# Sakurs Python Bindings
High-performance sentence boundary detection for Python using the Delta-Stack Monoid algorithm.
## Table of Contents
- [Installation](#installation)
- [Quick Start](#quick-start)
- [API Reference](#api-reference)
- [Functions](#functions)
- [Classes](#classes)
- [Supported Languages](#supported-languages)
- [Performance Tips](#performance-tips)
- [Benchmarks](#benchmarks)
- [Error Handling](#error-handling)
- [Development](#development)
- [Building from Source](#building-from-source)
- [Development Workflow](#development-workflow)
- [Troubleshooting](#troubleshooting)
- [License](#license)
## Installation
**Coming Soon** - Will be available on PyPI.
For now, build from source:
```bash
git clone https://github.com/sog4be/sakurs.git
cd sakurs/sakurs-py
uv pip install -e .
```
**Requirements**: Python 3.9 or later
## Quick Start
```python
import sakurs
# Simple sentence splitting
sentences = sakurs.split("Hello world. This is a test.")
print(sentences) # ['Hello world.', 'This is a test.']
# Process files directly
sentences = sakurs.split("document.txt") # Path as string
sentences = sakurs.split(Path("document.txt")) # Path object
# Specify language
sentences = sakurs.split("これは日本語です。テストです。", language="ja")
print(sentences) # ['これは日本語です。', 'テストです。']
# Get detailed output with offsets
results = sakurs.split(text, return_details=True)
for sentence in results:
print(f"{sentence.text} [{sentence.start}:{sentence.end}]")
# Memory-efficient processing for large files
for sentence in sakurs.split_large_file("huge_corpus.txt", max_memory_mb=50):
process(sentence) # Process each sentence as it's found
# Responsive iteration (loads all, yields incrementally)
for sentence in sakurs.iter_split("document.txt"):
print(sentence) # Get results as they're processed
```
## API Reference
### Table of Contents
- [Functions](#functions)
- [`sakurs.split`](#sakurssplit)
- [`sakurs.iter_split`](#sakursiter_split)
- [`sakurs.split_large_file`](#sakurssplit_large_file)
- [`sakurs.load`](#sakursload)
- [`sakurs.supported_languages`](#sakurssupported_languages)
- [Classes](#classes)
- [`SentenceSplitter`](#sakurssentencesplitter)
- [`Sentence`](#sakurssentence)
- [`LanguageConfig`](#sakurslanguageconfig)
### Functions
#### `sakurs.split`
Split text or file into sentences.
**Signature:**
```python
sakurs.split(
input,
*,
language=None,
language_config=None,
threads=None,
chunk_kb=None,
parallel=False,
execution_mode="adaptive",
return_details=False,
encoding="utf-8"
)
```
**Parameters:**
- `input` (str | Path | TextIO | BinaryIO): Text string, file path, or file-like object
- `language` (str, optional): Language code ("en", "ja")
- `language_config` (LanguageConfig, optional): Custom language configuration
- `threads` (int, optional): Number of threads (None for auto)
- `chunk_kb` (int, optional): Chunk size in KB (default: 256) for parallel processing (default: 256)
- `parallel` (bool): Force parallel processing even for small inputs
- `execution_mode` (str): "sequential", "parallel", or "adaptive" (default)
- `return_details` (bool): Return Sentence objects with metadata instead of strings
- `encoding` (str): Text encoding for file inputs (default: "utf-8")
**Returns:** List[str] or List[Sentence] if return_details=True
#### `sakurs.iter_split`
Process input and return sentences as an iterator. Loads entire input but yields incrementally.
**Signature:**
```python
sakurs.iter_split(
input,
*,
language=None,
language_config=None,
threads=None,
chunk_kb=None,
encoding="utf-8"
)
```
**Parameters:** Same as `split()` except no `return_details` parameter
**Returns:** Iterator[str] - Iterator yielding sentences
#### `sakurs.split_large_file`
Process large files with limited memory usage.
