- Improved error handling in ServeSession by rolling back temp IDs if encoding fails for new scripts.
- Added logic in applyPatch to remap echoed additions from the DescendantAdded handler, preventing duplicate instances.
- Updated API to resolve intermediate container temp IDs to real tree references, ensuring proper parent-child relationships for new scripts.
- Introduced new tests to verify the presence of server info flags and the resolution of temp IDs in scripts-only mode.
These changes aim to optimize script management and enhance the overall functionality of the session handling.
- Updated the TODO list to refine tasks related to sync performance in scripts only mode, emphasizing the need to prune non-script leaves.
- Enhanced the sourcemap command to handle non-terminal stdin during watch mode, allowing for graceful exit when stdin is closed.
These changes aim to improve project management and optimize the sourcemap functionality.
- Added tasks to investigate Tree reconciliation issues and improve sync performance in scripts only mode.
- Updated the project template to include ignore paths for ServerStorage/RBX_ANIMSAVES and MoonAnimator2Saves, enhancing version control practices.
These updates aim to streamline project management and optimize synchronization processes.
- Added a task to address the issue of unchanged commits in Git by suggesting the inclusion of the last Git hash.
- Included a note to add ServerStorage/RBX_ANIMSAVES to the ignore list.
These updates aim to improve task management and enhance the project's version control practices.
- Changed the format on save setting from true to false for both Lua and Luau in the VSCode configuration.
- Adjusted the editor ruler to maintain a consistent column width of 100.
These updates aim to refine formatting behavior and align with project preferences.
- Changed the editor ruler and stylua column width from 80 to 100 to accommodate wider code formatting.
These updates aim to improve code readability and align formatting preferences across the project.
- Added detailed tasks for implementing Atlas sync, including conditions for automatic acceptance and handling of connected-mode operations.
- Emphasized the need for the agent to await sync completion to ensure state integrity.
These updates aim to enhance the project's synchronization capabilities and improve task management.
- Added tasks for setting up a nightly build and suffixing the version for beta releases under the versioning section.
- Included a note on improving property diffs and merging support in the merging section.
These updates aim to enhance project management and development workflow.
- Added default formatter and enabled format on save for Lua and Luau in the VSCode settings for both the main project and the project template.
These changes aim to improve code formatting consistency and streamline the development workflow.
This commit updates the vscode-rojo submodule to the latest commit, ensuring the project incorporates the most recent changes and improvements from the submodule repository.
- Updated the handling of `Rojo_Id` attributes in the patch computation to improve reference resolution for instances.
- Refactored the `compute_property_patches` function to accept a context for specified IDs, enhancing the accuracy of property patching.
- Changed logging level from info to debug for prefetching file statistics to reduce log verbosity.
These changes aim to improve the accuracy of snapshot processing and optimize logging during file prefetching.
- Removed redundant Rojo reference attributes from snapshot comparisons to prevent spurious diffs during VFS events.
- Updated the logic for interned instance additions to ensure deterministic ordering by redacted parent and name.
These changes aim to improve the accuracy of snapshot comparisons and enhance the handling of interned instances.
- Changed project name from Rojo to Atlas in the README, reflecting the new branding.
- Expanded the README with detailed installation instructions, features, and CLI commands for Atlas.
- Added functionality in the CLI clone command to change the current working directory to the specified path after cloning.
These updates aim to improve documentation clarity and enhance user experience with the Atlas tool.
- Introduced a new upload workflow for the Atlas plugin to Roblox using Lune and OpenCloud.
- Added `UploadDetails.json` to store plugin metadata.
- Updated `build.rs` to include `UploadDetails.json` in the plugin snapshot.
- Enhanced `plugin.project.json` to reference the new `UploadDetails`.
- Added a new submodule for `opencloud-execute` to facilitate plugin uploads.
- Created scripts for publishing the plugin via both PowerShell and Bash.
- Updated GitHub Actions workflow to include steps for installing Rokit and uploading the plugin.
These changes aim to streamline the plugin release process and improve integration with Roblox's marketplace.
- Introduced a VFS prefetch cache to optimize file reads during project loading, reducing I/O overhead for large projects.
- Implemented parallel I/O in the VFS backend to improve serve startup and file processing performance.
- Replaced the `float-cmp` dependency with a custom float comparison and formatting solution for precise disk representation.
- Fixed NaN equality handling in `trueEquals` for compound types.
- Improved patch application logic for legacy references and resolved multiple live syncback issues.
