第2章：安装与环境配置

import geodatasets
world = gpd.read_file(geodatasets.get_path("naturalearth.land"))

import numpy as np
print(f"NumPy 版本: {np.__version__}")

# GeoPandas 中的坐标数据底层就是 numpy 数组
import geopandas as gpd
from shapely.geometry import Point

gdf = gpd.GeoDataFrame(geometry=[Point(0, 0), Point(1, 1)])
# bounds 返回的就是 numpy 数组
print(gdf.total_bounds)  # array([0., 0., 1., 1.])

import pandas as pd
print(f"pandas 版本: {pd.__version__}")

# GeoDataFrame 继承了 pandas 的所有功能
import geopandas as gpd
gdf = gpd.read_file(geodatasets.get_path("naturalearth.land"))

# 所有 pandas 操作都可用
print(type(gdf))          # <class 'geopandas.geodataframe.GeoDataFrame'>
print(isinstance(gdf, pd.DataFrame))  # True

import shapely
print(f"Shapely 版本: {shapely.__version__}")
print(f"GEOS 版本: {shapely.geos_version_string}")

# Shapely 提供几何对象
from shapely.geometry import Point, Polygon, LineString

point = Point(116.4, 39.9)       # 北京
polygon = Polygon([(0,0), (1,0), (1,1), (0,1)])
line = LineString([(0,0), (1,1), (2,0)])

# Shapely 2.0 的向量化操作
import numpy as np
points = shapely.points(np.random.rand(1000), np.random.rand(1000))
areas = shapely.area(shapely.buffer(points, 0.1))  # 批量缓冲区

import pyproj
print(f"pyproj 版本: {pyproj.__version__}")
print(f"PROJ 版本: {pyproj.proj_version_str}")

# CRS 对象
crs_4326 = pyproj.CRS.from_epsg(4326)   # WGS84 地理坐标系
crs_3857 = pyproj.CRS.from_epsg(3857)   # Web 墨卡托

print(f"WGS84: {crs_4326.name}")
print(f"Web Mercator: {crs_3857.name}")

# 坐标转换
transformer = pyproj.Transformer.from_crs(4326, 3857, always_xy=True)
x, y = transformer.transform(116.4, 39.9)  # 经纬度 → Web墨卡托
print(f"北京: ({x:.2f}, {y:.2f})")

import pyogrio
print(f"pyogrio 版本: {pyogrio.__version__}")
print(f"GDAL 版本: {pyogrio.__gdal_version__}")

# 列出支持的驱动
drivers = pyogrio.list_drivers()
print(f"支持的驱动数量: {len(drivers)}")

# pyogrio 提供底层读取功能
# GeoPandas 的 read_file() 内部调用 pyogrio

# matplotlib 静态绘图
import matplotlib.pyplot as plt
gdf.plot(column='population', cmap='YlOrRd', legend=True)
plt.savefig("static_map.png")

# folium 交互式地图
m = gdf.explore(column='population', cmap='YlOrRd', tooltip=['name', 'population'])
m.save("interactive_map.html")

# 使用 mapclassify 分级
gdf.plot(column='population', scheme='NaturalBreaks', k=5, legend=True)

from geopy.geocoders import Nominatim

geolocator = Nominatim(user_agent="my_app")
location = geolocator.geocode("北京市天安门")
print(f"经度: {location.longitude}, 纬度: {location.latitude}")

import geopandas as gpd
from sqlalchemy import create_engine

# 连接 PostGIS 数据库
engine = create_engine("postgresql+psycopg://user:password@localhost:5432/gisdb")

# 从 PostGIS 读取数据
gdf = gpd.read_postgis("SELECT * FROM buildings", engine, geom_col='geom')

# 写入 PostGIS
gdf.to_postgis("analysis_result", engine, if_exists='replace')

import geopandas as gpd

# 读写 Parquet 格式（高性能列式存储）
gdf = gpd.read_parquet("data.parquet")
gdf.to_parquet("output.parquet")

# 读写 Feather 格式（快速序列化）
gdf = gpd.read_feather("data.feather")
gdf.to_feather("output.feather")

# 在 Jupyter 中，GeoDataFrame 会自动以表格形式显示
import geopandas as gpd

world = gpd.read_file(geodatasets.get_path("naturalearth.land"))
world  # 自动渲染为 HTML 表格

