datacompose 0.2.8__tar.gz

datacompose-0.2.8/CHANGELOG.md

@@ -0,0 +1,227 @@
+# Changelog
+
+All notable changes to Datacompose will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [0.2.7.0] - 2025-09-11
+
+### Fixed
+- **SHA256 Transformer Memory Issues**: Fixed Java heap space OutOfMemoryError in email and phone number SHA256 hashing
+  - Set `standardize_first=False` by default in tests to avoid complex Spark query planning issues
+  - All SHA256 hashing tests now pass without memory errors
+  
+- **CLI Configuration Handling**: Improved config file error handling in add command
+  - Add command now properly fails with helpful error message when no config file exists
+  - Add command correctly handles malformed JSON config files
+  - "pyspark" is now the default target when explicitly called without config
+  
+- **Test Fixtures**: Added missing `diverse_test_data` fixture for conditional operator tests
+  - Created comprehensive test dataset with category, value, size, id, and text columns
+  - Fixed all conditional logic tests in `test_conditional_core.py`
+  - Fixed all real-world scenario tests in `test_conditional_real_world.py`
+  
+- **Test Assertions**: Updated test expectations to match actual behavior
+  - Fixed init command test to expect full command in error message ("datacompose init --force")
+  - Updated conditional test assertions for non-standardized hashing behavior
+
+### Changed
+- **Default Target Behavior**: ConfigLoader now returns "pyspark" as fallback when no config is provided programmatically
+
+## [0.2.6.0] - 2025-08-24
+
+### Added
+- **Automatic Conditional Detection**: Smart detection of conditional operators based on naming patterns
+  - Functions starting with `is_`, `has_`, `needs_`, `should_`, `can_`, `contains_`, `matches_`, `equals_`, `starts_with_`, `ends_with_` are automatically detected as conditionals
+  - Eliminates need for explicit `is_conditional=True` in most cases
+  - Explicit override still available when needed via `is_conditional` parameter
+- **Phone Number Processing Pipeline**: Complete phone number validation and formatting example
+  - Letter-to-number conversion (1-800-FLOWERS)
+  - NANP validation and formatting
+  - Toll-free number detection
+  - E.164 and parentheses formatting
+
+### Changed
+- **Conditional Operator Registration**: `is_conditional` parameter now optional with smart defaults
+- **Test Organization**: Consolidated conditional tests into three focused files:
+  - `test_conditional_core.py` - Core functionality, logic, errors, parameters, and performance
+  - `test_conditional_real_world.py` - Real-world pipeline scenarios
+  - `test_conditional_auto_detection.py` - Auto-detection feature tests
+
+### Fixed
+- **Phone Number Validation**: Updated NANP validation to be more flexible for testing scenarios
+
+## [0.2.5.3] - 2025-08-23
+
+### Added
+- **Compose Decorator Enhancement**: Auto-detection of PrimitiveRegistry instances in function globals
+  - Compose decorator now automatically discovers all namespace instances without explicit passing
+  - Improved namespace resolution using function's global scope instead of module globals
+  - Better support for multiple namespaces in composed functions
+
+### Fixed
+- **Namespace Resolution**: Fixed global namespace lookups to use function's own globals
+  - PipelineCompiler now correctly resolves namespaces from the decorated function's scope
+  - Fallback compose mode uses function globals for namespace discovery
+  - Prevents namespace resolution errors when registries are defined in different modules
+
+### Changed
+- **Phone Number Tests**: Updated test imports and formatting for phone number primitives
+- **Test Organization**: Added comprehensive conditional composition tests
+
+## [0.2.5.2] - 2025-08-22
+
+### Fixed
+- **Import Paths**: Updated import paths in phone_numbers pyspark primitives for clarity and consistency
+- **Documentation**: Improved docstring documentation across primitives
+
+## [0.2.5.1] - 2025-08-22
+
+### Changed
+- **Import Paths**: Renamed imports to be more transparent and clear
+
+### Added
+- **Documentation**: Added clear module-level docstrings throughout the codebase
+- **Unit Tests**: Added comprehensive unit tests for default initialization and datacompose.json configuration
+  - Tests for default target auto-selection with single target
+  - Tests for explicit target override behavior
+  - Tests for configuration file validation
+  - Tests for output path resolution from config
+
+### Fixed
+- **CLI Tests**: Fixed all failing default target configuration tests
+  - Added proper validation mocks for non-existent platforms in tests
+  - Fixed error message assertion for invalid platform validation
+  - Properly mocked generator class hierarchy for output path testing
+  - All 13 CLI default target tests now passing (100% pass rate)
+
+## [0.2.5] - 2025-08-21
+
+### Changed
+- **Documentation**: Streamlined README to be more concise
+  - Removed extensive code examples (now on website)
+  - Reduced from 390 lines to 44 lines
+  - Focused on core features and philosophy
+  - Added link to datacompose.io for detailed documentation
+
+### Fixed
+- **Test Suite**: Fixed failing CLI tests for `add` command
+  - Tests now properly mock ConfigLoader for isolated filesystem environments
+  - `test_add_invalid_transformer` correctly validates transformer not found error
