[python] Bump version: 0.3.1 → 0.3.2

Closes #594

The embedding functions are pydantic models so multiple instances with
the same parameters are considered ==, which means that if you have
multiple embedding columns it's possible for the embeddings to get
overwritten. Instead we use `is` instead of == to avoid this problem.

testing: modified unit test to include this case

Cargo.toml

@@ -5,9 +5,9 @@ exclude = ["python"]
 resolver = "2"
 
 [workspace.dependencies]
-lance = { "version" = "=0.8.6", "features" = ["dynamodb"] }
-lance-linalg = { "version" = "=0.8.6" }
-lance-testing = { "version" = "=0.8.6" }
+lance = { "version" = "=0.8.7", "features" = ["dynamodb"] }
+lance-linalg = { "version" = "=0.8.7" }
+lance-testing = { "version" = "=0.8.7" }
 # Note that this one does not include pyarrow
 arrow = { version = "47.0.0", optional = false }
 arrow-array = "47.0"
@@ -18,7 +18,7 @@ arrow-schema = "47.0"
 arrow-arith = "47.0"
 arrow-cast = "47.0"
 chrono = "0.4.23"
-half = { "version" = "=2.2.1", default-features = false, features = [
+half = { "version" = "=2.3.1", default-features = false, features = [
     "num-traits"
 ] }
 log = "0.4"

docs/mkdocs.yml

@@ -73,12 +73,14 @@ nav:
     - Vector Search: search.md
     - SQL filters: sql.md
     - Indexing: ann_indexes.md
+    - Versioning & Reproducibility: notebooks/reproducibility.ipynb
   - 🧬 Embeddings:
     - embeddings/index.md
     - Ingest Embedding Functions: embeddings/embedding_functions.md
     - Available Functions: embeddings/default_embedding_functions.md
     - Create Custom Embedding Functions: embeddings/api.md
-    - Example- MultiModal CLIP Embeddings: notebooks/DisappearingEmbeddingFunction.ipynb
+    - Example - Multi-lingual semantic search: notebooks/multi_lingual_example.ipynb
+    - Example - MultiModal CLIP Embeddings: notebooks/DisappearingEmbeddingFunction.ipynb
   - 🔍 Python full-text search: fts.md
   - 🔌 Integrations:
     - integrations/index.md
@@ -110,12 +112,14 @@ nav:
   - Vector Search: search.md
   - SQL filters: sql.md
   - Indexing: ann_indexes.md
+  - Versioning & Reproducibility: notebooks/reproducibility.ipynb
 - Embeddings:
   - embeddings/index.md
   - Ingest Embedding Functions: embeddings/embedding_functions.md
   - Available Functions: embeddings/default_embedding_functions.md
   - Create Custom Embedding Functions: embeddings/api.md
-  - Example- MultiModal CLIP Embeddings: notebooks/DisappearingEmbeddingFunction.ipynb
+  - Example - Multi-lingual semantic search: notebooks/multi_lingual_example.ipynb
+  - Example - MultiModal CLIP Embeddings: notebooks/DisappearingEmbeddingFunction.ipynb
 - Python full-text search: fts.md
 - Integrations:
   - integrations/index.md

docs/src/notebooks/DisappearingEmbeddingFunction.ipynb

@@ -1,5 +1,13 @@
 {
   "cells": [
+    {
+      "cell_type": "markdown",
+      "id": "88c1af18",
+      "metadata": {},
+      "source": [
+        "# Example - MultiModal CLIP Embeddings"
+      ]
+    },
     {
       "cell_type": "markdown",
       "id": "c6b5d346-2c2a-4341-a132-00e53543f8d1",

