feat: add list/create/delete/update/checkout tag API (#2353)

add the tag related API to list existing tags, attach tag to a version,
update the tag version, delete tag, get the version of the tag, and
checkout the version that the tag bounded to.

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

- **New Features**
- Introduced table version tagging, allowing users to create, update,
delete, and list human-readable tags for specific table versions.
  - Enabled checking out a table by either version number or tag name.
- Added new interfaces for tag management in both Python and Node.js
APIs, supporting synchronous and asynchronous workflows.

- **Bug Fixes**
  - None.

- **Documentation**
- Updated documentation to describe the new tagging features, including
usage examples.

- **Tests**
- Added comprehensive tests for tag creation, updating, deletion,
listing, and version checkout by tag in both Python and Node.js
environments.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

nodejs/__test__/table.test.ts

@@ -1178,6 +1178,73 @@ describe("when dealing with versioning", () => {
   });
 });
 
+describe("when dealing with tags", () => {
+  let tmpDir: tmp.DirResult;
+  beforeEach(() => {
+    tmpDir = tmp.dirSync({ unsafeCleanup: true });
+  });
+  afterEach(() => {
+    tmpDir.removeCallback();
+  });
+
+  it("can manage tags", async () => {
+    const conn = await connect(tmpDir.name, {
+      readConsistencyInterval: 0,
+    });
+
+    const table = await conn.createTable("my_table", [
+      { id: 1n, vector: [0.1, 0.2] },
+    ]);
+    expect(await table.version()).toBe(1);
+
+    await table.add([{ id: 2n, vector: [0.3, 0.4] }]);
+    expect(await table.version()).toBe(2);
+
+    const tagsManager = await table.tags();
+
+    const initialTags = await tagsManager.list();
+    expect(Object.keys(initialTags).length).toBe(0);
+
+    const tag1 = "tag1";
+    await tagsManager.create(tag1, 1);
+    expect(await tagsManager.getVersion(tag1)).toBe(1);
+
+    const tagsAfterFirst = await tagsManager.list();
+    expect(Object.keys(tagsAfterFirst).length).toBe(1);
+    expect(tagsAfterFirst).toHaveProperty(tag1);
+    expect(tagsAfterFirst[tag1].version).toBe(1);
+
+    await tagsManager.create("tag2", 2);
+    expect(await tagsManager.getVersion("tag2")).toBe(2);
+
+    const tagsAfterSecond = await tagsManager.list();
+    expect(Object.keys(tagsAfterSecond).length).toBe(2);
+    expect(tagsAfterSecond).toHaveProperty(tag1);
+    expect(tagsAfterSecond[tag1].version).toBe(1);
+    expect(tagsAfterSecond).toHaveProperty("tag2");
+    expect(tagsAfterSecond["tag2"].version).toBe(2);
+
+    await table.add([{ id: 3n, vector: [0.5, 0.6] }]);
+    await tagsManager.update(tag1, 3);
+    expect(await tagsManager.getVersion(tag1)).toBe(3);
+
+    await tagsManager.delete("tag2");
+    const tagsAfterDelete = await tagsManager.list();
+    expect(Object.keys(tagsAfterDelete).length).toBe(1);
+    expect(tagsAfterDelete).toHaveProperty(tag1);
+    expect(tagsAfterDelete[tag1].version).toBe(3);
+
+    await table.add([{ id: 4n, vector: [0.7, 0.8] }]);
+    expect(await table.version()).toBe(4);
+
+    await table.checkout(tag1);
+    expect(await table.version()).toBe(3);
+
+    await table.checkoutLatest();
+    expect(await table.version()).toBe(4);
+  });
+});
+
 describe("when optimizing a dataset", () => {
   let tmpDir: tmp.DirResult;
   let table: Table;

nodejs/lancedb/index.ts

@@ -23,6 +23,8 @@ export {
   OptimizeStats,
   CompactionStats,
   RemovalStats,
+  Tags,
+  TagContents,
 } from "./native.js";
 
 export {

nodejs/lancedb/table.ts

@@ -20,6 +20,7 @@ import {
   IndexConfig,
   IndexStatistics,
   OptimizeStats,
+  Tags,
   Table as _NativeTable,
 } from "./native";
 import {
@@ -374,7 +375,7 @@ export abstract class Table {
    *
    * Calling this method will set the table into time-travel mode. If you
    * wish to return to standard mode, call `checkoutLatest`.
-   * @param {number} version The version to checkout
+   * @param {number | string} version The version to checkout, could be version number or tag
    * @example
    * ```typescript
    * import * as lancedb from "@lancedb/lancedb"
@@ -390,7 +391,8 @@ export abstract class Table {
    * console.log(await table.version()); // 2
    * ```
    */
-  abstract checkout(version: number): Promise<void>;
+  abstract checkout(version: number | string): Promise<void>;
+
   /**
    * Checkout the latest version of the table. _This is an in-place operation._
    *
@@ -404,6 +406,23 @@ export abstract class Table {
    */
   abstract listVersions(): Promise<Version[]>;
 