+  - `test_complete_transformer_success` updated to match actual transformer names
+  - All CLI command tests passing with proper configuration mocking
+
+## [0.2.4] - 2025-08-13
+
+### Added
+- **Published to PyPI**: Package is now available via `pip install datacompose`
+- **Phone Number Primitives**: Complete set of 45+ phone number transformation functions
+  - NANP validation and formatting (North American Numbering Plan)
+  - International phone support with E.164 formatting
+  - Extension handling and toll-free detection
+  - Phone number extraction from text
+  - Letter-to-number conversion (1-800-FLOWERS support)
+- **Address Improvements**: Enhanced street extraction and standardization
+  - Fixed numbered street extraction ("5th Avenue" correctly returns "5th")
+  - Improved null handling in street extraction
+  - Custom mapping support for street suffix standardization
+- **Utils Export**: Generated code now includes `utils/primitives.py` for standalone deployment
+  - PrimitiveRegistry class embedded with generated code
+  - No runtime dependency on datacompose package
+  - Fallback imports for maximum compatibility
+- **Comprehensive Test Coverage**: Improved test coverage from 87% to 92%
+  - Added 18 new tests for primitives.py module (70% → 86% coverage)
+  - Created comprehensive test suites for all CLI commands
+  - Added full end-to-end integration tests (init → add → transform)
+  - validation.py achieved 100% coverage
+  - add.py improved to 99% coverage
+
+### Changed
+- **BREAKING**: Renamed `PrimitiveNameSpace` to `PrimitiveRegistry` throughout codebase
+- **Major Architecture Shift**: Removed YAML/spec file system entirely
+  - No more YAML specifications or JSON replacements
+  - Direct primitive file copying instead of template rendering
+  - Simplified discovery system works with transformer directories
+  - Removed `validate` command completely
+- **Import Strategy**: Primitives now try local utils import first, fall back to datacompose package
+- **File Naming**: Generated files use plural form with primitives suffix
+  - `emails` → `email_primitives.py`
+  - `addresses` → `address_primitives.py`
+  - `phone_numbers` → `phone_primitives.py`
+
+### Fixed
+- **Critical**: Fixed utils/primitives.py output location to be shared across all transformers
+  - Utils module now generates at top-level build/utils/ instead of per-transformer
+  - All transformers share the same PrimitiveRegistry implementation
+  - Prevents duplicate utils modules and ensures consistency
+- Phone `normalize_separators` now correctly handles parentheses: `(555)123-4567` → `555-123-4567`
+- Street extraction for numbered streets ("5th Avenue" issue)
+- Compose decorator now requires namespace to be passed explicitly for proper method resolution
+- `standardize_street_suffix` applies both custom and default mappings correctly
+- Test failures due to namespace resolution in compose decorator
+- Generator initialization error handling in add command
+
+### Removed
+- All YAML/spec file functionality
+- PostgreSQL generator references
+- Jinja2 template dependencies
+- `validate` command from CLI
+- Old Spark integration tests (replaced with end-to-end tests)
+
+## [0.2.0] - 2024-XX-XX
+
+### Added
+- **Primitive System**: New composable primitive architecture for building data pipelines
+  - `SmartPrimitive` class for partial application and parameter binding
+  - `PrimitiveRegistry` (originally PrimitiveNameSpace) for organizing related transformations
+  - Support for conditional primitives (boolean-returning functions)
+- **Conditional Compilation**: AST-based pipeline compilation with if/else support
+  - `PipelineCompiler` for parsing and compiling conditional logic
+  - `StablePipeline` for executing compiled pipelines
+  - Full support for nested conditionals and complex branching
+- **Comprehensive Testing**: 44+ tests covering conditional compilation scenarios
+  - Edge cases and null handling
+  - Complex nested logic
+  - Data-driven conditions
+  - Performance optimization tests
+  - Real-world use cases
+  - Parameter handling
+  - Error handling
+- **Improved Architecture**: Dual approach for different runtime constraints
+  - Primitives for flexible runtimes (Python, Spark, Scala)
+  - Templates for rigid targets (SQL, PostgreSQL)
+
+### Changed
+- Made PySpark an optional dependency
+- Reorganized test structure with focused test files and shared fixtures
+- Refined architecture to support both template-based and primitive-based approaches
+
+### Fixed
+- Import paths for pipeline compilation modules
+- Missing return statements in pipeline execution
+- Conditional logic to use accumulated results correctly
+
+## [0.1.4] - 2024-XX-XX
+
+### Added
+- Initial release of Datacompose
+- Core framework for generating data cleaning UDFs
+- Support for Spark, PostgreSQL, and Pandas targets
+- Built-in specifications for common data cleaning tasks:
+  - Email address cleaning
+  - Phone number normalization
+  - Address standardization
+  - Job title standardization
+  - Date/time parsing
+- CLI interface with commands:
+  - `datacompose init` - Initialize project
+  - `datacompose add` - Generate UDFs from specs
+  - `datacompose list` - List available targets and specs
+  - `datacompose validate` - Validate specification files
+- YAML-based specification format
+- Jinja2 templating for code generation
+- Comprehensive test suite
+- Documentation with Sphinx and Furo theme

