daesunp

This commit introduces **Quill.js-style delta payloads** for **array node events** within the **`packages/dds/tree`** eventing system. This **new capability** enhances the `AnchorSet`, `DeltaVisitor`, and `KernelEventBuffer` to capture and emit detailed `fieldMarks` for array changes, allowing consumers of both **`SharedTree`** and **`SimpleTree`** to subscribe to richer `nodeChanged` events. By providing granular `ArrayNodeDeltaOp` information, this change enables more efficient and accurate synchronization or UI rendering for array modifications. The implementation includes new `ArrayNodeDeltaOp` related types and `NodeChangedDataAlpha` interfaces, and is validated with new fuzz tests to ensure consistency across clients.

This commit introduces a **new capability** to the `packages/dds/tree` module by adding **composable labels for transactions**. It defines new types, `LabelTree` and `TransactionLabels`, which allow for tracking and associating metadata with **nested transactions**. The core logic for managing these labels, including `pushLabelFrame` and `popLabelFrame` methods, is integrated into the `SharedTree`'s transaction processing within `treeCheckout.ts` and `schematizingTreeView.ts`. This enhancement provides richer context for `LocalChangeMetadata` and `RemoteChangeMetadata`, significantly improving the ability to identify and process changes within complex, multi-level operations.

This commit introduces a **new capability** to the **SharedTreeCore**'s **transaction management** system. It adds a `size()` method to the `Transactor` interface and its `TransactionStack` implementation, allowing consumers to query the number of currently active transactions. This enhancement provides better visibility into the transaction state within the `shared-tree-core` package. New test cases have been added to verify the correct functionality and behavior of the `size()` method, including scenarios involving disposal.

This commit introduces a **new capability** to the **`tree-agent` framework** by implementing an `IdentifierCollisionResolver` to gracefully handle short name collisions during schema processing. This resolver automatically disambiguates identical short names (e.g., `foo` becomes `foo_1`, `foo_2`) to ensure uniqueness. The new logic is integrated into the **prompt generation**, **TypeScript schema rendering**, and **type generation** subsystems, affecting files such as `prompt.ts`, `renderSchemaTypeScript.ts`, and `typeGeneration.ts`. This enhancement prevents potential conflicts and improves the reliability of generated outputs by guaranteeing unique and unambiguous identifiers for schemas.

This commit introduces **native move operations for unhydrated nodes** within the **`simple-tree` DDS**, enhancing its core capabilities. It adds a `move` method to the `UnhydratedTreeSequenceFieldEditBuilder` and implements it in `UnhydratedSequenceField`, allowing for direct manipulation of unhydrated tree structures. This change also enables a **refactoring** of the `arrayNode`'s move implementation, eliminating the `withBufferedTreeEvents` workaround and improving event handling efficiency. New tests confirm correct event emission for both within-array and cross-field moves, ensuring the robustness of this **new feature**.

This commit introduces a **new capability** to the **SharedTree** and **SimpleTree** APIs, allowing users to associate an optional `label` object with transactions. This label is propagated through the system and can be accessed via `SchematizingSimpleTreeView.currentTransactionLabel` or within the `commitApplied` event callback. The primary purpose is to facilitate **transaction grouping**, particularly for advanced **undo/redo functionality**, by providing custom metadata for each commit. This change impacts `LocalChangeMetadata`, `RemoteChangeMetadata`, and `RunTransactionParams` interfaces, enhancing the extensibility of transaction management within the `packages/dds/tree` package.

This commit **increases the timeout** for the **state-based rebaser axioms stress test** within the `dds/tree` package. Specifically, the timeout for the `sequenceChangeRebaser.test.ts` file is extended from 80,000ms to 120,000ms. This **maintenance** change addresses recurring timeouts observed in the stress test pipeline, which were causing test failures due to insufficient execution time. By adjusting the timeout, the commit ensures the **reliability and stability of test execution** for the rebaser functionality, allowing comprehensive performance testing to complete successfully.

This commit introduces a **fix** within the **`tree-agent` package** to **sanitize friendly names for schema identifiers**, ensuring they conform to valid TypeScript naming conventions. The `unqualifySchema` function in `packages/framework/tree-agent/src/utils.ts` was modified to replace invalid characters with underscores and to prefix names starting with invalid characters with an underscore. This **maintenance** work prevents potential syntax errors when these identifiers are used in TypeScript code, such as `"com.fluidframework.foo!"` becoming `"foo_"`. New test cases were added to `prompt.spec.ts` and `schemaUtils.spec.ts` to verify the correct sanitization logic for various invalid schema names.

This commit introduces an **API usability improvement** by making the `options` parameter in the **`independentView`** function optional within the **`@fluidframework/tree`** package. This **refactoring** eliminates the need to explicitly pass an empty `{}` object when calling `independentView`, simplifying its usage. The change primarily affects the **`shared-tree`** module and its consumers, including various test files across the **`@fluidframework`** ecosystem. This enhancement streamlines development by reducing boilerplate and making the API more intuitive, with updated API reports reflecting the new signature.