**Signature:**
```python
sakurs.split_large_file(
file_path,
*,
language=None,
language_config=None,
max_memory_mb=100,
overlap_size=1024,
encoding="utf-8"
)
```
**Parameters:**
- `file_path` (str | Path): Path to the file
- `language` (str, optional): Language code
- `language_config` (LanguageConfig, optional): Custom language configuration
- `max_memory_mb` (int): Maximum memory to use in MB (default: 100)
- `overlap_size` (int): Bytes to overlap between chunks (default: 1024)
- `encoding` (str): File encoding (default: "utf-8")
**Returns:** Iterator[str] - Iterator yielding sentences
#### `sakurs.load`
Create a processor instance for repeated use.
**Signature:**
```python
sakurs.load(
language,
*,
threads=None,
chunk_kb=None,
execution_mode="adaptive"
)
```
**Parameters:**
- `language` (str): Language code ("en" or "ja")
- `threads` (int, optional): Number of threads
- `chunk_kb` (int, optional): Chunk size in KB (default: 256)
- `execution_mode` (str): Processing mode
**Returns:** SentenceSplitter instance
#### `sakurs.supported_languages`
Get list of supported languages.
**Signature:**
```python
sakurs.supported_languages()
```
**Returns:** List[str] - Supported language codes
### Classes
#### `sakurs.SentenceSplitter`
Main sentence splitter class for sentence boundary detection.
**Constructor Parameters:**
- `language` (str, optional): Language code
- `language_config` (LanguageConfig, optional): Custom language configuration
- `threads` (int, optional): Number of threads
- `chunk_kb` (int, optional): Chunk size in KB (default: 256)
- `execution_mode` (str): "sequential", "parallel", or "adaptive"
- `streaming` (bool): Enable streaming mode configuration
- `stream_chunk_mb` (int): Chunk size in MB for streaming mode
**Methods:**
- `split(input, *, return_details=False, encoding="utf-8")`: Split text or file into sentences
- `iter_split(input, *, encoding="utf-8")`: Return iterator over sentences
- `__enter__()` / `__exit__()`: Context manager support
#### `sakurs.Sentence`
Sentence with metadata (returned when `return_details=True`).
**Attributes:**
- `text` (str): The sentence text
- `start` (int): Character offset of sentence start
- `end` (int): Character offset of sentence end
- `confidence` (float): Confidence score (default: 1.0)
- `metadata` (dict): Additional metadata
#### `sakurs.LanguageConfig`
Language configuration for custom rules.
**Class Methods:**
- `from_toml(path)`: Load configuration from TOML file
- `to_toml(path)`: Save configuration to TOML file
**Attributes:**
- `code` (str): Language code
- `name` (str): Language name
- `terminators` (TerminatorConfig): Sentence terminator rules
- `ellipsis` (EllipsisConfig): Ellipsis handling rules
- `abbreviations` (AbbreviationConfig): Abbreviation rules
- `enclosures` (EnclosureConfig): Enclosure (quotes, parentheses) rules
- `suppression` (SuppressionConfig): Pattern suppression rules
## Supported Languages
- English (`en`, `english`)
- Japanese (`ja`, `japanese`)
## Performance Tips
1. **Choose the right function for your use case**:
```python
# For small to medium texts - use split()
sentences = sakurs.split(text)
# For responsive processing - use iter_split()
for sentence in sakurs.iter_split(document):
process_immediately(sentence)
# For huge files with memory constraints - use split_large_file()
for sentence in sakurs.split_large_file("10gb_corpus.txt", max_memory_mb=100):
index_sentence(sentence)
```
2. **Reuse SentenceSplitter instances**: Create once, use many times
```python
processor = sakurs.load("en", threads=4)
for document in documents:
sentences = processor.split(document)
```
3. **Configure for your workload**:
```python
# For CPU-bound batch processing
processor = sakurs.load("en", threads=8, execution_mode="parallel")
# For I/O-bound or interactive use
processor = sakurs.load("en", threads=2, execution_mode="adaptive")
# For memory-constrained environments
processor = sakurs.SentenceSplitter(language="en", streaming=True, stream_chunk_mb=5)
```
4. **Adjust chunk size for document characteristics**:
```python
# For texts with many short sentences
sentences = sakurs.split(text, chunk_kb=64)
# For texts with long sentences
sentences = sakurs.split(text, chunk_kb=512)
```
## Benchmarks
Sakurs demonstrates significant performance improvements over existing Python sentence segmentation libraries. Benchmarks are run automatically in CI and results are displayed in GitHub Actions job summaries.
