markback 0.1.0__tar.gz

markback-0.1.0/.gitignore

@@ -0,0 +1,207 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[codz]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py.cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#uv.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+#poetry.toml
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#   pdm recommends including project-wide configuration in pdm.toml, but excluding .pdm-python.
+#   https://pdm-project.org/en/latest/usage/project/#working-with-version-control
+#pdm.lock
+#pdm.toml
+.pdm-python
+.pdm-build/
+
+# pixi
+#   Similar to Pipfile.lock, it is generally recommended to include pixi.lock in version control.
+#pixi.lock
+#   Pixi creates a virtual environment in the .pixi directory, just like venv module creates one
+#   in the .venv directory. It is recommended not to include this directory in version control.
+.pixi
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.envrc
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+# Abstra
+# Abstra is an AI-powered process automation framework.
+# Ignore directories containing user credentials, local state, and settings.
+# Learn more at https://abstra.io/docs
+.abstra/
+
+# Visual Studio Code
+#  Visual Studio Code specific template is maintained in a separate VisualStudioCode.gitignore 
+#  that can be found at https://github.com/github/gitignore/blob/main/Global/VisualStudioCode.gitignore
+#  and can be added to the global gitignore or merged into this file. However, if you prefer, 
+#  you could uncomment the following to ignore the entire vscode folder
+# .vscode/
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+# Cursor
+#  Cursor is an AI-powered code editor. `.cursorignore` specifies files/directories to
+#  exclude from AI features like autocomplete and code analysis. Recommended for sensitive data
+#  refer to https://docs.cursor.com/context/ignore-files
+.cursorignore
+.cursorindexingignore
+
+# Marimo
+marimo/_static/
+marimo/_lsp/
+__marimo__/

