow
 arrow = { version = "52.2", optional = false }
 arrow-array = "52.2"

python/.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.14.0"
+current_version = "0.14.1-beta.0"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

python/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-python"
-version = "0.14.0"
+version = "0.14.1-beta.0"
 edition.workspace = true
 description = "Python bindings for LanceDB"
 license.workspace = true

rust/lancedb/src/connection.rs

@@ -196,22 +196,6 @@ impl<T: IntoArrow> CreateTableBuilder<true, T> {
         };
         Ok((data, builder))
     }
-
-    pub fn add_embedding(mut self, definition: EmbeddingDefinition) -> Result<Self> {
-        // Early verification of the embedding name
-        let embedding_func = self
-            .parent
-            .embedding_registry()
-            .get(&definition.embedding_name)
-            .ok_or_else(|| Error::EmbeddingFunctionNotFound {
-                name: definition.embedding_name.clone(),
-                reason: "No embedding function found in the connection's embedding_registry"
-                    .to_string(),
-            })?;
-
-        self.embeddings.push((definition, embedding_func));
-        Ok(self)
-    }
 }
 
 // Builder methods that only apply when we do not have initial data
@@ -329,6 +313,26 @@ impl<const HAS_DATA: bool, T: IntoArrow> CreateTableBuilder<HAS_DATA, T> {
         };
         self
     }
+
+    /// Add an embedding definition to the table.
+    ///
+    /// The `embedding_name` must match the name of an embedding function that
+    /// was previously registered with the connection's [`EmbeddingRegistry`].
+    pub fn add_embedding(mut self, definition: EmbeddingDefinition) -> Result<Self> {
+        // Early verification of the embedding name
+        let embedding_func = self
+            .parent
+            .embedding_registry()
+            .get(&definition.embedding_name)
+            .ok_or_else(|| Error::EmbeddingFunctionNotFound {
+                name: definition.embedding_name.clone(),
+                reason: "No embedding function found in the connection's embedding_registry"
+                    .to_string(),
+            })?;
+
+        self.embeddings.push((definition, embedding_func));
+        Ok(self)
+    }
 }
 
 #[derive(Clone, Debug)]

rust/lancedb/src/index.rs

@@ -144,7 +144,7 @@ impl std::str::FromStr for IndexType {
             "BTREE" => Ok(Self::BTree),
             "BITMAP" => Ok(Self::Bitmap),
             "LABEL_LIST" | "LABELLIST" => Ok(Self::LabelList),
-            "FTS" => Ok(Self::FTS),
+            "FTS" | "INVERTED" => Ok(Self::FTS),
             "IVF_PQ" => Ok(Self::IvfPq),
             "IVF_HNSW_PQ" => Ok(Self::IvfHnswPq),
             "IVF_HNSW_SQ" => Ok(Self::IvfHnswSq),

rust/lancedb/src/table.rs

@@ -2110,7 +2110,6 @@ mod tests {
     use arrow_schema::{DataType, Field, Schema, TimeUnit};
     use futures::TryStreamExt;
     use lance::dataset::{Dataset, WriteMode};
-    use lance::index::DatasetIndexInternalExt;
     use lance::io::{ObjectStoreParams, WrappingObjectStore};
     use rand::Rng;
     use tempfile::tempdir;
@@ -3002,22 +3001,8 @@ mod tests {
         let index_configs = table.list_indices().await.unwrap();
         assert_eq!(index_configs.len(), 1);
         let index = index_configs.into_iter().next().unwrap();
-        // TODO: Fix via https://github.com/lancedb/lance/issues/2039
-        // assert_eq!(index.index_type, crate::index::IndexType::Bitmap);
+        assert_eq!(index.index_type, crate::index::IndexType::Bitmap);
         assert_eq!(index.columns, vec!["category".to_string()]);
-
-        // For now, just open the index to verify its type
-        let lance_dataset = table.as_native().unwrap().dataset.get().await.unwrap();
-        let indices = lance_dataset
-            .load_indices_by_name(&index.name)
-            .await
-            .unwrap();
-        let index_meta = &indices[0];
-        let idx = lance_dataset
-            .open_scalar_index("category", &index_meta.uuid.to_string())
-            .await
-            .unwrap();
-        assert_eq!(idx.index_type(), IndexType::Bitmap);
     }
 
     #[tokio::test]
@@ -3086,22 +3071,57 @@ mod tests {
         let index_configs = table.list_indices().await.unwrap();
         assert_eq!(index_configs.len(), 1);
         let index = index_configs.into_iter().next().unwrap();
-        // TODO: Fix via https://github.com/lancedb/lance/issues/2039
-        // assert_eq!(index.index_type, crate::index::IndexType::LabelList);
+        assert_eq!(index.index_type, crate::index::IndexType::LabelList);
         assert_eq!(index.columns, vec!["tags".to_string()]);
+    }
 
-        // For now, just open the index to verify its type
-        let lance_dataset = table.as_native().unwrap().dataset.get().await.unwrap();
-        let indices = lance_dataset
-            .load_indices_by_name(&index.name)
+    #[tokio::test]
+    async fn test_create_inverted_index() {
+        let tmp_dir = tempdir().unwrap();
+        let uri = tmp_dir.path().to_str().unwrap();
+
+        let conn = ConnectBuilder::new(uri).execute().await.unwrap();
+        const WORDS: [&str; 3] = ["cat", "dog", "fish"];
+        let mut text_builder = StringBuilder::new();
+        let num_rows = 120;
+        for i in 0..num_rows {
+            text_builder.append_value(WORDS[i % 3]);
+        }
+        let text = Arc::new(text_builder.finish());
+
+        let schema = Arc::new(Schema::new(vec![
+            Field::new("id", DataType::Int32, false),
+            Field::new("text", DataType::Utf8, true),
+        ]));
+        let batch = RecordBatch::try_new(
+            schema.clone(),
+            vec![
+                Arc::new(Int32Array::from_iter_values(0..num_rows as i32)),
+                text,
+            ],
+        )
+        .unwrap();
+
+        let table = conn
+            .create_table(
+                "test_bitmap",
+                RecordBatchIterator::new(vec![Ok(batch.clone())], batch.schema()),
+            )
+            .execute()
             .await
             .unwrap();
-        let index_meta = &indices[0];
-        let idx = lance_dataset
-            .open_scalar_index("tags", &index_meta.uuid.to_string())
+
+        table
+            .create_index(&["text"], Index::FTS(Default::default()))
+            .execute()
             .await
             .unwrap();
-        assert_eq!(idx.index_type(), IndexType::LabelList);
+        let index_configs = table.list_indices().await.unwrap();
+        assert_eq!(index_configs.len(), 1);
+        let index = index_configs.into_iter().next().unwrap();
+        assert_eq!(index.index_type, crate::index::IndexType::FTS);
+        assert_eq!(index.columns, vec!["text".to_string()]);
+        assert_eq!(index.name, "text_idx");
     }
 
     #[tokio::test]