### Running Benchmarks Locally
To run performance benchmarks comparing sakurs with other libraries:
```bash
# Install benchmark dependencies
uv pip install -e ".[benchmark]"
# Run all benchmarks
pytest benchmarks/ --benchmark-only
# Run specific language benchmarks
pytest benchmarks/test_benchmark_english.py --benchmark-only
pytest benchmarks/test_benchmark_japanese.py --benchmark-only
```
### Benchmark Libraries
- **English**: Compared against [PySBD](https://github.com/nipunsadvilkar/pySBD)
- **Japanese**: Compared against [ja_sentence_segmenter](https://github.com/wwwcojp/ja_sentence_segmenter)
## Error Handling
```python
import sakurs
# Language errors
try:
processor = sakurs.load("unsupported_language")
except sakurs.InvalidLanguageError as e:
print(f"Language error: {e}")
# File errors
try:
sentences = sakurs.split("nonexistent.txt")
except sakurs.FileNotFoundError as e:
print(f"File error: {e}")
# Configuration errors
try:
config = sakurs.LanguageConfig.from_toml("invalid.toml")
except sakurs.ConfigurationError as e:
print(f"Config error: {e}")
# The library handles edge cases gracefully
sentences = sakurs.split("") # Returns []
sentences = sakurs.split("No punctuation") # Returns ["No punctuation"]
```
## Development
This package is built with PyO3 and maturin.
### Building from Source
For development, we recommend building and installing wheels rather than using editable installs:
```bash
# Build the wheel
maturin build --release --features extension-module
# Install the wheel (force reinstall to ensure updates)
uv pip install --force-reinstall target/wheels/*.whl
```
**Important Note**: Avoid using `pip install -e .` or `maturin develop` as they can lead to stale binaries that don't reflect Rust code changes. The editable install mechanism doesn't properly track changes in the compiled Rust extension module.
### Development Workflow
1. Make changes to the Rust code
2. Build the wheel: `maturin build --release --features extension-module`
3. Install the wheel: `uv pip install --force-reinstall target/wheels/*.whl`
4. Run tests: `python -m pytest tests/`
For convenience, you can use the Makefile from the project root:
```bash
make py-dev # Builds and installs the wheel
make py-test # Builds, installs, and runs tests
```
### Troubleshooting
If your changes aren't reflected after rebuilding:
- Check if you have an editable install: `uv pip show sakurs` (look for "Editable project location")
- Uninstall completely: `uv pip uninstall sakurs -y`
- Reinstall from wheel as shown above
- Use `.venv/bin/python` directly instead of `uv run` to avoid automatic editable install restoration
## License
MIT License - see LICENSE file for details.
