- Simplified the `isReadable` condition in `encodeService` for better readability.
- Updated the `ServiceChunk` structure to conditionally include properties and references only if they exist, enhancing data efficiency.
- Introduced a `MAX_SCORING_DEPTH` constant in the matching module to limit recursion depth, improving performance and preventing potential stack overflow issues.
- Refactored the `computePairCost` and `matchChildren` functions to incorporate the new depth parameter, ensuring accurate cost calculations during matching.
These changes enhance the clarity and efficiency of the encoding and matching processes, contributing to overall performance improvements.
- Removed the encoding of attributes and tags from the `encodeService` function, consolidating their inclusion into the properties map as `Variant::Attributes(...)` and `Variant::Tags(...)`.
- Updated the `ServiceChunk` struct to reflect these changes, ensuring a more efficient serialization process during syncback operations.
- Adjusted related tests to align with the new structure, enhancing overall data management and consistency in service encoding.
These modifications improve the clarity and performance of the service encoding process.
- Updated the `encodeService` function to return a more concise result object by conditionally including properties, attributes, tags, and references only if they are present.
- This change enhances the efficiency of the encoding process by reducing unnecessary data in the output, improving overall performance during syncback operations.
- Introduced a new `encodeService` helper in the plugin to serialize service properties, attributes, tags, and references alongside children data.
- Updated the `performSyncback` function to utilize the new encoding method, ensuring all services are processed without child-only restrictions.
- Extended the `ServiceChunk` struct to include fields for properties, attributes, tags, and references, improving data handling during syncback.
- Modified the `build_dom_from_chunks` function to apply service properties and resolve references when reconstructing the service instances.
- Added comprehensive tests to validate the new service properties handling, ensuring accurate synchronization and data integrity.
These enhancements significantly improve the live syncback process, allowing for better service data management and consistency.
- Renamed test case to clarify that children with duplicate names are now encoded rather than skipped.
- Adjusted assertions to ensure all children are included in the encoded output, aligning with the updated encoding logic.
- Refactored related test cases for improved clarity and maintainability.
These changes enhance the accuracy of the ChangeBatcher tests, ensuring they correctly validate the new encoding behavior.
- Updated the `encodeInstance.stress.spec.lua` tests to eliminate checks for skipped counts, as duplicates are now included alongside unique children.
- Adjusted expectations in the tests to reflect the new behavior of the encoding process, ensuring accurate validation of encoded instances.
- Removed legacy assertions related to skipped counts, streamlining the test cases for better clarity and maintainability.
- Finalized the transition from a duplicate-name blacklist system to a comprehensive 3-pass matching algorithm for both syncback and forward operations, enhancing the handling of ambiguous paths.
- Integrated a new `matching.lua` module to facilitate instance matching during hydration, utilizing unique name matching, class name narrowing, and structural fingerprinting.
- Updated various components, including `ChangeBatcher`, `Reconciler`, and `snapshot` modules, to leverage the new matching logic, ensuring robust file management and improved performance.
- Removed legacy duplicate handling code and streamlined the encoding process for instances with duplicate names.
These changes significantly enhance the clarity and efficiency of file operations, paving the way for improved project management and user experience.
- Added `onInstanceInserted` callback to `InstanceMap` to notify when new instances are tracked, allowing `ChangeBatcher` to resolve pending unresolved Ref properties.
- Updated `createPatchSet` and `encodePatchUpdate` to support handling unresolved Ref properties, deferring their resolution until the target instance is available.
- Implemented logic in `ChangeBatcher` to manage pending unresolved Ref properties and re-queue them when the target instance is inserted.
- Introduced new tests to verify behavior with ambiguous paths and ensure stability when handling duplicate-named siblings.
These changes improve the handling of Ref properties in the system, enhancing synchronization reliability and robustness.
- Introduced a comprehensive audit guide for the production-grade sync feature, detailing the evaluation criteria for round-trip identity and code quality.
- Added a new plan for auditing the two-way sync Ref property feature, outlining specific tasks and expected behaviors.
- Updated `ServeSession.lua` to include the instance map in patch updates, ensuring accurate property encoding.
- Enhanced `encodeInstance.lua` to prevent encoding Ref properties during instance addition, delegating this to the patch update process.
- Implemented regression tests to highlight known limitations in Ref handling, ensuring better coverage and identification of existing bugs.
These changes improve the clarity of the sync feature's audit process and strengthen the handling of Ref properties in the system.
- Simplified the formatting of function calls in `encodePatchUpdate.spec.lua` for better readability.
- Added multiple new snapshot files to enhance end-to-end testing for reference path changes, including scenarios for adding, removing, and handling nonexistent attributes.
- Updated test cases in `apply.rs`, `compute.rs`, and `api.rs` to ensure consistent handling of reference properties and improve clarity in property updates.
These changes enhance the maintainability of the test suite and improve the robustness of snapshot testing.
- Promoted sync-related logging from Debug/Trace to Info level across multiple files, ensuring visibility of changes in the Studio output console.
- Added a new plan for expanding the connected mode test suite to cover multi-step scenarios and round-trip identity verification.
- Introduced a plan to fix stale tree desynchronization by implementing a validation mechanism on plugin connection.
- Created a plan to bump sync logging to Info level, detailing todos for various components in the plugin.
This update improves the clarity of logs and enhances testing coverage for connected mode scenarios.