datacompose-0.2.8/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 datacompose
+
+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.

datacompose-0.2.8/MANIFEST.in

@@ -0,0 +1,38 @@
+# Include documentation
+include README.md
+include LICENSE
+include CHANGELOG.md
+
+# Include all YAML specifications
+recursive-include datacompose/transformers *.yaml
+
+# Include all Jinja2 templates
+recursive-include datacompose/transformers *.j2
+recursive-include datacompose/generators *.j2
+
+# Include type hints
+recursive-include datacompose py.typed
+
+# Include test data (optional, remove if you don't want tests in distribution)
+recursive-include tests *.py
+recursive-include tests *.csv
+recursive-include tests *.yaml
+
+# Exclude unnecessary files
+global-exclude *.pyc
+global-exclude *.pyo
+global-exclude __pycache__
+global-exclude .DS_Store
+global-exclude .git*
+global-exclude *.swp
+global-exclude *~
+
+# Exclude development files
+exclude .pre-commit-config.yaml
+exclude .gitignore
+exclude docker-compose.yml
+exclude Dockerfile
+exclude Makefile
+prune docs/build
+prune notebooks
+prune scripts

datacompose-0.2.8/PKG-INFO

@@ -0,0 +1,176 @@
+Metadata-Version: 2.4
+Name: datacompose
+Version: 0.2.8
+Summary: Copy-pasteable data transformation primitives for PySpark. Inspired by shadcn-svelte.
+Author: Datacompose Contributors
+Maintainer: Datacompose Contributors
+License: MIT
+Project-URL: Homepage, https://github.com/tc-cole/datacompose
+Project-URL: Documentation, https://github.com/tc-cole/datacompose/tree/main/docs
+Project-URL: Repository, https://github.com/tc-cole/datacompose.git
+Project-URL: Issues, https://github.com/tc-cole/datacompose/issues
+Project-URL: Changelog, https://github.com/tc-cole/datacompose/blob/main/CHANGELOG.md
+Keywords: data-cleaning,data-quality,udf,spark,postgres,code-generation,data-pipeline,etl
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Topic :: Software Development :: Code Generators
+Classifier: Topic :: Database
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: jinja2>=3.0.0
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: click>=8.0.0
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Requires-Dist: black>=23.0.0; extra == "dev"
+Requires-Dist: mypy>=1.0.0; extra == "dev"
+Requires-Dist: ruff>=0.1.0; extra == "dev"
+Provides-Extra: docs
+Requires-Dist: mkdocs>=1.5.3; extra == "docs"
+Requires-Dist: mkdocs-material>=9.5.0; extra == "docs"
+Requires-Dist: mkdocs-material-extensions>=1.3; extra == "docs"
+Requires-Dist: mkdocs-minify-plugin>=0.7.1; extra == "docs"
+Requires-Dist: mkdocs-redirects>=1.2.1; extra == "docs"
+Requires-Dist: mike>=2.0.0; extra == "docs"
+Requires-Dist: pymdown-extensions>=10.5; extra == "docs"
+Requires-Dist: pygments>=2.17.0; extra == "docs"
+Requires-Dist: mkdocs-git-revision-date-localized-plugin>=1.2.2; extra == "docs"
+Requires-Dist: mkdocs-glightbox>=0.3.5; extra == "docs"
+Dynamic: license-file
+
+# DataCompose
+
+PySpark transformations you can actually own and modify. No black boxes.
+
+## Before vs After
+
+```python
+# Before: Regex nightmare for addresses
+df = df.withColumn("state_clean",
+    F.when(F.col("address").rlike(".*\\b(NY|N\\.Y\\.|New York|NewYork|Newyork)\\b.*"), "NY")
+    .when(F.col("address").rlike(".*\\b(CA|Cal\\.|Calif\\.|California)\\b.*"), "CA")
+    .when(F.col("address").rlike(".*\\b(IL|Ill\\.|Illinois|Illinios)\\b.*"), "IL")
+    .when(F.upper(F.col("address")).contains("NEW YORK"), "NY")
+    .when(F.regexp_extract(F.col("address"), ",\\s*([A-Z]{2})\\s+\\d{5}", 1) == "NY", "NY")
+    .when(F.regexp_extract(F.col("address"), "\\s+([A-Z]{2})\\s*$", 1) == "NY", "NY")
+    # ... handle "N.Y 10001" vs "NY, 10001" vs "New York 10001"
+    # ... handle misspellings like "Californai" or "Illnois"  
