data.py•2.61 kB
# 数据采样和处理函数
from fastmcp import FastMCP
from turf_mcp.utils import call_js_script
data_mcp = FastMCP("data")
@data_mcp.tool
async def sample(feature_collection: str, num: int) -> str:
"""
从特征集合中随机采样指定数量的特征。
此功能从输入的特征集合中随机选择指定数量的特征，返回一个新的特征集合。
Args:
feature_collection: 输入特征集合
- 类型: str (JSON 字符串格式的 GeoJSON FeatureCollection)
- 格式: 任何有效的 GeoJSON FeatureCollection
- 示例: '{"type": "FeatureCollection", "features": [{"type": "Feature", "geometry": {"type": "Point", "coordinates": [-75.343, 39.984]}}, ...]}'
num: 采样数量
- 类型: int
- 描述: 要从输入集合中随机选择特征的数量
- 示例: 5
Returns:
str: JSON 字符串格式的 GeoJSON FeatureCollection
- 类型: GeoJSON FeatureCollection
- 格式: {"type": "FeatureCollection", "features": [...]}
- 示例: '{"type": "FeatureCollection", "features": [{"type": "Feature", "geometry": {"type": "Point", "coordinates": [-75.343, 39.984]}}, ...]}'
Raises:
Exception: 当 JavaScript 执行失败、超时或输入数据格式错误时抛出异常
Example:
>>> import asyncio
>>> feature_collection = '{"type": "FeatureCollection", "features": [{"type": "Feature", "geometry": {"type": "Point", "coordinates": [-75.343, 39.984]}}, ...]}'
>>> result = asyncio.run(sample(feature_collection, 5))
>>> print(result)
'{"type": "FeatureCollection", "features": [{"type": "Feature", "geometry": {"type": "Point", "coordinates": [-75.343, 39.984]}}, ...]}'
Notes:
- 输入参数 feature_collection 必须是有效的 JSON 字符串
- 采样数量不能超过输入特征集合中的特征总数
- 如果采样数量为0，返回空的特征集合
- 采样是随机进行的，每次调用可能得到不同的结果
- 依赖于 Turf.js 库和 Node.js 环境
"""
js_script = f"""
const turf = require('@turf/turf');
const featureCollection = JSON.parse('{feature_collection}');
const num = parseInt({num});
const result = turf.sample(featureCollection, num);
console.log(JSON.stringify(result));
"""
try:
return await call_js_script(js_script)
except Exception as e:
raise Exception(f"执行异常: {str(e)}")