taster 3.1.0__tar.gz

taster-3.1.0/.claude/settings.local.json

@@ -0,0 +1,48 @@
+{
+  "permissions": {
+    "allow": [
+      "Bash(python:*)",
+      "Bash(nul)",
+      "Bash(for dir in \"E:\\Dropbox\\Camera Roll\\2025_sorted\"/*)",
+      "Bash(done)",
+      "Bash(xargs -I {} python -c \"from PIL import Image; print(Image.open(''''{}'''').size)\")",
+      "WebFetch(domain:ai.google.dev)",
+      "Bash(mkdir:*)",
+      "Bash(git mv test_*.py analyze_missing_files.py check_babygate_cache.py check_gemini_cache_babygate.py find_failed_embeddings.py find_missing_files.py find_missing_videos.py trace_missing_files.py missing_files_list.txt old/tests/)",
+      "Bash(mv:*)",
+      "Bash(git add:*)",
+      "Bash(git commit:*)",
+      "Bash(dir /s \"E:\\OneDrive\\Evie and Elle\\LLM Taste Cloner\" /b)",
+      "Bash(dir:*)",
+      "Bash(tree:*)",
+      "Bash(tar:*)",
+      "Bash(powershell -Command:*)",
+      "Bash(powershell -ExecutionPolicy Bypass -File \"E:\\OneDrive\\Evie and Elle\\Sommelier\\create_zip.ps1\")",
+      "WebSearch",
+      "Bash(cat:*)",
+      "Bash(git push:*)",
+      "Bash(gh pr create --title \"Add error tracking and automatic retry for classification failures\" --body \"$\\(cat <<''EOF''\n## Summary\n\n- **Error Tracking**: Failed classifications now include structured error metadata \\(`is_error_fallback`, `error_type`, `error_message`, `retry_count`\\) enabling easy identification and reprocessing\n- **Automatic Retry**: API failures \\(timeouts, rate limits, connection errors\\) trigger up to 2 automatic retries with exponential backoff \\(2s, 4s\\)\n- **Burst-Aware Reprocessing**: New `clear_failed_for_retry.py` script clears entire bursts when any photo in the burst failed, preserving grouping context\n- **Context-Aware Prompts**: When retrying burst photos as singletons, includes context about sibling classifications\n\n## Changes\n\n### Core Classification\n- `src/classification/classifier.py`: Added `_create_fallback_response\\(\\)` with error fields, `_execute_with_retry\\(\\)` wrapper, burst context storage\n- `src/classification/prompt_builder.py`: Added `build_singleton_with_burst_context\\(\\)` for context-aware retry\n\n### Configuration\n- `src/core/config.py`: Added `classification_retries`, `retry_delay_seconds`, `retry_on_errors` settings\n- `config.yaml`: Added retry configuration section\n- `src/core/cache.py`: Added `burst_context` cache type\n\n### Reporting\n- `taste_classify.py`: Updated CSV reports to include error tracking columns\n\n### New Scripts\n- `clear_failed_for_retry.py`: Identifies photos with error fallbacks, clears cache and sorted folders for re-classification\n- `reprocess_failed.py`: Standalone reprocessing with burst context awareness\n\n### Documentation & Tests\n- `README.md`: Added Error Tracking & Retry section, updated version to 2.2.0\n- `tests/test_error_tracking.py`: 20+ unit tests for error tracking and retry functionality\n\n## Test plan\n\n- [x] Unit tests pass \\(`pytest tests/test_error_tracking.py -v`\\)\n- [ ] Run `clear_failed_for_retry.py` with `--dry-run` on a sorted folder with known failures\n- [ ] Clear failed photos and verify re-classification works correctly\n- [ ] Verify CSV reports include new error tracking columns\n\n🤖 Generated with [Claude Code]\\(https://claude.com/claude-code\\)\nEOF\n\\)\")",
+      "Bash(gh pr view:*)",
+      "Bash(gh pr diff:*)",
+      "Bash(gh pr merge:*)",
+      "Bash(pip install:*)",
+      "Bash(curl:*)",
+      "Bash(taskkill:*)",
+      "Bash(pkill:*)",
+      "Bash(tasklist:*)",
+      "Bash(gh repo view:*)",
+      "Bash(gh repo rename:*)",
+      "Bash(gh repo edit:*)",
+      "Bash(pip show:*)",
+      "Bash(printf:*)",
+      "Bash(echo:*)",
+      "Bash(git log:*)",
+      "Bash(find:*)",
+      "Bash(ls:*)",
+      "Bash(cd:*)",
+      "Bash(grep:*)"
+    ],
+    "deny": [],
+    "ask": []
+  }
+}