+    # ... 50 more states × 10 variations each
+)
+
+# After: One line
+from builders.transformers.addresses import addresses
+df = df.withColumn("state", addresses.standardize_state(F.col("address")))
+```
+
+## Installation
+
+```bash
+pip install datacompose
+```
+
+## How It Works
+
+```bash
+# Copy transformers into YOUR repo
+datacompose add phones
+datacompose add addresses
+datacompose add emails
+```
+
+```python
+# Use them like any Python module - this is your code now
+from transformers.pyspark.addresses import addresses
+
+df = (df
+    .withColumn("street_number", addresses.extract_street_number(F.col("address")))
+    .withColumn("street_name", addresses.extract_street_name(F.col("address")))
+    .withColumn("city", addresses.extract_city(F.col("address")))
+    .withColumn("state", addresses.standardize_state(F.col("address")))
+    .withColumn("zip", addresses.extract_zip_code(F.col("address")))
+)
+
+# Result:
++----------------------------------------+-------------+------------+-----------+-----+-------+
+|address                                 |street_number|street_name |city       |state|zip    |
++----------------------------------------+-------------+------------+-----------+-----+-------+
+|123 Main St, New York, NY 10001        |123          |Main        |New York   |NY   |10001  |
+|456 Oak Ave Apt 5B, Los Angeles, CA 90001|456        |Oak         |Los Angeles|CA   |90001  |
+|789 Pine Blvd, Chicago, IL 60601       |789          |Pine        |Chicago    |IL   |60601  |
++----------------------------------------+-------------+------------+-----------+-----+-------+
+```
+
+The code lives in your repo. Modify it. Delete what you don't need. No external dependencies.
+
+## Why Copy-to-Own?
+
+- **Your data is weird** - Phone numbers with "ask for Bob"? We can't predict that. You can fix it.
+- **No breaking changes** - Library updates can't break your pipeline at 2 AM
+- **Actually debuggable** - Stack traces point to YOUR code, not site-packages
+- **No dependency hell** - It's just PySpark. If Spark runs, this runs.
+
+## Available Transformers
+
+**Phones** - Standardize formats, extract from text, validate, handle extensions  
+**Addresses** - Parse components, standardize states, validate zips, detect PO boxes  
+**Emails** - Validate, extract domains, fix typos (gmial→gmail), standardize
+
+More coming based on what you need.
+
+## Real Example
+
+```python
+# Messy customer data
+df = spark.createDataFrame([
+    ("(555) 123-4567 ext 89", "john.doe@gmial.com", "123 Main St Apt 4B"),
+    ("555.987.6543", "JANE@COMPANY.COM", "456 Oak Ave, NY, NY 10001")
+])
+
+# Clean it
+clean_df = (df
+    .withColumn("phone", phones.standardize_phone(F.col("phone")))
+    .withColumn("email", emails.fix_common_typos(F.col("email")))
+    .withColumn("street", addresses.extract_street_address(F.col("address")))
+)
+```
+
+## The Philosophy
+
+```
+█████████████ 60% - Already clean
+████████ 30% - Common patterns (formatting, typos)
+██ 8% - Edge cases (weird but fixable)
+▌ 2% - Complete chaos (that's what interns are for)
+```
+
+We handle the 38% with patterns. You handle the 2% chaos.
+
+## Documentation
+
+Full docs at [datacompose.io](https://datacompose.io)
+
+## Key Features
+
+- **Zero dependencies** - Just PySpark code that runs anywhere Spark runs
+- **Fully modifiable** - It's in your repo. Change whatever you need
+- **Battle-tested patterns** - Built from real production data cleaning challenges  
+- **Composable functions** - Chain simple operations into complex pipelines
+- **No breaking changes** - You control when and how to update
+
+## License
+
+MIT - It's your code now.
+
+---
+
+*Inspired by [shadcn/ui](https://ui.shadcn.com/) and [Svelte](https://svelte.dev/)'s approach to components - copy, don't install.*