This commit introduces a **new capability** for the **`tree-agent`** by implementing **property binding**, allowing it to **expose properties of schema classes directly to the Large Language Model (LLM)**. This enhancement significantly broadens the LLM's contextual understanding and interaction possibilities, as it can now access and interpret schema properties in addition to methods. The work involves defining new API types and mechanisms in `packages/framework/tree-agent/src/propertyBinding.ts`, integrating this into the `typeGeneration.ts` and `prompt.ts` logic to include properties in generated TypeScript schemas, and updating the `getZodSchemaAsTypeScript` utility for correct formatting. This ensures that the LLM's prompt accurately reflects the full state of bindable objects, improving the agent's ability to reason and operate on complex data structures.

This commit **updates the documentation** for **schema evolution** within the `data-structures/tree` section. It specifically **revises code examples** in `docs/docs/data-structures/tree/schema-evolution/allowed-types-rollout.mdx` to reflect the progression of the API. The change involves migrating from the deprecated `SchemaFactoryAlpha` to the current **beta API**, `SchemaFactoryBeta`. This **documentation update** ensures users are guided to implement schema evolution using the most up-to-date and supported interfaces, improving the accuracy and relevance of the provided examples.

This commit **promotes core schema definition APIs** within the **`@fluidframework/tree`** package from alpha to beta stability. The `staged`, `types`, `stagedRecursive`, and `typesRecursive` methods are now exposed on `SchemaFactoryBeta` and `SchemaStaticsBeta`, making them officially available for use in **beta applications**. This **API evolution** enables developers to define and manage **staged and recursive schema types** with increased confidence and stability. The change involves updating API reports for both `@fluidframework/tree` and `@fluidframework/fluid-framework`, adjusting module exports, and revising documentation to reflect the transition from alpha to beta interfaces for schema evolution.

This commit introduces a **new feature** to the **`simple-tree` DDS**, enabling the use of the `delete` keyword on properties of **`ObjectNode`** instances. This allows for the semantic removal of properties from these object nodes, enhancing their mutability. The core implementation modifies the `createProxyHandler` and adds a new `applyFieldChange` function within `objectNode.ts` to correctly process these deletions. This change improves the flexibility of `ObjectNode` manipulation, with updated tests ensuring proper behavior, especially for optional fields.

This commit introduces a **behavior change** to the **Simple Tree's record nodes**, where setting a value to `undefined` via the `set` function now explicitly behaves like a **delete operation**. This update ensures consistency in how `undefined` values are handled within **`recordNode`** structures, effectively removing the entry rather than storing an `undefined` value. The change is accompanied by **updated documentation** in `packages/dds/tree/src/simple-tree/node-kinds/record/recordNodeTypes.ts` and a **new test case** in `packages/dds/tree/src/test/simple-tree/node-kinds/recordNode.spec.ts` to verify this intended behavior. This **clarifies and standardizes** the interaction with record node properties, improving the predictability of the **`simple-tree`** API.

This commit **fixes a critical bug** in the **`DetachedFieldIndex`** module within the `dds/tree` package, where the `createEntry` method incorrectly generated only one `rootId` when attempting to create multiple entries (`count > 1`). The fix ensures proper incrementation of root and minor IDs, preventing data integrity issues for detached tree nodes. Additionally, **comprehensive unit tests** have been added for `DetachedFieldIndex` methods, including `createEntry`, `deleteEntry`, `loadData`, and `clone`. This work is a **bug fix** and **test enhancement**, significantly improving the **reliability and validation** of this core tree data structure responsible for managing detached tree data.

This commit introduces a **new capability** by adding comprehensive **identifier API methods** to the **TreeAlpha** interface within the `packages/dds/tree` package. These new utilities, exposed via `TreeAlpha.identifier` and `TreeIdentifierUtils`, enable **creating, retrieving, shortening, and lengthening node identifiers** across both **SharedTree** and **SimpleTree** contexts. Specifically, functions like `getIdentifierFromNode`, `shorten`, `lengthen`, and `create` provide flexible ways to manage node references, alongside a **bug fix** for `mockNodeIdentifierManager.ts` to handle out-of-bounds local ID offsets. This enhancement significantly improves the ability to **programmatically manage and interact with node identifiers**, ensuring more robust and flexible tree data structure manipulation.

Dropdown to change column type in table example app (#24637)

This commit introduces a **new capability** by adding the `TreeAlpha.key2` method within the **`@fluidframework/tree` DDS package**. This method provides a refined way to retrieve a node's key, specifically returning `undefined` for root nodes, and is intended to eventually replace the existing `Tree.key` method. The work also includes **refactoring** of `simple-tree` APIs to export key-related utility functions like `getStoredKey` and `getPropertyKeyFromStoredKey`. This change lays the groundwork for a more consistent and robust tree node keying mechanism, with updated API reports in both `tree` and `fluid-framework` packages reflecting the new public API.

Table example app using TableSchema (#24337)

This commit significantly **enhances the unit test coverage** for the **`tableSchema` module** within `packages/dds/tree`, adding comprehensive tests for various table operations like insertion, removal, and cell manipulation in `tableSchema.spec.ts`. It also **refactors** the `tableSchema` API by renaming `delete` methods to `remove` (e.g., `removeCell`, `removeRows`) for improved **API consistency** across `IRow`, `Row`, `ITable`, and `Table` interfaces. Furthermore, the `createTable` method gains new overloads, allowing it to be called without explicitly providing `Row` or `Column` schemas, simplifying its usage. Some new tests are intentionally skipped, marking areas where future policy decisions are required for specific edge cases. This work improves the **robustness and usability** of the `tableSchema` component.