taster-3.1.0/.github/workflows/claude-code-review.yml

@@ -0,0 +1,57 @@
+name: Claude Code Review
+
+on:
+  pull_request:
+    types: [opened, synchronize]
+    # Optional: Only run on specific file changes
+    # paths:
+    #   - "src/**/*.ts"
+    #   - "src/**/*.tsx"
+    #   - "src/**/*.js"
+    #   - "src/**/*.jsx"
+
+jobs:
+  claude-review:
+    # Optional: Filter by PR author
+    # if: |
+    #   github.event.pull_request.user.login == 'external-contributor' ||
+    #   github.event.pull_request.user.login == 'new-developer' ||
+    #   github.event.pull_request.author_association == 'FIRST_TIME_CONTRIBUTOR'
+
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: read
+      issues: read
+      id-token: write
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+
+      - name: Run Claude Code Review
+        id: claude-review
+        uses: anthropics/claude-code-action@v1
+        with:
+          claude_code_oauth_token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
+          prompt: |
+            REPO: ${{ github.repository }}
+            PR NUMBER: ${{ github.event.pull_request.number }}
+
+            Please review this pull request and provide feedback on:
+            - Code quality and best practices
+            - Potential bugs or issues
+            - Performance considerations
+            - Security concerns
+            - Test coverage
+
+            Use the repository's CLAUDE.md for guidance on style and conventions. Be constructive and helpful in your feedback.
+
+            Use `gh pr comment` with your Bash tool to leave your review as a comment on the PR.
+
+          # See https://github.com/anthropics/claude-code-action/blob/main/docs/usage.md
+          # or https://docs.claude.com/en/docs/claude-code/cli-reference for available options
+          claude_args: '--allowed-tools "Bash(gh issue view:*),Bash(gh search:*),Bash(gh issue list:*),Bash(gh pr comment:*),Bash(gh pr diff:*),Bash(gh pr view:*),Bash(gh pr list:*)"'
+

taster-3.1.0/.github/workflows/claude.yml

@@ -0,0 +1,50 @@
+name: Claude Code
+
+on:
+  issue_comment:
+    types: [created]
+  pull_request_review_comment:
+    types: [created]
+  issues:
+    types: [opened, assigned]
+  pull_request_review:
+    types: [submitted]
+
+jobs:
+  claude:
+    if: |
+      (github.event_name == 'issue_comment' && contains(github.event.comment.body, '@claude')) ||
+      (github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '@claude')) ||
+      (github.event_name == 'pull_request_review' && contains(github.event.review.body, '@claude')) ||
+      (github.event_name == 'issues' && (contains(github.event.issue.body, '@claude') || contains(github.event.issue.title, '@claude')))
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: read
+      issues: read
+      id-token: write
+      actions: read # Required for Claude to read CI results on PRs
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+
+      - name: Run Claude Code
+        id: claude
+        uses: anthropics/claude-code-action@v1
+        with:
+          claude_code_oauth_token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
+
+          # This is an optional setting that allows Claude to read CI results on PRs
+          additional_permissions: |
+            actions: read
+
+          # Optional: Give a custom prompt to Claude. If this is not specified, Claude will perform the instructions specified in the comment that tagged it.
+          # prompt: 'Update the pull request description to include a summary of changes.'
+
+          # Optional: Add claude_args to customize behavior and configuration
+          # See https://github.com/anthropics/claude-code-action/blob/main/docs/usage.md
+          # or https://docs.claude.com/en/docs/claude-code/cli-reference for available options
+          # claude_args: '--allowed-tools Bash(gh pr:*)'
+

