Skip to main content

fast-tavern(Python 版)

Repo: Lianues/fast-tavern

这是主项目 fast-tavern 的 Python 移植版,目标是对齐 TypeScript 实现的行为(提示词组装与多阶段调试输出)。

安装(开发期)

在本目录下执行:

pip install -e .[dev]
pytest

打包/发布

在本目录下执行:

pip install build twine
python -m build
twine upload dist/*

快速开始(与 TS 用法对齐)

from fast_tavern import build_prompt, History

result = build_prompt(
    preset=preset,
    character=character,
    globals={"worldBooks": world_books, "regexScripts": regex_scripts},
    history=History.openai(
        [
            {"role": "system", "content": "You are a helpful assistant."},
            {"role": "user", "content": "Hello"},
            {"role": "assistant", "content": "Hi!"},
        ]
    ),
    view="model",
    macros={"user": "Bob"},
    variables={"score": 1},
    output_format="openai",
    system_role_policy="keep",
)

print(result["stages"]["tagged"]["afterPostRegex"])

### 旧酒馆格式直接调用(包装入口)

```python
from fast_tavern import build_prompt_from_silly_tavern

result = build_prompt_from_silly_tavern(
    preset=legacy_preset_settings,   # 旧 ST preset/settings
    character=legacy_character,      # 可选,旧 ST character
    globals={
        "worldBooks": legacy_worldbooks,
        "regexScripts": legacy_regexes,
    },
    history=legacy_chat,
    view="model",
    outputFormat="openai",
)

也可单独调用转换函数:

  • convert_preset_from_silly_tavern / convert_worldbook_from_silly_tavern
  • convert_regex_from_silly_tavern / convert_character_from_silly_tavern / convert_history_from_silly_tavern

print(result["stages"]["output"]["afterPostRegex"])


### Regex flags 说明(与 TS 的差异点)

- `findRegex` 支持 `"/pattern/flags"` 与 `"pattern"` 两种写法。
- flags 映射:`i/m/s` -> Python `re` 对应 flags;`g` 用于决定“替换一次/全部”;`u` 默认等价;`y` 不支持(若遇到会按普通正则处理)。

### 发布后安装

```bash
pip install fast-tavern

Metadata

Release files for fast-tavern 0.1.8

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for fast-tavern 0.1.8
File Size Uploaded Provenance Actions
fast_tavern-0.1.8.tar.gz 30.8 kB Without provenance Details for fast_tavern-0.1.8.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for fast-tavern 0.1.8
File Interpreter ABI Platform Actions
fast_tavern-0.1.8-py3-none-any.whl Python 3 none any Details for fast_tavern-0.1.8-py3-none-any.whl

Total release size: 69.5 kB

Release files / fast_tavern-0.1.8.tar.gz

Download URL fast_tavern-0.1.8.tar.gz
Size 30.8 kB
Tags Source
SHA-256 checksum
How to use checksums
e3ddd82f38da3436666471b08c2d83f8ef8a462a0187e2500d95f18c1de1ea70
BLAKE2b-256 checksum
How to use checksums
10eeba7e9b89775bec2fb5fb7e3986189e403cbd59a386d253d486896a01b6bb
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.3

Release files / fast_tavern-0.1.8-py3-none-any.whl

Download URL fast_tavern-0.1.8-py3-none-any.whl
Size 38.7 kB
Tags Python 3
SHA-256 checksum
How to use checksums
2558b0fe8fb5655a805889153218cbfaebd371a8354de73aacb5fdc791d3195d
BLAKE2b-256 checksum
How to use checksums
d54c2b8b5255a800a696330a1de7feb4f57acc0151d1f8f265b44880b848f70e
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.3

Security for fast-tavern 0.1.8

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page