- Added yield intervals for plugin diffing to prevent UI freezes during large syncs.
- Enhanced syncback functionality with automatic dedup suffix stripping and improved error handling.
These changes aim to enhance performance, reliability, and user experience in file handling and synchronization processes.
- Updated the `BASE_DURATION_MS` constant from 30.0 to 100.0 to allow for longer wait times during server startup.
- Increased the `MAX_TRIES` constant from 5 to 10 to provide more attempts for the server to come online before failing.
These changes aim to improve the reliability of server initialization by allowing for more extended backoff periods and retries.
This commit updates the rbx-dom submodule to the latest commit, ensuring that the project is using the most recent changes and improvements from the submodule repository.
- Added a check for `watch_enabled` before calling `unwatch` in the `remove_file` and `remove_dir_all` methods to prevent unnecessary operations when watching is disabled.
- Updated error handling in the `StdBackend` to include a new condition for `Generic` errors related to watch status, improving logging clarity.
These changes aim to optimize the file system operations and enhance error handling in the virtual file system.
- Updated the `vscode-rojo` submodule to the latest commit.
- Modified the syncback confirmation dialog title to indicate its beta status and adjusted its dimensions for better usability.
- Enhanced the warning message to clarify the differences between live syncback and CLI syncback, helping users make informed decisions.
These changes aim to improve user experience and provide clearer guidance during the syncback process.
- Added `GIT_TERMINAL_PROMPT` environment variable set to `0` to suppress terminal prompts during Git operations.
- Redirected standard input to null to prevent any interactive prompts, ensuring smoother initialization processes.
These changes aim to enhance the user experience by streamlining the Git initialization workflow.
- Updated the `vscode-rojo` submodule to indicate a dirty state.
- Modified `DomLabel.lua` to adjust size and position calculations for better layout.
- Removed redundant code in `init.lua` related to visible nodes.
- Introduced new functions in `Confirming.lua` to create statistical tags and bars for change information, improving the visual representation of changes in the confirmation page.
These updates aim to enhance the user interface and provide clearer insights into changes made during the patch visualization process.
- Updated the `vscode-rojo` submodule to the latest commit.
- Added a new plan for addressing issues found during the audit of parallelization and optimization commits, including fixes for a data loss bug in the orphan scanner, a duplicate-children bug in the prefetch cache, and a nested-detection correctness issue.
- Implemented a small DRY refactor by extracting a shared helper function for UNC path stripping.
- Enhanced the `syncback` logic to ensure hidden directories are preserved during clean mode operations and added tests to validate these changes.
These updates aim to improve the reliability and correctness of the parallel I/O functionality while maintaining code clarity and performance.
- Changed several logging statements from `info` to `debug` across various modules, including `StdBackend`, `RefPathIndex`, `ServeSession`, and `syncback_loop_with_stats`.
- This adjustment aims to reduce log verbosity during normal operations while retaining detailed logs for debugging purposes, enhancing overall performance and maintainability.
- Deleted the `project_port` and `serve_address` methods from the `ServeSession` implementation as they were no longer needed.
- Removed the `scan_directory` and `normalize_existing_path` helper functions from the `syncback_loop_with_stats` function, streamlining the code and improving maintainability.
These changes aim to enhance code clarity and reduce unnecessary complexity in the session and syncback functionalities.
- Updated the `ServeCommand` to load the initial project directly, improving clarity and reducing unnecessary session handling.
- Introduced a new `replace_placeholders` function in `FsSnapshot` to efficiently replace `__ROJO_REF_<hex>__` placeholders using a HashMap, optimizing the substitution process.
- Enhanced the syncback logic to handle both `__ROJO_REF_` and `Rojo_Ref_` markers, ensuring comprehensive path replacements during the syncback process.
These changes aim to improve the performance and maintainability of the serve and syncback functionalities.
- Added a `children` field to the `PrefetchCache` structure to store sorted child paths for directories, optimizing directory read operations.
- Updated the `read_dir` method in `VfsInner` to utilize the new `children` cache, improving performance by reducing backend calls for cached directories.
- Refactored the `prefetch_project_files` function to populate the `children` map during prefetching, enhancing the efficiency of path collection.
- Adjusted tests to accommodate the new `children` field in `PrefetchCache`, ensuring comprehensive coverage of the updated functionality.
These changes aim to improve the performance and reliability of the VFS by refining the prefetching mechanism and enhancing directory handling.