Raw data
{
"_id": null,
"home_page": "https://github.com/sog4be/sakurs",
"name": "sakurs",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": "nlp, sentence, boundary, detection, tokenization, parallel",
"author": "sog4be <163720533+sog4be@users.noreply.github.com>",
"author_email": "Ryo Sogabe <163720533+sog4be@users.noreply.github.com>",
"download_url": null,
"platform": null,
"description": "# Sakurs Python Bindings\n\nHigh-performance sentence boundary detection for Python using the Delta-Stack Monoid algorithm.\n\n## Table of Contents\n\n- [Installation](#installation)\n- [Quick Start](#quick-start)\n- [API Reference](#api-reference)\n - [Functions](#functions)\n - [Classes](#classes)\n- [Supported Languages](#supported-languages)\n- [Performance Tips](#performance-tips)\n- [Benchmarks](#benchmarks)\n- [Error Handling](#error-handling)\n- [Development](#development)\n - [Building from Source](#building-from-source)\n - [Development Workflow](#development-workflow)\n - [Troubleshooting](#troubleshooting)\n- [License](#license)\n\n## Installation\n\n**Coming Soon** - Will be available on PyPI.\n\nFor now, build from source:\n```bash\ngit clone https://github.com/sog4be/sakurs.git\ncd sakurs/sakurs-py\nuv pip install -e .\n```\n\n**Requirements**: Python 3.9 or later\n\n## Quick Start\n\n```python\nimport sakurs\n\n# Simple sentence splitting\nsentences = sakurs.split(\"Hello world. This is a test.\")\nprint(sentences) # ['Hello world.', 'This is a test.']\n\n# Process files directly\nsentences = sakurs.split(\"document.txt\") # Path as string\nsentences = sakurs.split(Path(\"document.txt\")) # Path object\n\n# Specify language\nsentences = sakurs.split(\"\u3053\u308c\u306f\u65e5\u672c\u8a9e\u3067\u3059\u3002\u30c6\u30b9\u30c8\u3067\u3059\u3002\", language=\"ja\")\nprint(sentences) # ['\u3053\u308c\u306f\u65e5\u672c\u8a9e\u3067\u3059\u3002', '\u30c6\u30b9\u30c8\u3067\u3059\u3002']\n\n# Get detailed output with offsets\nresults = sakurs.split(text, return_details=True)\nfor sentence in results:\n print(f\"{sentence.text} [{sentence.start}:{sentence.end}]\")\n\n# Memory-efficient processing for large files\nfor sentence in sakurs.split_large_file(\"huge_corpus.txt\", max_memory_mb=50):\n process(sentence) # Process each sentence as it's found\n\n# Responsive iteration (loads all, yields incrementally) \nfor sentence in sakurs.iter_split(\"document.txt\"):\n print(sentence) # Get results as they're processed\n```\n\n## API Reference\n\n### Table of Contents\n- [Functions](#functions)\n - [`sakurs.split`](#sakurssplit)\n - [`sakurs.iter_split`](#sakursiter_split)\n - [`sakurs.split_large_file`](#sakurssplit_large_file)\n - [`sakurs.load`](#sakursload)\n - [`sakurs.supported_languages`](#sakurssupported_languages)\n- [Classes](#classes)\n - [`SentenceSplitter`](#sakurssentencesplitter)\n - [`Sentence`](#sakurssentence)\n - [`LanguageConfig`](#sakurslanguageconfig)\n\n### Functions\n\n#### `sakurs.split`\nSplit text or file into sentences.\n\n**Signature:**\n```python\nsakurs.split(\n input,\n *,\n language=None,\n language_config=None,\n threads=None,\n chunk_kb=None,\n parallel=False,\n execution_mode=\"adaptive\",\n return_details=False,\n encoding=\"utf-8\"\n)\n```\n\n**Parameters:**\n- `input` (str | Path | TextIO | BinaryIO): Text string, file path, or file-like object\n- `language` (str, optional): Language code (\"en\", \"ja\")\n- `language_config` (LanguageConfig, optional): Custom language configuration\n- `threads` (int, optional): Number of threads (None for auto)\n- `chunk_kb` (int, optional): Chunk size in KB (default: 256) for parallel processing (default: 256)\n- `parallel` (bool): Force parallel processing even for small inputs\n- `execution_mode` (str): \"sequential\", \"parallel\", or \"adaptive\" (default)\n- `return_details` (bool): Return Sentence objects with metadata instead of strings\n- `encoding` (str): Text encoding for file inputs (default: \"utf-8\")\n\n**Returns:** List[str] or List[Sentence] if return_details=True\n\n#### `sakurs.iter_split`\nProcess input and return sentences as an iterator. Loads entire input but yields incrementally.