markback-0.1.0/IMPLEMENTATION_NOTES.md

@@ -0,0 +1,119 @@
+# Implementation Notes
+
+## Design Decisions
+
+### Parser Architecture
+
+The parser uses a state-machine approach, processing lines sequentially and classifying each line into one of:
+- `COMPACT_RECORD` - `@source ... <<<` on one line
+- `HEADER` - `@keyword value`
+- `FEEDBACK` - `<<< ...`
+- `SEPARATOR` - `---`
+- `BLANK` - empty line
+- `CONTENT` - anything else
+
+This allows handling all format variants (single, multi, compact, paired) with the same core logic.
+
+**Tradeoff:** The parser is single-pass but accumulates state. An alternative two-pass approach (first split on separators, then parse each segment) was considered but rejected because compact records don't require separators.
+
+### Feedback Parsing
+
+Feedback parsing supports three modes:
+1. **Raw** - Return the feedback string as-is
+2. **Structured** - Parse into label, attributes, and comment
+3. **JSON** - Parse `json:{...}` prefixed feedback
+
+The structured parser splits on `; ` (semicolon + space) and classifies segments:
+- Segments with `=` are key-value attributes
+- First non-attribute segment is the label
+- Subsequent non-attribute segments become the comment
+
+**Tradeoff:** This heuristic-based parsing may occasionally misclassify freeform text that happens to contain `=`. A stricter approach would require escaping, but that conflicts with the "easy to type" design goal.
+
+### Compact Record Detection
+
+A line is classified as a compact record if it:
+1. Starts with `@source`
+2. Contains `<<<`
+
+This is done before checking if it's a regular header to ensure compact records are handled correctly.
+
+**Edge case:** A `@source` path containing `<<<` would be misinterpreted. This is documented as a limitation - paths should not contain the feedback delimiter.
+
+### Paired File Discovery
+
+Paired files are discovered by:
+1. Finding all files in a directory
+2. Identifying label files by suffix (`.label.txt`, `.feedback.txt`, `.mb`)
+3. Matching content files to label files by basename
+
+**Tradeoff:** This simple approach doesn't handle nested directories or complex naming patterns. A future version could support glob patterns or manifest files.
+
+### Writer Canonical Format
+
+The writer produces deterministic output by:
+1. Normalizing line endings to LF
+2. Ordering headers (`@uri` before `@source`)
+3. Trimming trailing whitespace
+4. Using compact format for source-only records when `prefer_compact=True`
+
+**Tradeoff:** The compact format preference is configurable because some users may prefer the more explicit full format even for simple records.
+
+### LLM Abstraction
+
+The `LLMClient` abstraction supports:
+- OpenAI-compatible APIs (most common)
+- Mock client for testing
+
+The factory pattern allows injecting mock clients during tests without modifying the workflow code.
+
+**Tradeoff:** Only synchronous HTTP is supported. Async support was considered but adds complexity without clear benefit for the typical use case (small datasets, infrequent calls).
+
+### Evaluation Heuristics
+
+The v1 evaluation uses simple heuristics:
+- Parse the expected feedback for a label
+- Check if the label is in `positive_labels` or `negative_labels`
+- Look for sentiment indicators in the operator output
+
+**Tradeoff:** This is intentionally simple and deterministic. A more sophisticated approach would use an LLM for evaluation, but that adds cost and non-determinism. The simple approach is easy to test and understand.
+
+### File Mode Handling
+
+Two modes are supported:
+- **git** - Modify files in place (suitable for version-controlled projects)
+- **versioned** - Never overwrite, create timestamped versions
+
+**Tradeoff:** The versioned mode uses timestamps rather than sequential numbers. This ensures uniqueness without needing to scan existing files, but timestamps can be less intuitive for ordering.
+
+## Known Limitations
+
+1. **Path restrictions:** Source paths cannot contain the `<<<` delimiter
+2. **Binary content:** Binary files are referenced but not embedded
+3. **Encoding:** Only UTF-8 is supported
+4. **Line classification:** Content cannot start with `@source ... <<<` on the first line after headers without a blank line
+5. **Evaluation:** Simple heuristic-based, not semantic understanding
+
+## Testing Strategy
+
+### Unit Tests
+- Parser tests cover all format variants from the spec
+- Writer tests verify roundtrip stability
+- Linter tests cover all error and warning codes
+- Type tests verify core data structures
+
+### Integration Tests
+- CLI tests use Typer's test runner
+- Workflow tests use mock LLM clients
+- File operations use temporary directories
+
+### Fixtures
+All spec examples are included as test fixtures to ensure the implementation matches the specification.
+
+## Future Considerations
+
+1. **Async HTTP:** Could improve throughput for large datasets
+2. **Streaming parser:** For very large files
+3. **Diff output:** Show what normalization changed
+4. **Watch mode:** Auto-lint on file changes
+5. **IDE integration:** LSP server for real-time linting

markback-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 dandriscoll
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