- Simplified the iteration over directory entries in the `prefetch_project_files` function by using `into_iter().flatten()` instead of a nested loop.
- This change enhances code clarity while maintaining the existing functionality of collecting valid directory entries.
These modifications aim to streamline the code and improve maintainability in the prefetching process.
- Updated the `PrefetchCache` structure to include metadata about whether paths are files or directories, optimizing the cache's utility during VFS operations.
- Refactored the `prefetch_project_files` function to accept a `Project` object, allowing for more flexible path collection and improved logging of the prefetching process.
- Added tests to validate the behavior of the prefetch cache, ensuring correct handling of file reads, canonicalization, and cache depletion scenarios.
- Enhanced existing tests to cover new functionality, including concurrent reads and stress tests for snapshot determinism.
These changes aim to improve the performance and reliability of the VFS by refining the prefetching mechanism and ensuring accurate metadata handling.
- Introduced additional Git configuration settings: `core.splitIndex`, `core.untrackedcache`, and `feature.manyFiles` to enhance repository initialization.
- These changes aim to improve the default setup for new repositories, optimizing performance and functionality from the start.
- Introduced a `PrefetchCache` structure to store pre-read file contents and canonical paths, enhancing the speed of initial reads during snapshot building.
- Implemented parallel file reading using `rayon` and `walkdir` to populate the cache efficiently.
- Updated the `VfsInner` struct to include the prefetch cache and modified read methods to utilize it, reducing I/O operations.
- Added functions to set and clear the prefetch cache in the `Vfs` API, improving memory management during live operations.
- Enhanced tests to validate the functionality of the prefetch cache, ensuring correct behavior during file reads and canonicalization.
These changes aim to significantly improve the performance of the VFS by minimizing file access times during initial operations.
- Introduced `DIFF_YIELD_INTERVAL` and `HYDRATE_YIELD_INTERVAL` constants to manage execution pacing during the diffing and hydrating processes, respectively.
- Implemented logic to pause execution every specified interval to improve performance and prevent potential blocking in long-running operations.
- Enhanced the `diff` and `hydrate` functions to track and yield based on their respective counts, ensuring smoother operation during intensive tasks.
These changes aim to enhance the responsiveness of the system during heavy processing tasks by introducing controlled yielding mechanisms.
- Simplified the comparison logic for `Font` variants in the `variant_eq` and `variant_eq_disk` functions by consolidating multiple conditions into single lines for improved readability.
- Removed the handling of `cached_face_id` from the `hash_variant` function, streamlining the hashing process for `Font` variants.
These changes aim to enhance code clarity and maintainability while preserving the existing functionality.
- Introduced a new plan to address port-in-use race conditions in both server and test environments.
- Implemented a retry mechanism with exponential backoff for binding to known ports in `LiveServer::start()`.
- Added a `respawn_with_new_port()` method in `TestServeSession` to handle transient port conflicts during tests.
- Modified `wait_to_come_online()` to retry with a new port if the server process exits due to `AddrInUse`.
- Updated `fresh_rebuild_read()` to include port retry logic, enhancing robustness against port conflicts.
These changes aim to improve the reliability of server and test interactions by mitigating race conditions related to port usage.
- Enhanced the syncback process to prevent concurrent executions by implementing a locking mechanism.
- Updated error responses to provide clearer feedback when a syncback is already in progress.
- Refactored related functions to streamline the syncback initiation logic, improving overall reliability.
- Added new tests to validate the concurrency handling and error messaging, ensuring robustness in edge cases.
These changes aim to enhance the reliability and user experience of the syncback feature by addressing concurrency issues and improving error handling.
- Introduced a new plan to automatically strip `~N` dedup suffixes from filenames during forward sync, eliminating the need for redundant `.meta.json5` files.
- Implemented the `strip_dedup_suffix()` utility in `dedup_suffix.rs` to handle suffix removal.
- Updated `file_name_for_path()` and `get_dir_middleware()` to utilize the new stripping functionality.
- Modified syncback logic to skip writing meta `name` fields for dedup-only cases, ensuring cleaner instance naming.
- Enhanced existing tests and added new tests to cover the updated dedup behavior, ensuring robustness and correctness.
These changes aim to streamline the handling of deduplicated instances, improving overall system efficiency and clarity in naming conventions.
- Made `format_f32` and `format_f64` functions public to allow external access.
- Introduced new constants for disk options and buffer sizes for `f32` and `f64`.