\n\n**Signature:**\n```python\nsakurs.iter_split(\n input,\n *,\n language=None,\n language_config=None,\n threads=None,\n chunk_kb=None,\n encoding=\"utf-8\"\n)\n```\n\n**Parameters:** Same as `split()` except no `return_details` parameter\n\n**Returns:** Iterator[str] - Iterator yielding sentences\n\n#### `sakurs.split_large_file`\nProcess large files with limited memory usage.\n\n**Signature:**\n```python\nsakurs.split_large_file(\n file_path,\n *,\n language=None,\n language_config=None,\n max_memory_mb=100,\n overlap_size=1024,\n encoding=\"utf-8\"\n)\n```\n\n**Parameters:**\n- `file_path` (str | Path): Path to the file\n- `language` (str, optional): Language code\n- `language_config` (LanguageConfig, optional): Custom language configuration \n- `max_memory_mb` (int): Maximum memory to use in MB (default: 100)\n- `overlap_size` (int): Bytes to overlap between chunks (default: 1024)\n- `encoding` (str): File encoding (default: \"utf-8\")\n\n**Returns:** Iterator[str] - Iterator yielding sentences\n\n#### `sakurs.load`\nCreate a processor instance for repeated use.\n\n**Signature:**\n```python\nsakurs.load(\n language,\n *,\n threads=None,\n chunk_kb=None,\n execution_mode=\"adaptive\"\n)\n```\n\n**Parameters:**\n- `language` (str): Language code (\"en\" or \"ja\")\n- `threads` (int, optional): Number of threads\n- `chunk_kb` (int, optional): Chunk size in KB (default: 256)\n- `execution_mode` (str): Processing mode\n\n**Returns:** SentenceSplitter instance\n\n#### `sakurs.supported_languages`\nGet list of supported languages.\n\n**Signature:**\n```python\nsakurs.supported_languages()\n```\n\n**Returns:** List[str] - Supported language codes\n\n### Classes\n\n#### `sakurs.SentenceSplitter`\nMain sentence splitter class for sentence boundary detection.\n\n**Constructor Parameters:**\n- `language` (str, optional): Language code\n- `language_config` (LanguageConfig, optional): Custom language configuration\n- `threads` (int, optional): Number of threads\n- `chunk_kb` (int, optional): Chunk size in KB (default: 256)\n- `execution_mode` (str): \"sequential\", \"parallel\", or \"adaptive\"\n- `streaming` (bool): Enable streaming mode configuration\n- `stream_chunk_mb` (int): Chunk size in MB for streaming mode\n\n**Methods:**\n- `split(input, *, return_details=False, encoding=\"utf-8\")`: Split text or file into sentences\n- `iter_split(input, *, encoding=\"utf-8\")`: Return iterator over sentences\n- `__enter__()` / `__exit__()`: Context manager support\n\n#### `sakurs.Sentence`\nSentence with metadata (returned when `return_details=True`).\n\n**Attributes:**\n- `text` (str): The sentence text\n- `start` (int): Character offset of sentence start\n- `end` (int): Character offset of sentence end\n- `confidence` (float): Confidence score (default: 1.0)\n- `metadata` (dict): Additional metadata\n\n#### `sakurs.LanguageConfig`\nLanguage configuration for custom rules.\n\n**Class Methods:**\n- `from_toml(path)`: Load configuration from TOML file\n- `to_toml(path)`: Save configuration to TOML file\n\n**Attributes:**\n- `code` (str): Language code\n- `name` (str): Language name\n- `terminators` (TerminatorConfig): Sentence terminator rules\n- `ellipsis` (EllipsisConfig): Ellipsis handling rules\n- `abbreviations` (AbbreviationConfig): Abbreviation rules\n- `enclosures` (EnclosureConfig): Enclosure (quotes, parentheses) rules\n- `suppression` (SuppressionConfig): Pattern suppression rules\n\n## Supported Languages\n\n- English (`en`, `english`)\n- Japanese (`ja`, `japanese`)\n\n## Performance Tips\n\n1. **Choose the right function for your use case**:\n ```python\n # For small to medium