# 交互式地图直接在 notebook 中显示
world.explore()  # 自动渲染为交互式地图

import geopandas as gpd
import pandas as pd
import numpy as np
import shapely
import pyproj
import pyogrio

print("=" * 50)
print("GeoPandas 环境信息")
print("=" * 50)
print(f"GeoPandas 版本: {gpd.__version__}")
print(f"pandas 版本:    {pd.__version__}")
print(f"numpy 版本:     {np.__version__}")
print(f"shapely 版本:   {shapely.__version__}")
print(f"pyproj 版本:    {pyproj.__version__}")
print(f"pyogrio 版本:   {pyogrio.__version__}")
print(f"GEOS 版本:      {shapely.geos_version_string}")
print(f"PROJ 版本:      {pyproj.proj_version_str}")
print(f"GDAL 版本:      {pyogrio.__gdal_version__}")

import geopandas
geopandas.show_versions()

import geopandas as gpd
from shapely.geometry import Point, Polygon
import pandas as pd

print("测试 1: 创建 GeoDataFrame...")
gdf = gpd.GeoDataFrame({
    '城市': ['北京', '上海', '广州', '深圳'],
    '人口_万': [2189, 2487, 1868, 1756],
    'geometry': [
        Point(116.40, 39.90),
        Point(121.47, 31.23),
        Point(113.26, 23.13),
        Point(114.06, 22.54)
    ]
}, crs="EPSG:4326")
print(f"  ✅ 创建成功，{len(gdf)} 条记录")

print("\n测试 2: CRS 转换...")
gdf_mercator = gdf.to_crs(epsg=3857)
print(f"  ✅ 转换成功，CRS: {gdf_mercator.crs.to_epsg()}")

print("\n测试 3: 缓冲区分析...")
gdf_proj = gdf.to_crs(epsg=32650)
buffered = gdf_proj.buffer(50000)  # 50公里缓冲区
print(f"  ✅ 缓冲区创建成功")

print("\n测试 4: 空间索引...")
sindex = gdf.sindex
print(f"  ✅ 空间索引创建成功")

print("\n测试 5: 空间关系...")
bbox = Polygon([(113, 22), (115, 22), (115, 24), (113, 24)])
result = gdf[gdf.within(bbox)]
print(f"  ✅ 空间查询成功，找到 {len(result)} 个城市在范围内")

print("\n测试 6: 数据导出...")
gdf.to_file("test_output.geojson", driver="GeoJSON")
gdf_read = gpd.read_file("test_output.geojson")
print(f"  ✅ GeoJSON 读写成功")

# 清理测试文件
import os
os.remove("test_output.geojson")

print("\n" + "=" * 50)
print("🎉 所有测试通过！GeoPandas 安装正确！")
print("=" * 50)

def check_optional_dep(name, min_version=None):
    """检查可选依赖是否安装"""
    try:
        mod = __import__(name)
        version = getattr(mod, '__version__', '未知')
        print(f"  ✅ {name:20s} {version}")
        return True
    except ImportError:
        print(f"  ❌ {name:20s} 未安装")
        return False

print("可选依赖检查:")
print("-" * 40)
check_optional_dep('matplotlib')
check_optional_dep('folium')
check_optional_dep('mapclassify')
check_optional_dep('xyzservices')
check_optional_dep('geopy')
check_optional_dep('sqlalchemy')
check_optional_dep('pyarrow')
check_optional_dep('scipy')