taster-3.1.0/.github/workflows/publish.yml

@@ -0,0 +1,32 @@
+name: Publish to PyPI
+
+on:
+  push:
+    tags: ["v*"]
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    permissions:
+      id-token: write   # Trusted Publishing (OIDC)
+      contents: write   # GitHub Release creation
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install build tools
+        run: pip install build
+
+      - name: Build sdist and wheel
+        run: python -m build
+
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1
+
+      - name: Create GitHub Release
+        uses: softprops/action-gh-release@v2
+        with:
+          generate_release_notes: true

taster-3.1.0/.gitignore

@@ -0,0 +1,53 @@
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+env/
+venv/
+ENV/
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# IDEs
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# OS
+.DS_Store
+Thumbs.db
+desktop.ini
+
+# Environment variables
+.env
+.env.local
+
+# Project specific
+.taste_cache/
+old/
+archive/
+Share/
+Storage/
+Review/
+training_data/
+*.log
+*.pkl
+*.json.bak
+.coverage

taster-3.1.0/CHANGELOG.md

@@ -0,0 +1,38 @@
+# Changelog
+
+## 3.0.0 (2026-02-17)
+
+**Sommelier is now a distributable PyPI package.**
+
+### Breaking Changes
+
+- **Package renamed:** `src/` is now `sommelier/`. All imports change from `from src.*` to `from sommelier.*`.
+- **`requirements.txt` removed.** Dependencies are now declared in `pyproject.toml` with optional groups.
+- **`setup_sommelier.py` removed.** Replaced by `sommelier init`.
+
+### Added
+
+- **`pip install sommelier`** -- proper Python package with `pyproject.toml` (hatchling build).
+- **CLI entry point** -- `sommelier classify`, `sommelier train`, `sommelier serve`, `sommelier status`, `sommelier init`.
+- **Platform-aware config directory** -- `%APPDATA%\sommelier` (Windows), `~/Library/Application Support/sommelier` (macOS), `$XDG_CONFIG_HOME/sommelier` (Linux).
+- **Config resolution order** -- explicit path > `./config.yaml` > user config dir > built-in defaults. Sommelier works out of the box with zero config.
+- **Optional dependency groups** -- `pip install sommelier[gemini]`, `sommelier[ml]`, `sommelier[all]`, etc. Clear error messages when an optional dep is missing.
+- **`sommelier/compat.py`** -- `require()` helper for graceful optional dependency handling.
+- **`sommelier/dirs.py`** -- platform-aware directory resolution for config, profiles, and .env.
+- **New tests** -- `test_dirs.py`, `test_cli.py`, `test_compat.py`, `test_config_resolution.py`.
+
+### Changed
+
+- `load_config()` returns defaults when no config file is found (previously raised `FileNotFoundError`).
+- Standalone scripts (`taste_classify.py`, `mcp_server.py`, etc.) still work as thin wrappers for repo-based usage.
+- README updated for cross-platform paths and `pip install` workflow.
+
+### Migration
+
+If you were using Sommelier from a git clone:
+
+1. Pull the latest code.
+2. Run `pip install -e ".[gemini]"` (or whichever provider you use).
+3. Replace `py -3.12 taste_classify.py` with `sommelier classify`.
+4. Replace `py -3.12 setup_sommelier.py` with `sommelier init`.
+5. If you import from Sommelier in your own code, change `from src.*` to `from sommelier.*`.