datacompose-0.2.8/README.md

@@ -0,0 +1,126 @@
+# DataCompose
+
+PySpark transformations you can actually own and modify. No black boxes.
+
+## Before vs After
+
+```python
+# Before: Regex nightmare for addresses
+df = df.withColumn("state_clean",
+    F.when(F.col("address").rlike(".*\\b(NY|N\\.Y\\.|New York|NewYork|Newyork)\\b.*"), "NY")
+    .when(F.col("address").rlike(".*\\b(CA|Cal\\.|Calif\\.|California)\\b.*"), "CA")
+    .when(F.col("address").rlike(".*\\b(IL|Ill\\.|Illinois|Illinios)\\b.*"), "IL")
+    .when(F.upper(F.col("address")).contains("NEW YORK"), "NY")
+    .when(F.regexp_extract(F.col("address"), ",\\s*([A-Z]{2})\\s+\\d{5}", 1) == "NY", "NY")
+    .when(F.regexp_extract(F.col("address"), "\\s+([A-Z]{2})\\s*$", 1) == "NY", "NY")
+    # ... handle "N.Y 10001" vs "NY, 10001" vs "New York 10001"
+    # ... handle misspellings like "Californai" or "Illnois"  
+    # ... 50 more states × 10 variations each
+)
+
+# After: One line
+from builders.transformers.addresses import addresses
+df = df.withColumn("state", addresses.standardize_state(F.col("address")))
+```
+
+## Installation
+
+```bash
+pip install datacompose
+```
+
+## How It Works
+
+```bash
+# Copy transformers into YOUR repo
+datacompose add phones
+datacompose add addresses
+datacompose add emails
+```
+
+```python
+# Use them like any Python module - this is your code now
+from transformers.pyspark.addresses import addresses
+
+df = (df
+    .withColumn("street_number", addresses.extract_street_number(F.col("address")))
+    .withColumn("street_name", addresses.extract_street_name(F.col("address")))
+    .withColumn("city", addresses.extract_city(F.col("address")))
+    .withColumn("state", addresses.standardize_state(F.col("address")))
+    .withColumn("zip", addresses.extract_zip_code(F.col("address")))
+)
+
+# Result:
++----------------------------------------+-------------+------------+-----------+-----+-------+
+|address                                 |street_number|street_name |city       |state|zip    |
++----------------------------------------+-------------+------------+-----------+-----+-------+
+|123 Main St, New York, NY 10001        |123          |Main        |New York   |NY   |10001  |
+|456 Oak Ave Apt 5B, Los Angeles, CA 90001|456        |Oak         |Los Angeles|CA   |90001  |
+|789 Pine Blvd, Chicago, IL 60601       |789          |Pine        |Chicago    |IL   |60601  |
++----------------------------------------+-------------+------------+-----------+-----+-------+
+```
+
+The code lives in your repo. Modify it. Delete what you don't need. No external dependencies.
+
+## Why Copy-to-Own?
+
+- **Your data is weird** - Phone numbers with "ask for Bob"? We can't predict that. You can fix it.
+- **No breaking changes** - Library updates can't break your pipeline at 2 AM
+- **Actually debuggable** - Stack traces point to YOUR code, not site-packages
+- **No dependency hell** - It's just PySpark. If Spark runs, this runs.
+
+## Available Transformers
+
+**Phones** - Standardize formats, extract from text, validate, handle extensions  
+**Addresses** - Parse components, standardize states, validate zips, detect PO boxes  
+**Emails** - Validate, extract domains, fix typos (gmial→gmail), standardize
+
+More coming based on what you need.
+
+## Real Example
+
+```python
+# Messy customer data
+df = spark.createDataFrame([
+    ("(555) 123-4567 ext 89", "john.doe@gmial.com", "123 Main St Apt 4B"),
+    ("555.987.6543", "JANE@COMPANY.COM", "456 Oak Ave, NY, NY 10001")
+])
+
+# Clean it
+clean_df = (df
+    .withColumn("phone", phones.standardize_phone(F.col("phone")))
+    .withColumn("email", emails.fix_common_typos(F.col("email")))
+    .withColumn("street", addresses.extract_street_address(F.col("address")))
+)
+```
+
+## The Philosophy
+
+```
+█████████████ 60% - Already clean
+████████ 30% - Common patterns (formatting, typos)
+██ 8% - Edge cases (weird but fixable)
+▌ 2% - Complete chaos (that's what interns are for)
+```
+
+We handle the 38% with patterns. You handle the 2% chaos.
+
+## Documentation
+
+Full docs at [datacompose.io](https://datacompose.io)
+
+## Key Features
+
+- **Zero dependencies** - Just PySpark code that runs anywhere Spark runs
+- **Fully modifiable** - It's in your repo. Change whatever you need
+- **Battle-tested patterns** - Built from real production data cleaning challenges  
+- **Composable functions** - Chain simple operations into complex pipelines
+- **No breaking changes** - You control when and how to update
+
+## License
+
+MIT - It's your code now.
+
+---
+
+*Inspired by [shadcn/ui](https://ui.shadcn.com/) and [Svelte](https://svelte.dev/)'s approach to components - copy, don't install.*