texts - use split()\n sentences = sakurs.split(text)\n \n # For responsive processing - use iter_split()\n for sentence in sakurs.iter_split(document):\n process_immediately(sentence)\n \n # For huge files with memory constraints - use split_large_file()\n for sentence in sakurs.split_large_file(\"10gb_corpus.txt\", max_memory_mb=100):\n index_sentence(sentence)\n ```\n\n2. **Reuse SentenceSplitter instances**: Create once, use many times\n ```python\n processor = sakurs.load(\"en\", threads=4)\n for document in documents:\n sentences = processor.split(document)\n ```\n\n3. **Configure for your workload**: \n ```python\n # For CPU-bound batch processing\n processor = sakurs.load(\"en\", threads=8, execution_mode=\"parallel\")\n \n # For I/O-bound or interactive use\n processor = sakurs.load(\"en\", threads=2, execution_mode=\"adaptive\")\n \n # For memory-constrained environments\n processor = sakurs.SentenceSplitter(language=\"en\", streaming=True, stream_chunk_mb=5)\n ```\n\n4. **Adjust chunk size for document characteristics**:\n ```python\n # For texts with many short sentences\n sentences = sakurs.split(text, chunk_kb=64)\n \n # For texts with long sentences\n sentences = sakurs.split(text, chunk_kb=512)\n ```\n\n## Benchmarks\n\nSakurs demonstrates significant performance improvements over existing Python sentence segmentation libraries. Benchmarks are run automatically in CI and results are displayed in GitHub Actions job summaries.\n\n### Running Benchmarks Locally\n\nTo run performance benchmarks comparing sakurs with other libraries:\n\n```bash\n# Install benchmark dependencies\nuv pip install -e \".[benchmark]\"\n\n# Run all benchmarks\npytest benchmarks/ --benchmark-only\n\n# Run specific language benchmarks\npytest benchmarks/test_benchmark_english.py --benchmark-only\npytest benchmarks/test_benchmark_japanese.py --benchmark-only\n```\n\n### Benchmark Libraries\n\n- **English**: Compared against [PySBD](https://github.com/nipunsadvilkar/pySBD)\n- **Japanese**: Compared against [ja_sentence_segmenter](https://github.com/wwwcojp/ja_sentence_segmenter)\n\n## Error Handling\n\n```python\nimport sakurs\n\n# Language errors\ntry:\n processor = sakurs.load(\"unsupported_language\")\nexcept sakurs.InvalidLanguageError as e:\n print(f\"Language error: {e}\")\n\n# File errors\ntry:\n sentences = sakurs.split(\"nonexistent.txt\")\nexcept sakurs.FileNotFoundError as e:\n print(f\"File error: {e}\")\n\n# Configuration errors\ntry:\n config = sakurs.LanguageConfig.from_toml(\"invalid.toml\")\nexcept sakurs.ConfigurationError as e:\n print(f\"Config error: {e}\")\n\n# The library handles edge cases gracefully\nsentences = sakurs.split(\"\") # Returns []\nsentences = sakurs.split(\"No punctuation\") # Returns [\"No punctuation\"]\n```\n\n## Development\n\nThis package is built with PyO3 and maturin.\n\n### Building from Source\n\nFor development, we recommend building and installing wheels rather than using editable installs:\n\n```bash\n# Build the wheel\nmaturin build --release --features extension-module\n\n# Install the wheel (force reinstall to ensure updates)\nuv pip install --force-reinstall target/wheels/*.whl\n```\n\n**Important Note**: Avoid using `pip install -e .` or `maturin develop` as they can lead to stale binaries that don't reflect Rust code changes. The editable install mechanism doesn't properly track changes in the compiled Rust extension module.\n\n### Development Workflow\n\n1. Make changes to the Rust code\n2. Build the wheel: `maturin build --release --features extension-module`\n3. Install the wheel: `uv pip install --force-reinstall target/wheels/*.whl`\n4. Run tests: `python -m pytest tests/`\n\nFor convenience, you can use the Makefile from the project root:\n```bash\nmake py-dev # Builds and installs the wheel\nmake py-test # Builds, installs, and runs tests\n```\n\n### Troubleshooting\n\nIf your changes aren't reflected after rebuilding:\n- Check if you have an editable install: `uv pip show sakurs` (look for \"Editable project location\")\n- Uninstall completely: `uv pip uninstall sakurs -y`\n- Reinstall from wheel as shown above\n- Use `.venv/bin/python` directly instead of `uv run` to avoid automatic editable install restoration\n\n## License\n\nMIT License - see LICENSE file for details.\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Fast, parallel sentence boundary detection using Delta-Stack Monoid algorithm",