markback-0.1.0/PKG-INFO

@@ -0,0 +1,251 @@
+Metadata-Version: 2.4
+Name: markback
+Version: 0.1.0
+Summary: A compact, human-writable format for storing content paired with feedback/labels
+Project-URL: Homepage, https://github.com/dandriscoll/markback
+Project-URL: Repository, https://github.com/dandriscoll/markback
+Project-URL: Documentation, https://github.com/dandriscoll/markback#readme
+Project-URL: Issues, https://github.com/dandriscoll/markback/issues
+Author: Dan Driscoll
+License-Expression: MIT
+License-File: LICENSE
+Keywords: annotation,data-labeling,feedback,labeling,llm,markdown
+Classifier: Development Status :: 3 - Alpha
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Software Development :: Quality Assurance
+Classifier: Topic :: Text Processing :: Markup
+Requires-Python: >=3.10
+Requires-Dist: httpx>=0.25.0
+Requires-Dist: python-dotenv>=1.0.0
+Requires-Dist: rich>=13.0.0
+Requires-Dist: typer>=0.9.0
+Provides-Extra: dev
+Requires-Dist: build>=1.0.0; extra == 'dev'
+Requires-Dist: pytest-cov>=4.0.0; extra == 'dev'
+Requires-Dist: pytest>=7.0.0; extra == 'dev'
+Requires-Dist: twine>=5.0.0; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# MarkBack
+
+A compact, human-writable format for storing content paired with feedback/labels.
+
+## Installation
+
+```bash
+pip install -e .
+```
+
+## Quick Start
+
+### Parse a MarkBack file
+
+```python
+from markback import parse_file, parse_string
+
+# Parse a file
+result = parse_file("labels.mb")
+for record in result.records:
+    print(f"{record.uri}: {record.feedback}")
+
+# Parse a string
+text = """
+@uri local:example
+
+Some content here.
+<<< positive; good quality
+"""
+result = parse_string(text)
+```
+
+### Write MarkBack files
+
+```python
+from markback import Record, SourceRef, write_file, OutputMode
+
+records = [
+    Record(feedback="good", uri="local:1", content="First item"),
+    Record(feedback="bad", uri="local:2", content="Second item"),
+]
+
+# Write multi-record file
+write_file("output.mb", records, mode=OutputMode.MULTI)
+
+# Write compact label list
+write_file("labels.mb", records, mode=OutputMode.COMPACT)
+```
+
+### Lint files
+
+```python
+from markback import lint_file
+
+result = lint_file("myfile.mb")
+if result.has_errors:
+    for d in result.diagnostics:
+        print(d)
+```
+
+## CLI Usage
+
+### Initialize configuration
+
+```bash
+markback init
+```
+
+Creates a `.env` file with all configuration options.
+
+### Lint files
+
+```bash
+# Lint a single file
+markback lint myfile.mb
+
+# Lint a directory
+markback lint ./data/
+
+# JSON output
+markback lint myfile.mb --json
+```
+
+### Normalize to canonical format
+
+```bash
+# Output to stdout
+markback normalize input.mb
+
+# Output to file
+markback normalize input.mb output.mb
+
+# In-place normalization
+markback normalize input.mb --in-place
+```
+
+### List records
+
+```bash
+markback list myfile.mb
+markback list ./data/ --json
+```
+
+### Convert between formats
+
+```bash
+# Convert to multi-record format
+markback convert input.mb output.mb --to multi
+
+# Convert to compact label list
+markback convert input.mb output.mb --to compact
+
+# Convert to paired files
+markback convert input.mb ./output_dir/ --to paired
+```
+
+### Run LLM workflow
+
+```bash
+# Run editor/operator workflow
+markback workflow run dataset.mb --prompt "Initial prompt" --output results.json
+
+# View evaluation results
+markback workflow evaluate results.json
+
+# Extract refined prompt
+markback workflow prompt results.json --output refined_prompt.txt
+```
+
+## File Formats
+
+### Single Record
+
+```
+@uri local:example
+
+Content goes here.
+<<< positive; quality=high
+```
+
+### Multi-Record
+
+```
+@uri local:item-1
+
+First content.
+<<< good
+
+---
+@uri local:item-2
+
+Second content.
+<<< bad; needs improvement
+```
+
+### Compact Label List
+
+```
+@source ./images/001.jpg <<< approved; scene=beach
+@source ./images/002.jpg <<< rejected; too dark
+@source ./images/003.jpg <<< approved; scene=mountain
+```
+
+### Paired Files
+
+**content.txt:**
+```
+The actual content goes here.
+```
+
+**content.label.txt:**
+```
+@uri local:content-id
+<<< approved; reviewer=alice
+```
+
+## Configuration
+
+Configuration is loaded from `.env`:
+
+```bash
+# File handling mode
+FILE_MODE=git  # or "versioned"
+
+# Label file discovery
+LABEL_SUFFIXES=.label.txt,.feedback.txt,.mb
+
+# Editor LLM
+EDITOR_API_BASE=https://api.openai.com/v1
+EDITOR_API_KEY=your-key
+EDITOR_MODEL=gpt-4
+
+# Operator LLM
+OPERATOR_API_BASE=https://api.openai.com/v1
+OPERATOR_API_KEY=your-key
+OPERATOR_MODEL=gpt-4
+```
+
+## Development
+
+### Run tests
+
+```bash
+pip install -e ".[dev]"
+pytest
+```
+
+### Run with coverage
+
+```bash
+pytest --cov=markback
+```
+
+## License
+
+MIT