- Introduced comprehensive tests for the `PatchSet` module, covering methods such as `hasRemoves`, `hasAdditions`, `hasUpdates`, and various ID and instance checks.
- Added a new test suite for `ServeSession`, focusing on WebSocket message handling and confirmation behavior, ensuring correct merging and application of patches.
- Implemented edge case tests in `ChangeBatcher` to verify data integrity during rapid changes and instance removals.
- Enhanced `applyPatch` tests to handle scenarios involving missing instances and ignored class names, ensuring robust patch application logic.
- Marked all todos in the comprehensive_plugin_tests plan as completed, indicating that all planned tests have been implemented.
- Introduced new stress test files for InstanceMap, PatchSet, and encoding functionalities, covering various scenarios including large batch operations, conflict resolution, and deep/wide hierarchies.
- Enhanced the testing framework to ensure thorough validation of plugin functionalities and edge cases.
- Introduced a new CI pipeline documentation file outlining steps for auto-fixing formatting, running linters, building, and testing.
- Documented the Rojo syncback architecture, detailing improvements to prevent duplicate file creation and format mismatches.
- Updated syncback logic to ensure existing instances are updated instead of creating duplicates, enhancing project file management.
- Added tests for encoding instances and handling nested children to improve reliability in syncback operations.
- Added detailed logging for syncback operations, including additions, removals, and updates, to improve traceability and debugging.
- Implemented summary logs for patch applications in both the client and server, capturing the number of changes made.
- Enhanced logging in the ChangeBatcher and ApiService to provide insights into property changes and instance updates during syncback processes.
- Improved error handling and logging for VFS events to ensure better visibility of file operations.
- Introduced a new function, `findClassDescriptor`, to retrieve class properties and metadata, including inherited properties, from the database.
- Updated `encodeInstance` to include encoding for Tags alongside Attributes, ensuring comprehensive property serialization.
- Improved property filtering logic to skip non-serializable properties and those matching default values, aligning with syncback behavior.
- Enhanced comments for clarity on property handling and encoding processes.
- Updated the encodeInstance function to include additional script properties (Disabled, LinkedSource, etc.) for encoding into the .meta.json5 file.
- Improved handling of properties by skipping those that are already managed or should be ignored, ensuring only serializable properties are encoded.
- Clarified comments to indicate the necessity of encoding the Source property.
- Added support for encoding the RunContext property in the Script class to determine file suffixes (.server.luau, .client.luau, .plugin.luau, .legacy.luau).
- Updated the Confirming.lua file to improve UI consistency in the confirmation message display.
- Modified snapshot middleware and syncback logic to accommodate new file extensions and ensure proper handling of plugin scripts.
- Refactored tests to align with the new naming conventions for script suffixes and improved path uniqueness checks.
- Updated the `ServeSession` to implement logic for pulling instances that exist in Studio but not in Rojo's filesystem, creating appropriate files during sync confirmation.
- Introduced a new `encodeInstance` function to serialize Studio instances for transmission, supporting all instance types.
- Modified the `PatchTree` to allow explicit user selection for push/pull/skip actions, enhancing user control during sync operations.
- Added tests for the new encoding functionality and updated existing tests to ensure proper behavior with the new syncback logic.
- Added support for a new sync mode in ServeSession that filters property changes to only include the Source property when the server indicates it is in source-only mode.
- Updated ChangeBatcher to handle the new syncSourceOnly parameter, ensuring that only relevant property changes are processed based on the server's capabilities.
- Enhanced API response to include sync_source_only flag, indicating the server's sync capabilities.
- Adjusted related functions to accommodate the new sync logic, improving data synchronization efficiency.
* Switch git submodules to Wally packages
* Update build snapshot
* Add wally to foreman and use latest versions
* Install packages in CI runners
* Fix indents
* Install packages in the correct directory
* Install packages in correct dir of release action too
* Remove submodules from ci checkout
* Remove submodules from release checkout
* Update selene with latest fix
* Fix whitespace
Co-authored-by: Lucien Greathouse <me@lpghatguy.com>
* Implement ChangeBatcher
* Use ChangeBatcher for two-way sync
* Pause updates during patch application
* I can English good
* Break after encountering a nil Parent change
This prevents __flush from erroring out when an instance's Parent is
changed to nil and it has other property changes in the same batch.
* Update rbx_dom_lua
* Don't connect changed listeners in a running game
#468 made me realize how bad of an idea this is in general...
* Update TestEZ and fix sibling Ref reification test
* Add ChangeBatcher tests
* Test instance unpausing by breaking functionality out to __cycle
* Break up the module a bit and improve tests
* Shuffle requires around and edit comment
* Break out more stuff, rename createChangePatch -> createPatchSet
* Make ChangeBatcher responsible for unpausing all paused instances
This somewhat improves the situation (of course, it would preferrable
to not have to hack around this problem with Source at all). It also
sets us up nicely if we come across any other properties that do
anything similar.
* Remove old reference to pausedBatchInstances
* Use RenderStepped instead of Heartbeat and trash multi-frame pauses
I probably should have done this in the first place...
ChangeBatcher still needs to unpause instances, but we don't need to
hold pauses for any longer than one cycle.
* Remove useless branch
* if not next(x) -> if next(x) == nil
* Add InstanceMap:unpauseAllInstances, use it in ChangeBatcher
* Move IsRunning check to InstanceMap:__maybeFireInstanceChanged
