Add support for book covers

.gitignore

@@ -1,8 +1,15 @@
+covers/
 data/
 libro-test.db
 libro.db
 libro.bak
+node_modules/
+dist/
+dist-electron/
+package-lock.json
 
-.ruff_cache
-uv.lock
-.venv/
+*.log
+*.local
+.DS_Store
+.vscode/*
+.vite

AGENTS.md

@@ -1,43 +1,19 @@
-# AGENTS.md
+# CLAUDE.md
 
 This file provides guidance to LLMs working with code in this repository.
 
 ## Project Overview
 
-Libro is a terminal-based reading tracker written in Python that stores data in a local SQLite database. It is a pure CLI tool, interactive input uses `prompt_toolkit` for field prompts with validation and autocomplete. Output is formatted with `rich` Python library.
+Libro is a desktop reading tracker built with Electron + React + Vite. It uses a local SQLite database.
 
 ## Development Commands
 
-### Running
-
-```bash
-# Run the application
-uv run libro              # Show books read this year
-uv run libro add          # Add a book + review
-uv run libro report       # Same as default
-uv run libro --db ./libro.db report  # Use specific database file
-```
-
-### Using Justfile
-
 ```bash
-just run          # Run the app
-just test         # Run pytest tests
-just lint         # Check with ruff
-just lint-fix     # Fix lint issues
-just type-check   # Run mypy
-just ci           # Run all checks (lint, type-check, test)
-just install      # Install dependencies with uv sync
-just dev-install  # Install dev dependencies
-just build        # Build the package
-```
-
-### Testing and Quality
-
-```bash
-uv run python -m pytest tests/ -v
-uv run python -m ruff check src/libro/
-uv run python -m mypy --package libro
+npm run dev        # Start Vite renderer + Electron (hot-reload)
+npm run build      # Typecheck, build renderer, build Electron main/preload
+npm run package    # Full build + electron-builder → .dmg
+npm run typecheck  # tsc on both renderer and Electron tsconfigs
+npm run lint       # eslint
 ```
 
 ## Code Architecture
@@ -46,27 +22,41 @@ uv run python -m mypy --package libro
 
 ```
 libro/
-├── src/libro/
-│   ├── main.py           # Entry point, command routing via match/case
-│   ├── config.py         # Argument parsing and database location logic
-│   ├── models.py         # Dataclasses: Book, Review, BookReview, ReadingList, ReadingListBook
-│   └── actions/
-│       ├── db.py         # Database init and migration
-│       ├── show.py       # Display functions (rich tables)
-│       ├── report.py     # Report views (author stats, yearly charts)
-│       ├── modify.py     # Add/edit books and reviews (prompt_toolkit input)
-│       ├── lists.py      # Reading list management
-│       └── importer.py   # CSV import (Goodreads format and generic CSV)
-└── tests/
+├── electron/
+│   ├── main.ts       # Electron main process, BrowserWindow setup
+│   ├── preload.ts    # Context bridge — exposes LibroApi to renderer
+│   ├── ipc.ts        # IPC handler registration
+│   └── db/
+│       ├── client.ts # Database singleton, path resolution, migration call
+│       ├── migrate.ts # Schema migrations
+│       ├── books.ts  # Book/review queries
+│       ├── reports.ts # Year counts, author counts, review filters
+│       └── lists.ts  # Reading list CRUD
+├── shared/
+│   └── types.ts      # Shared TypeScript interfaces (LibroApi, all input/output types)
+└── src/              # React renderer
+    ├── App.tsx        # Shell with nav (Books, Reports, Search, Lists, Add Book)
+    ├── lib/
+    │   ├── api.ts     # Typed wrapper around window.libro (the context bridge)
+    │   ├── ratings.ts # Star rating helpers
+    │   ├── ReviewTable.tsx # Shared review table component (rows open BookDetailModal)
+    │   └── BookDetailModal.tsx # Modal showing full book details + all reviews
+    └── features/
+        ├── books/BooksByYearView.tsx
+        ├── reports/ReportsView.tsx
+        ├── search/SearchView.tsx
+        ├── lists/ListsView.tsx
+        └── add/AddBookReviewForm.tsx
 ```
 
 ### Key Technologies
 
-- **CLI Parsing**: argparse with subparsers
-- **Interactive Input**: prompt_toolkit (PromptSession with validators and completers)
-- **Terminal Output**: rich (tables, progress bars, styled text)
-- **Database**: SQLite3 (stdlib)
-- **Package Manager**: uv
+- **Framework**: Electron 33 + React 19 + Vite 5
+- **Database**: better-sqlite3 (synchronous, runs in main process only)
+- **Styling**: Halfmoon CSS
+- **Build**: tsup for Electron main/preload, Vite for renderer
+- **Packaging**: electron-builder (targets macOS DMG)
+- **Package Manager**: npm
 