datacompose-0.2.8/datacompose/__init__.py

@@ -0,0 +1 @@
+"""Datacompose source package."""

datacompose-0.2.8/datacompose/cli/__init__.py

@@ -0,0 +1,5 @@
+"""
+Datacompose CLI - Command-line interface for generating data cleaning UDFs.
+"""
+
+__version__ = "0.2.7.0"

datacompose-0.2.8/datacompose/cli/colors.py

@@ -0,0 +1,80 @@
+"""
+Simple color utilities for CLI output.
+"""
+
+import os
+import sys
+
+
+class Colors:
+    """ANSI color codes for terminal output."""
+
+    # Text colors
+    RED = "\033[91m"
+    GREEN = "\033[92m"
+    YELLOW = "\033[93m"
+    BLUE = "\033[94m"
+    MAGENTA = "\033[95m"
+    CYAN = "\033[96m"
+    WHITE = "\033[97m"
+    GRAY = "\033[90m"
+
+    # Styles
+    BOLD = "\033[1m"
+    DIM = "\033[2m"
+    UNDERLINE = "\033[4m"
+
+    # Reset
+    RESET = "\033[0m"
+
+    @classmethod
+    def is_enabled(cls) -> bool:
+        """Check if colors should be enabled."""
+        # Disable colors if NO_COLOR env var is set
+        if os.getenv("NO_COLOR"):
+            return False
+
+        # Disable colors if not in a TTY
+        if not sys.stdout.isatty():
+            return False
+
+        return True
+
+
+def colorize(text: str, color: str = "", style: str = "") -> str:
+    """Colorize text if colors are enabled."""
+    if not Colors.is_enabled():
+        return text
+
+    prefix = style + color
+    return f"{prefix}{text}{Colors.RESET}"
+
+
+def success(text: str) -> str:
+    """Green text for success messages."""
+    return colorize(text, Colors.GREEN, Colors.BOLD)
+
+
+def error(text: str) -> str:
+    """Red text for error messages."""
+    return colorize(text, Colors.RED, Colors.BOLD)
+
+
+def warning(text: str) -> str:
+    """Yellow text for warning messages."""
+    return colorize(text, Colors.YELLOW, Colors.BOLD)
+
+
+def info(text: str) -> str:
+    """Blue text for info messages."""
+    return colorize(text, Colors.BLUE)
+
+
+def highlight(text: str) -> str:
+    """Cyan text for highlighted text."""
+    return colorize(text, Colors.CYAN, Colors.BOLD)
+
+
+def dim(text: str) -> str:
+    """Dimmed text for less important info."""
+    return colorize(text, Colors.GRAY)

datacompose-0.2.8/datacompose/cli/commands/__init__.py

@@ -0,0 +1,3 @@
+"""
+CLI commands for Datacompose.
+"""