"version": "0.1.1",
"project_urls": {
"Documentation": "https://github.com/sog4be/sakurs#readme",
"Homepage": "https://github.com/sog4be/sakurs",
"Issues": "https://github.com/sog4be/sakurs/issues",
"Repository": "https://github.com/sog4be/sakurs"
},
"split_keywords": [
"nlp",
" sentence",
" boundary",
" detection",
" tokenization",
" parallel"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "66bdc377a4e1fefab9c186248b3411b1d2f98605eb2f2c3d8e2f9342aef4daea",
"md5": "273a630adef5beb3647dcee14d827cad",
"sha256": "772a4c110824e1e1b0b854bfb3bf1f1483eafc32b26fdbfb85ff857e5592d5d6"
},
"downloads": -1,
"filename": "sakurs-0.1.1-cp39-abi3-macosx_10_12_x86_64.whl",
"has_sig": false,
"md5_digest": "273a630adef5beb3647dcee14d827cad",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.9",
"size": 1661999,
"upload_time": "2025-07-27T15:33:39",
"upload_time_iso_8601": "2025-07-27T15:33:39.734545Z",
"url": "https://files.pythonhosted.org/packages/66/bd/c377a4e1fefab9c186248b3411b1d2f98605eb2f2c3d8e2f9342aef4daea/sakurs-0.1.1-cp39-abi3-macosx_10_12_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "2227904d8778604700d313a323c5d0fb9c3dedf5f5b9a2e7347f623772923618",
"md5": "8c47344859b5c2aa375baebc021365fe",
"sha256": "42e05f07250e4dda4be0bc85dac9d86e1ee489a149a2384d108319759c87cc9f"
},
"downloads": -1,
"filename": "sakurs-0.1.1-cp39-abi3-macosx_11_0_arm64.whl",
"has_sig": false,
"md5_digest": "8c47344859b5c2aa375baebc021365fe",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.9",
"size": 1569320,
"upload_time": "2025-07-27T15:33:41",
"upload_time_iso_8601": "2025-07-27T15:33:41.587524Z",
"url": "https://files.pythonhosted.org/packages/22/27/904d8778604700d313a323c5d0fb9c3dedf5f5b9a2e7347f623772923618/sakurs-0.1.1-cp39-abi3-macosx_11_0_arm64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "4b220e71ea68d1c9e5680fdced717e86afb44fb04f12ee62e1e04cbafffdb2e3",
"md5": "dd1d5f0eb3e2bcb570bb381d82af3b29",
"sha256": "901d55d48fe9aadccf4ec4f7a3b12f471e115b4f44f7d0e2e4bdbc72ad5f6fb0"
},
"downloads": -1,
"filename": "sakurs-0.1.1-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "dd1d5f0eb3e2bcb570bb381d82af3b29",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.9",
"size": 1811187,
"upload_time": "2025-07-27T15:33:43",
"upload_time_iso_8601": "2025-07-27T15:33:43.310163Z",
"url": "https://files.pythonhosted.org/packages/4b/22/0e71ea68d1c9e5680fdced717e86afb44fb04f12ee62e1e04cbafffdb2e3/sakurs-0.1.1-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "8402ae6f7544be4ef64fb8c2cf07a54f72a95de3ccb5e9867fa79b3fe1d90ff2",
"md5": "c24ee49b356ca29687e412ec00739b0c",
"sha256": "d38a36077bd9b76028f48c497634e1323197c9acfd09b439b23e3ccfe4b24977"
},
"downloads": -1,
"filename": "sakurs-0.1.1-cp39-abi3-win_amd64.whl",
"has_sig": false,
"md5_digest": "c24ee49b356ca29687e412ec00739b0c",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.9",
"size": 1431002,
"upload_time": "2025-07-27T15:33:45",
"upload_time_iso_8601": "2025-07-27T15:33:45.656175Z",
"url": "https://files.pythonhosted.org/packages/84/02/ae6f7544be4ef64fb8c2cf07a54f72a95de3ccb5e9867fa79b3fe1d90ff2/sakurs-0.1.1-cp39-abi3-win_amd64.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-27 15:33:39",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "sog4be",
"github_project": "sakurs",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "sakurs"
}
JSON