 ## Data Model
 
@@ -82,14 +72,7 @@ The application separates **books** from **reviews** to support:
 3. `reading_lists` - Curated book collections
 4. `reading_list_books` - Many-to-many join table (with priority and added_date)
 
-**Models** (all dataclasses in `models.py`):
-- `Book` - Single book entity with `insert()`, `get_by_id()`, `find_by_title_author()`
-- `Review` - Single review entity with `insert()`
-- `BookReview` - Combined book + review, most commonly used for display; has `get_by_id()`
-- `ReadingList` - List entity with full CRUD methods
-- `ReadingListBook` - Association with `get_books_in_list()`, `get_list_stats()`
-
-### Database Schema
+**Database Schema:**
 
 ```sql
 CREATE TABLE books (
@@ -129,128 +112,35 @@ CREATE TABLE reading_list_books (
 );
 ```
 
-## Application Flow
-
-### Entry Point
-
-`main.py` calls `init_args()` from `config.py`, connects to the database, runs `migrate_db()`, then routes to the correct action via `match args["command"]`.
-
-### Database Location Priority
-
-1. `--db` command-line flag
-2. `libro.db` in current directory
-3. `LIBRO_DB` environment variable
-4. Platform-specific data directory (via `appdirs`):
-   - Linux: `~/.local/share/Libro/mkaz/libro.db`
-   - macOS: `~/Library/Application Support/Libro/mkaz/libro.db`
-   - Windows: `%APPDATA%\mkaz\Libro\libro.db`
-
-## CLI Commands
-
-```bash
-# Default - show books read this year
-libro
-libro report
-
-# Year chart of books read
-libro report --chart
-
-# Author stats (all authors with counts)
-libro report --author
-# Books by specific author
-libro report --author "Stephen King"
-
-# Show book/review detail by review ID
-libro report <id>
-
-# Filter by year
-libro report --year 2024
-
-# Add book + review (interactive prompts)
-libro add
-
-# Book management
-libro book                    # Show recent books
-libro book <id>               # Show specific book
-libro book add                # Add book only (no review)
-libro book edit <id>          # Edit book details
-libro book --author "King"    # Filter by author
-libro book --title "Found"    # Filter by title
-libro book --year 2024        # Filter by year
-
-# Review management
-libro review                  # Show recent reviews
-libro review <id>             # Show review detail
-libro review add <book_id>    # Add review to existing book
-libro review edit <id>        # Edit review
-libro review --author "King"  # Filter by author
-libro review --year 2024      # Filter by year
-
-# Reading list management
-libro list create "Name" --description "..."
-libro list show               # All lists
-libro list show <id>          # Specific list
-libro list add <list_id> [book_ids...]  # Add books to list (interactive if no IDs)
-libro list remove <list_id> <book_id>
-libro list stats              # All list stats
-libro list stats <id>         # Specific list stats
-libro list edit <id> --name "New Name"
-libro list delete <id>
-libro list import <file.csv> --name "List Name"
-
-# Import from Goodreads CSV
-libro import <file.csv>
-```
-
-## Important Patterns
-
-### Interactive Input (modify.py)
-
-`add_book_review()`, `add_book()`, `add_review()`, `edit_book()`, `edit_review()` all use `prompt_toolkit.PromptSession` with:
-- **Validators**: `NonEmptyValidator`, `IntValidator`, `RatingValidator` (1-5), `DateValidator` (YYYY-MM-DD)
-- **Completers**: `AuthorCompleter` (sorted by frequency), `GenreCompleter`
-- Tab completion works in any field with a completer attached
-
-### Date Handling
-
-Dates are stored as ISO 8601 strings. `models.py` registers a SQLite adapter/converter:
-
-```python
-sqlite3.register_adapter(date, adapt_date_iso)     # date -> "YYYY-MM-DD"
-sqlite3.register_converter("date", convert_date)   # bytes -> date
-```
-
-The connection must be opened with `detect_types=sqlite3.PARSE_DECLTYPES` for converters to fire, but the current code uses `db.row_factory = sqlite3.Row` without `detect_types`. Date fields come back as strings; convert with `date.fromisoformat()` if needed.
-
-### Adding a New Action
-
-1. Add argument parsing to `config.py` (argparse subparser or flag)
-2. Create or update a function in the appropriate `actions/` file
-3. Add a case in `main.py`'s `match args["command"]` block
-4. Use `rich.Table` or `rich.console` for output; use `PromptSession` for interactive input
-
-### Adding a New Store Query
+## IPC Architecture
 
-SQL lives directly in the model methods (`models.py`) and action files. There is no separate store layer. Keep SELECT queries in model classmethods when returning typed objects; put display-only queries in `show.py` or `report.py`.
+All database access happens in the main process. The renderer never touches the database directly.
 