+  /**
+   * Get a tags manager for this table.
+   *
+   * Tags allow you to label specific versions of a table with a human-readable name.
+   * The returned tags manager can be used to list, create, update, or delete tags.
+   *
+   * @returns {Tags} A tags manager for this table
+   * @example
+   * ```typescript
+   * const tagsManager = await table.tags();
+   * await tagsManager.create("v1", 1);
+   * const tags = await tagsManager.list();
+   * console.log(tags); // { "v1": { version: 1, manifestSize: ... } }
+   * ```
+   */
+  abstract tags(): Promise<Tags>;
+
   /**
    * Restore the table to the currently checked out version
    *
@@ -699,8 +718,11 @@ export class LocalTable extends Table {
     return await this.inner.version();
   }
 
-  async checkout(version: number): Promise<void> {
-    await this.inner.checkout(version);
+  async checkout(version: number | string): Promise<void> {
+    if (typeof version === "string") {
+      return this.inner.checkoutTag(version);
+    }
+    return this.inner.checkout(version);
   }
 
   async checkoutLatest(): Promise<void> {
@@ -719,6 +741,10 @@ export class LocalTable extends Table {
     await this.inner.restore();
   }
 
+  async tags(): Promise<Tags> {
+    return await this.inner.tags();
+  }
+
   async optimize(options?: Partial<OptimizeOptions>): Promise<OptimizeStats> {
     let cleanupOlderThanMs;
     if (

nodejs/src/table.rs

@@ -249,6 +249,14 @@ impl Table {
             .default_error()
     }
 
+    #[napi(catch_unwind)]
+    pub async fn checkout_tag(&self, tag: String) -> napi::Result<()> {
+        self.inner_ref()?
+            .checkout_tag(tag.as_str())
+            .await
+            .default_error()
+    }
+
     #[napi(catch_unwind)]
     pub async fn checkout_latest(&self) -> napi::Result<()> {
         self.inner_ref()?.checkout_latest().await.default_error()
@@ -281,6 +289,13 @@ impl Table {
         self.inner_ref()?.restore().await.default_error()
     }
 
+    #[napi(catch_unwind)]
+    pub async fn tags(&self) -> napi::Result<Tags> {
+        Ok(Tags {
+            inner: self.inner_ref()?.clone(),
+        })
+    }
+
     #[napi(catch_unwind)]
     pub async fn optimize(
         &self,
@@ -546,3 +561,78 @@ pub struct Version {
     pub timestamp: i64,
     pub metadata: HashMap<String, String>,
 }
+
+#[napi]
+pub struct TagContents {
+    pub version: i64,
+    pub manifest_size: i64,
+}
+
+#[napi]
+pub struct Tags {
+    inner: LanceDbTable,
+}
+
+#[napi]
+impl Tags {
+    #[napi]
+    pub async fn list(&self) -> napi::Result<HashMap<String, TagContents>> {
+        let rust_tags = self.inner.tags().await.default_error()?;
+        let tag_list = rust_tags.as_ref().list().await.default_error()?;
+        let tag_contents = tag_list
+            .into_iter()
+            .map(|(k, v)| {
+                (
+                    k,
+                    TagContents {
+                        version: v.version as i64,
+                        manifest_size: v.manifest_size as i64,
+                    },
+                )
+            })
+            .collect();
+
+        Ok(tag_contents)
+    }
+
+    #[napi]
+    pub async fn get_version(&self, tag: String) -> napi::Result<i64> {
+        let rust_tags = self.inner.tags().await.default_error()?;
+        rust_tags
+            .as_ref()
+            .get_version(tag.as_str())
+            .await
+            .map(|v| v as i64)
+            .default_error()
+    }
+
+    #[napi]
+    pub async unsafe fn create(&mut self, tag: String, version: i64) -> napi::Result<()> {
+        let mut rust_tags = self.inner.tags().await.default_error()?;
+        rust_tags
+            .as_mut()
+            .create(tag.as_str(), version as u64)
+            .await
+            .default_error()
+    }
+
+    #[napi]
+    pub async unsafe fn delete(&mut self, tag: String) -> napi::Result<()> {
+        let mut rust_tags = self.inner.tags().await.default_error()?;
+        rust_tags
+            .as_mut()
+            .delete(tag.as_str())
+            .await
+            .default_error()
+    }
+
+    #[napi]
+    pub async unsafe fn update(&mut self, tag: String, version: i64) -> napi::Result<()> {
+        let mut rust_tags = self.inner.tags().await.default_error()?;
+        rust_tags
+            .as_mut()
+            .update(tag.as_str(), version as u64)
+            .await
+            .default_error()
+    }
+}