- Introduced creation of `pub.p5i` files for publishers to maintain compatibility with older repository formats.
- Implemented saving repository configuration in legacy INI format (`pkg5.repository`).
- Updated tests to validate the generation and structure of legacy files.
- Added new dependencies (`rust-ini`) and updated `Cargo.toml` and `Cargo.lock` accordingly.
- Eliminate obsolete assertions for the root `file` directory as it is no longer created.
- Update tests to validate file paths in `publisher`-specific directories, aligning with the updated repository structure.
- Simplify `create_directories` by removing unnecessary `file` directory creation.
- Add `obsoleted.rs` module to handle storing, metadata management, and operations for obsoleted packages.
- Implement commands for marking, listing, searching, restoring, exporting, and importing obsoleted packages (`pkg6repo`).
- Enhance `RepositoryError` with `From` implementations for various error types to manage database and serialization-related errors.
- Introduce reusable data structures for obsoleted package metadata and export representation.
- Update `Cargo.toml` and `Cargo.lock` to include new dependencies (`redb`, `bincode`, etc.).
- Document obsoleted package workflow and integration details in `doc/obsoleted_packages.md` for contributors.
- Refactor repository internals to integrate obsoleted package support without disrupting existing workflow.
- Add robust error handling, logging, and pagination for enhanced usability and scalability.
- Replace manual recursive directory scanning with `walkdir` for simplified and efficient file traversal.
- Update dependencies in `Cargo.toml` and `Cargo.lock` to include `walkdir`.
- Enhance code maintainability by replacing redundant logic with streamlined traversal and processing.
- Eliminate redundant `test_manifest_parsing.rs` as its functionality is covered during `FileBackend` implementation.
- Define `PackageContentVectors` for streamlined data organization.
- Enhance manifest handling in `FileBackend` with recursive search, improved content extraction, and robust error handling.
- Simplify and centralize manifest parsing logic for maintainability and code clarity.
- Standardize formatting by aligning and cleaning up indentation, spacing, and line breaks across multiple modules.
- Enhance logging with improved message formatting for better context and readability.
- Simplify error handling in `pkg6dev`, `pkg6repo`, and core libraries by consolidating redundant patterns and improving clarity.
- Update test cases to reflect formatting and logging changes while extending coverage of edge cases.
- Perform general cleanup, including removing unused imports, refining comments, and ensuring consistent style.
- Extract manifest path construction and recursive manifest search into helper methods (`construct_manifest_path` and `find_manifests_recursive`).
- Enhance package information extraction by separating stem and version details from FMRI.
- Adjust directory creation logic to include per-package subdirectories for better organization.
- Simplify and centralize manifest parsing logic, improving error handling and maintainability.
- Modify `rebuild` and `refresh` methods to take `&self` instead of `&mut self`.
- Replace mutable repository instances in `pkg6dev`, `pkg6repo`, and core library with immutable ones.
- Introduce `debug` logging for enhanced traceability during metadata rebuild and refresh.
- Extract and simplify catalog rebuild logic into a dedicated `rebuild_catalog` method.
- General cleanup: remove redundant methods, reorganize logic, and refine comments across affected files.
- Update `FileBackend` to store files in a multi-level directory structure (`xx/yy/hash`) based on hash prefixes.
- Adjust method signatures and logic across `FileBackend`, `pkg6dev`, and `pkg6repo` to support mutable repository operations.
- Add test cases and scripts (`test_file_structure.rs`, `test_repo_operations.sh`) to validate the new file structure.
- Refactor catalog manager to use `RefCell` for interior mutability to enable lazy initialization.
- Clean up redundant and dead code, enhance comments, and replace outdated logic for file handling and metadata operations.