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.3.2",
+  "version": "0.3.3",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.3.2",
+      "version": "0.3.3",
       "cpu": [
         "x64",
         "arm64"
@@ -53,11 +53,11 @@
         "uuid": "^9.0.0"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.3.2",
-        "@lancedb/vectordb-darwin-x64": "0.3.2",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.3.2",
-        "@lancedb/vectordb-linux-x64-gnu": "0.3.2",
-        "@lancedb/vectordb-win32-x64-msvc": "0.3.2"
+        "@lancedb/vectordb-darwin-arm64": "0.3.3",
+        "@lancedb/vectordb-darwin-x64": "0.3.3",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.3.3",
+        "@lancedb/vectordb-linux-x64-gnu": "0.3.3",
+        "@lancedb/vectordb-win32-x64-msvc": "0.3.3"
       }
     },
     "node_modules/@apache-arrow/ts": {
@@ -317,9 +317,9 @@
       }
     },
     "node_modules/@lancedb/vectordb-darwin-arm64": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.2.tgz",
-      "integrity": "sha512-CDh+sU2k4xVfWauwDZnybma8AJ+Q2i0SzHg05BwgDcani7I0k60NjJ5GobpgQ38xOiEmwHllES1xs4NRh+1YkA==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.3.tgz",
+      "integrity": "sha512-nvyj7xNX2/wb/PH5TjyhLR/NQ1jVuoBw2B5UaSg7qf8Tnm5SSXWQ7F25RVKcKwh72fz1qB+CWW24ftZnRzbT/Q==",
       "cpu": [
         "arm64"
       ],
@@ -329,9 +329,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-darwin-x64": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.2.tgz",
-      "integrity": "sha512-xevyA+M/UE8ttaNkx68AyIUKlyWMhIzOECx0hbyN1zfShJe2UcunQcmbM1NxUi7EywodByyiP7bfMI1ZR1Y4Mw==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.3.tgz",
+      "integrity": "sha512-7CW+nILyPHp6cua0Rl0xaTDWw/vajEn/jCsEjFYgDmE+rtf5Z5Fum41FxR9C2TtIAvUK+nWb5mkYeOLqU6vRvg==",
       "cpu": [
         "x64"
       ],
@@ -341,9 +341,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.2.tgz",
-      "integrity": "sha512-mSKkQ/p6UTSLwWzfZMBS7wA6Gf335KljXLaOhdT4TUI/jC6e9/cvZKkXRgpdE9/gvfl4/WVzKY7sg3+azDYQ+A==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.3.tgz",
+      "integrity": "sha512-MmhwbacKxZPkLwwOqysVY8mUb8lFoyFIPlYhSLV4xS1C8X4HWALljIul1qMl1RYudp9Uc3PsOzRexl+OvCGfUw==",
       "cpu": [
         "arm64"
       ],
@@ -353,9 +353,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.2.tgz",
-      "integrity": "sha512-S1D0VwdidwyfIKE58t94rD+EEb5B64ORMVkTw5FBZJirShkk82+0G9H3jNgWrRMt1PB3Qn1286/wqDLQ9+fTsA==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.3.tgz",
+      "integrity": "sha512-OrNlsKi/QPw59Po040oRKn8IuqFEk4upc/4FaFKqVkcmQjjZrMg5Kgy9ZfWIhHdAnWXXggZZIPArpt0X1B0ceA==",
       "cpu": [
         "x64"
       ],
@@ -365,9 +365,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.2.tgz",
-      "integrity": "sha512-tnct1hf9GAlMchhYU6Lqmbm2nUKPO8apS7tuTIiucQh6gx+vbHmFZHFNHhw1AUJTpsj/eH2Z9iNayuC5Scdvhw==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.3.tgz",
+      "integrity": "sha512-lIT0A7a6eqX51IfGyhECtpXXgsr//kgbd+HZbcCdPy2GMmNezSch/7V22zExDSpF32hX8WfgcTLYCVWVilggDQ==",
       "cpu": [
         "x64"
       ],
@@ -4869,33 +4869,33 @@
       }
     },
     "@lancedb/vectordb-darwin-arm64": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.2.tgz",
-      "integrity": "sha512-CDh+sU2k4xVfWauwDZnybma8AJ+Q2i0SzHg05BwgDcani7I0k60NjJ5GobpgQ38xOiEmwHllES1xs4NRh+1YkA==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.3.3.tgz",
+      "integrity": "sha512-nvyj7xNX2/wb/PH5TjyhLR/NQ1jVuoBw2B5UaSg7qf8Tnm5SSXWQ7F25RVKcKwh72fz1qB+CWW24ftZnRzbT/Q==",
       "optional": true
     },
     "@lancedb/vectordb-darwin-x64": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.2.tgz",
-      "integrity": "sha512-xevyA+M/UE8ttaNkx68AyIUKlyWMhIzOECx0hbyN1zfShJe2UcunQcmbM1NxUi7EywodByyiP7bfMI1ZR1Y4Mw==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.3.3.tgz",
+      "integrity": "sha512-7CW+nILyPHp6cua0Rl0xaTDWw/vajEn/jCsEjFYgDmE+rtf5Z5Fum41FxR9C2TtIAvUK+nWb5mkYeOLqU6vRvg==",
       "optional": true
     },
     "@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.2.tgz",
-      "integrity": "sha512-mSKkQ/p6UTSLwWzfZMBS7wA6Gf335KljXLaOhdT4TUI/jC6e9/cvZKkXRgpdE9/gvfl4/WVzKY7sg3+azDYQ+A==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.3.3.tgz",
+      "integrity": "sha512-MmhwbacKxZPkLwwOqysVY8mUb8lFoyFIPlYhSLV4xS1C8X4HWALljIul1qMl1RYudp9Uc3PsOzRexl+OvCGfUw==",
       "optional": true
     },
     "@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.2.tgz",
-      "integrity": "sha512-S1D0VwdidwyfIKE58t94rD+EEb5B64ORMVkTw5FBZJirShkk82+0G9H3jNgWrRMt1PB3Qn1286/wqDLQ9+fTsA==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.3.3.tgz",
+      "integrity": "sha512-OrNlsKi/QPw59Po040oRKn8IuqFEk4upc/4FaFKqVkcmQjjZrMg5Kgy9ZfWIhHdAnWXXggZZIPArpt0X1B0ceA==",
       "optional": true
     },
     "@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.3.2",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.2.tgz",
-      "integrity": "sha512-tnct1hf9GAlMchhYU6Lqmbm2nUKPO8apS7tuTIiucQh6gx+vbHmFZHFNHhw1AUJTpsj/eH2Z9iNayuC5Scdvhw==",
+      "version": "0.3.3",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.3.3.tgz",
+      "integrity": "sha512-lIT0A7a6eqX51IfGyhECtpXXgsr//kgbd+HZbcCdPy2GMmNezSch/7V22zExDSpF32hX8WfgcTLYCVWVilggDQ==",
       "optional": true
     },
     "@neon-rs/cli": {