-## Troubleshooting
+- `preload.ts` exposes a typed `LibroApi` object on `window.libro` via `contextBridge`
+- `src/lib/api.ts` re-exports this as `api` for use in React components
+- `ipc.ts` registers `ipcMain.handle` handlers that call into `electron/db/` functions
 
-### Duplicate Book on Import
+IPC channel naming: `namespace:verb-noun` (e.g. `books:add-book-review`, `lists:get-by-id`).
 
-`importer.py` calls `Book.find_by_title_author()` before inserting. If a book already exists, it skips the insert and adds a review to the existing book.
+## Database Location
 
-### Date Shows Wrong Value
+Priority order:
 
-Check that the date string is valid ISO format (`YYYY-MM-DD`). The `DateValidator` in `modify.py` enforces this on input. For database reads without `detect_types`, call `date.fromisoformat(row["date_read"])` explicitly.
+1. `libro.db` in cwd (if it exists)
+2. `LIBRO_DB` environment variable
+3. `{appData}/Libro/mkaz/libro.db`
 
-### Reading List Migration
+## Adding a New Feature
 
-Old databases without reading list tables are handled by `migrate_db()` in `actions/db.py`, which runs on every startup and creates missing tables idempotently.
+1. Add types to `shared/types.ts` (input, output, and the `LibroApi` interface entry)
+2. Add the query function in the appropriate `electron/db/*.ts` file
+3. Register an IPC handler in `electron/ipc.ts`
+4. Wire it up in `preload.ts` and `src/lib/api.ts`
+5. Build the React component in `src/features/`
 
 ## Code Style
 
-- Python 3.10+ (`match/case`, `list[str]` type hints)
-- Dataclasses for models
-- Type hints throughout; mypy enforced via `just type-check`
-- ruff for linting and formatting
-- No global state; `db` connection passed explicitly to every function
+- Strict TypeScript; all types defined in `shared/types.ts`
+- Two separate tsconfig targets: `tsconfig.app.json` (renderer) and `tsconfig.electron.json` (main/preload)
+- No ORM; raw SQL via better-sqlite3 synchronous API
+- eslint with react-hooks and react-refresh plugins

CONTRIBUTING.md

@@ -1,62 +1,23 @@
 # Contributing to Libro
 
-Hey there! Thanks for your interest in contributing to Libro. This is just a hobby project I work on in my spare time, but I'm totally open to help from others who find it useful or interesting.
+Libro is a hobby project, but contributions are welcome.
 
 ## Getting Started
 
-The easiest way to get up and running:
+1. Fork and clone the repo
+2. Install dependencies: `npm install`
+3. Run the dev build: `just dev`
 
-1. Fork the repo
-2. Clone your fork locally
-3. Make sure you have Python 3.10+ and install `uv` if you don't have it
-4. Run `just install` to set up dependencies
-5. Run `uv run libro --help` to make sure everything works
+## Development
 
-## Making Changes
-
-### Development Workflow
-
-- `just lint` - Check your code style (uses ruff)
-- `just run <args>` - Test your changes with the CLI
-- `just build` - Full build process before submitting
-
-### What I'm Looking For
+`just dev` builds the Electron main process, starts the Vite dev server, then launches the app. Vite handles hot-reload for renderer changes. If you change anything in `electron/`, restart `just dev`.
 
-I'm pretty relaxed about contributions, but here are some things that would be especially welcome:
-
-- **Bug fixes** - If something's broken, I'd love help fixing it
-- **Small features** - New commands, import formats, display options, etc.
-- **Documentation improvements** - Better help text, examples, etc.
-- **Testing** - The project could use more automated tests
-- **Performance improvements** - Making things faster is always good
-
-### What Probably Won't Get Merged
-
-- Major architectural changes without discussion first
-- Features that significantly complicate the codebase
-- Anything that breaks backward compatibility without a really good reason
+## Making Changes
 
-## Code Style
+Follow the pattern in CLAUDE.md for adding new features — types in `shared/types.ts`, queries in `electron/db/`, IPC handler in `electron/ipc.ts`, wired up in `preload.ts` and `src/lib/api.ts`, then the React component in `src/features/`.
 
-I use ruff for formatting and linting, so just run `just lint` and fix any issues it finds. The existing code style is pretty straightforward - try to match what's already there.
+Run `npm run typecheck` and `npm run lint` before submitting.
 
 ## Submitting Changes
 
-1. Create a branch for your changes
-2. Make your changes and test them
-3. Run `just lint` to make sure everything looks good
-4. Open a pull request with a clear description of what you changed and why
-
-No need for fancy commit message formats or elaborate PR templates. Just explain what you did in plain English.
-
-## Questions or Ideas?
-
-Feel free to open an issue if you want to discuss something before diving in. I'm usually pretty responsive, though sometimes it might take a few days if I'm busy with other stuff.
-
-## License
-
-By contributing, you agree that your contributions will be licensed under the same license as the project (check the LICENSE file).
-
----
-
-Thanks for considering contributing! Even small improvements are appreciated. 🙂
+Open a pull request with a plain description of what you changed and why. No fancy templates needed.