- Added `cleanup_f32` function to clean up float values before comparison.
- Implemented `disk_eq_f32` and `disk_eq_f64` functions to compare float values based on their disk representation.
- Updated `variant_eq_disk` to utilize the new disk equality functions, improving accuracy in matching algorithms.
These changes aim to enhance the precision of float comparisons and ensure consistent behavior across different representations in the system.
- Added a new section in `audit.md` for the Live Syncback Pipeline, detailing the plugin encoding, transport, and server handling processes.
- Included entries for newly introduced files `encodeHelpers.lua` and `encodeService.lua` in the audit documentation.
- Updated the audit checklist to cover live syncback parity and related concerns.
- Revised rule MDC files (`atlas-api.mdc`, `atlas-plugin.mdc`, `atlas-dedup.mdc`) to reflect recent changes, including the addition of `encodeHelpers.lua` and clarifications on `SyncbackConfirm`.
- Ensured comprehensive coverage of new test suites in the audit documentation.
These changes aim to enhance the clarity and accuracy of the audit documentation and rules, aligning them with the latest codebase updates.
- Introduced a new audit plan addressing six identified issues during the live syncback audit, including resource leaks, error handling, concurrent request management, input validation, documentation clarity, and test coverage.
- Implemented fixes such as replacing `mem::forget` with `drop` for better memory management, enhancing error recovery to prevent server crashes, and validating chunk counts against blob sizes.
- Updated documentation to reflect the audit findings and standards, ensuring clarity on the requirements for each fix.
- Added five new integration tests to cover various edge cases and ensure robustness in the syncback functionality.
These changes aim to improve the reliability and correctness of the live syncback feature while providing a structured approach to addressing identified issues.
- Updated versions of several dependencies in `Cargo.lock`, including `async-compression` (0.4.40), `aws-lc-rs` (1.16.0), `bumpalo` (3.20.1), and `compression-codecs` (0.4.37), ensuring compatibility and access to the latest features and fixes.
- Refactored the `fire` method in `SyncbackSignal` to return a boolean indicating whether a syncback was successfully initiated, preventing concurrent syncbacks.
- Enhanced error handling in the API syncback process to return a conflict response if a syncback is already in progress.
- Improved the `serve.rs` file by replacing `forget` with `drop` for better memory management and added checks to ensure service chunk integrity during live syncback operations.
- Added new tests for edge cases and resilience in live syncback functionality, ensuring robustness against malformed data and rapid sequential requests.
These changes aim to enhance the reliability and performance of the syncback feature while keeping dependencies up to date.
- Introduced a new audit plan detailing the "Ambiguous Path Handling" feature, including three parallel matching algorithm implementations (two in Rust and one in Lua) and their integration with the Reconciler.
- Documented findings related to scoring discrepancies between Lua and Rust implementations, including fixes for stale comments, scoring granularity for Tags/Attributes, and performance improvements in cost computation.
- Enhanced the `trueEquals` function to support fuzzy equality for additional types (UDim, UDim2, NumberRange, Rect, NumberSequence, ColorSequence) with corresponding unit tests to ensure accuracy.
These changes aim to improve the reliability and performance of the matching algorithms while ensuring consistency across implementations.
- Eliminated the `float-cmp` crate from the project, replacing its functionality with custom float comparison functions that match Lua's epsilon-based equality checks.
- Updated the `variant_eq` function to utilize the new fuzzy equality methods for `Float32` and `Float64` variants, ensuring consistent behavior across Rust and Lua implementations.
- Enhanced the matching algorithms in both forward sync and syncback processes to account for refined scoring of `Ref`, `Tags`, and `Attributes`, improving scoring accuracy and parity between implementations.
- Introduced a new plan document detailing the changes made to the matching algorithms, including the rationale for the updates and the expected impact on performance and correctness.
These changes aim to improve the reliability and maintainability of the matching algorithms while ensuring parity with Lua's behavior.
- Introduced a comprehensive audit plan for the git-based sync defaults feature, detailing the interaction between server and plugin components, including metadata handling and auto-staging logic.
- Documented the audit findings, highlighting performance considerations, path normalization issues, and code quality improvements.
- Included a sequence diagram to illustrate the workflow and a detailed table of audited files, enhancing clarity and maintainability of the feature.
These changes aim to ensure a thorough understanding and evaluation of the git sync defaults implementation, promoting better performance and reliability in the system.