node/src/integration_test/test.ts

@@ -65,8 +65,8 @@ describe('LanceDB Mirrored Store Integration test', function () {
     const mirroredPath = path.join(dir, `${tableName}.lance`)
     fs.readdir(mirroredPath, { withFileTypes: true }, (err, files) => {
       if (err != null) throw err
-      // there should be two dirs
-      assert.equal(files.length, 2)
+      // there should be three dirs
+      assert.equal(files.length, 3)
       assert.isTrue(files[0].isDirectory())
       assert.isTrue(files[1].isDirectory())
 
@@ -76,6 +76,12 @@ describe('LanceDB Mirrored Store Integration test', function () {
         assert.isTrue(files[0].name.endsWith('.txn'))
       })
 
+      fs.readdir(path.join(mirroredPath, '_versions'), { withFileTypes: true }, (err, files) => {
+        if (err != null) throw err
+        assert.equal(files.length, 1)
+        assert.isTrue(files[0].name.endsWith('.manifest'))
+      })
+
       fs.readdir(path.join(mirroredPath, 'data'), { withFileTypes: true }, (err, files) => {
         if (err != null) throw err
         assert.equal(files.length, 1)
@@ -88,8 +94,8 @@ describe('LanceDB Mirrored Store Integration test', function () {
 
     fs.readdir(mirroredPath, { withFileTypes: true }, (err, files) => {
       if (err != null) throw err
-      // there should be two dirs
-      assert.equal(files.length, 3)
+      // there should be four dirs
+      assert.equal(files.length, 4)
       assert.isTrue(files[0].isDirectory())
       assert.isTrue(files[1].isDirectory())
       assert.isTrue(files[2].isDirectory())
@@ -128,12 +134,13 @@ describe('LanceDB Mirrored Store Integration test', function () {
 
     fs.readdir(mirroredPath, { withFileTypes: true }, (err, files) => {
       if (err != null) throw err
-      // there should be two dirs
-      assert.equal(files.length, 4)
+      // there should be five dirs
+      assert.equal(files.length, 5)
       assert.isTrue(files[0].isDirectory())
       assert.isTrue(files[1].isDirectory())
       assert.isTrue(files[2].isDirectory())
       assert.isTrue(files[3].isDirectory())
+      assert.isTrue(files[4].isDirectory())
 
       // Three TXs now
       fs.readdir(path.join(mirroredPath, '_transactions'), { withFileTypes: true }, (err, files) => {

python/.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.3.1
+current_version = 0.3.2
 commit = True
 message = [python] Bump version: {current_version} → {new_version}
 tag = True

python/lancedb/pydantic.py

@@ -327,7 +327,12 @@ class LanceModel(pydantic.BaseModel):
         for vec, func in vec_and_function:
             for source, field_info in cls.safe_get_fields().items():
                 src_func = get_extras(field_info, "source_column_for")
-                if src_func == func:
+                if src_func is func:
+                    # note we can't use == here since the function is a pydantic
+                    # model so two instances of the same function are ==, so if you
+                    # have multiple vector columns from multiple sources, both will
+                    # be mapped to the same source column
+                    # GH594
                     configs.append(
                         EmbeddingFunctionConfig(
                             source_column=source, vector_column=vec, function=func

python/pyproject.toml

@@ -1,9 +1,9 @@
 [project]
 name = "lancedb"
-version = "0.3.1"
+version = "0.3.2"
 dependencies = [
     "deprecation",
-    "pylance==0.8.6",
+    "pylance==0.8.7",
     "ratelimiter~=1.0",
     "retry>=0.9.2",
     "tqdm>=4.1.0",

python/tests/test_embeddings_slow.py

@@ -33,10 +33,13 @@ def test_sentence_transformer(alias, tmp_path):
     db = lancedb.connect(tmp_path)
     registry = get_registry()
     func = registry.get(alias).create()
+    func2 = registry.get(alias).create()
 
     class Words(LanceModel):
         text: str = func.SourceField()
+        text2: str = func2.SourceField()
         vector: Vector(func.ndims()) = func.VectorField()
+        vector2: Vector(func2.ndims()) = func2.VectorField()
 
     table = db.create_table("words", schema=Words)
     table.add(
@@ -50,7 +53,16 @@ def test_sentence_transformer(alias, tmp_path):
                     "foo",
                     "bar",
                     "baz",
-                ]
+                ],
+                "text2": [
+                    "to be or not to be",
+                    "that is the question",
+                    "for whether tis nobler",
+                    "in the mind to suffer",
+                    "the slings and arrows",
+                    "of outrageous fortune",
+                    "or to take arms",
+                ],
             }
         )
     )
@@ -62,6 +74,13 @@ def test_sentence_transformer(alias, tmp_path):
     expected = table.search(vec).limit(1).to_pydantic(Words)[0]
     assert actual.text == expected.text
     assert actual.text == "hello world"
+    assert not np.allclose(actual.vector, actual.vector2)
+
+    actual = (
+        table.search(query, vector_column_name="vector2").limit(1).to_pydantic(Words)[0]
+    )
+    assert actual.text != "hello world"
+    assert not np.allclose(actual.vector, actual.vector2)
 
 
 @pytest.mark.slow

rust/vectordb/src/io/object_store.rs

@@ -57,7 +57,7 @@ trait PrimaryOnly {
 
 impl PrimaryOnly for Path {
     fn primary_only(&self) -> bool {
-        self.to_string().contains("manifest")
+        self.filename().unwrap_or("") == "_latest.manifest"
     }
 }
 
@@ -118,8 +118,10 @@ impl ObjectStore for MirroringObjectStore {
         self.primary.head(location).await
     }
 
-    // garbage collection on secondary will happen async from other means
     async fn delete(&self, location: &Path) -> Result<()> {
+        if !location.primary_only() {
+            self.secondary.delete(location).await?;
+        }
         self.primary.delete(location).await
     }
 
@@ -132,7 +134,7 @@ impl ObjectStore for MirroringObjectStore {
     }
 
     async fn copy(&self, from: &Path, to: &Path) -> Result<()> {
-        if from.primary_only() {
+        if to.primary_only() {
             self.primary.copy(from, to).await
         } else {
             self.secondary.copy(from, to).await?;
@@ -142,6 +144,9 @@ impl ObjectStore for MirroringObjectStore {
     }
 
     async fn copy_if_not_exists(&self, from: &Path, to: &Path) -> Result<()> {
+        if !to.primary_only() {
+            self.secondary.copy(from, to).await?;
+        }
         self.primary.copy_if_not_exists(from, to).await
     }
 }
@@ -379,7 +384,7 @@ mod test {
             let primary_f = primary_elem.unwrap().unwrap();
             // hit manifest, skip, _versions contains all the manifest and should not exist on secondary
             let primary_raw_path = primary_f.file_name().to_str().unwrap();
-            if primary_raw_path.contains("manifest") || primary_raw_path.contains("_versions") {
+            if primary_raw_path.contains("_latest.manifest") {
                 primary_elem = primary_iter.next();
                 continue;
             }

rust/vectordb/src/table.rs

@@ -153,6 +153,22 @@ impl Table {
         })
     }
 
+    pub async fn checkout_latest(&self) -> Result<Self> {
+        let latest_version_id = self.dataset.latest_version_id().await?;
+        let dataset = if latest_version_id == self.dataset.version().version {
+            self.dataset.clone()
+        } else {
+            Arc::new(self.dataset.checkout_version(latest_version_id).await?)
+        };
+
+        Ok(Table {
+            name: self.name.clone(),
+            uri: self.uri.clone(),
+            dataset,
+            store_wrapper: self.store_wrapper.clone(),
+        })
+    }
+
     fn get_table_name(uri: &str) -> Result<String> {
         let path = Path::new(uri);
         let name = path