---
title: PostGIS 矢量瓦片
date: 2020-09-11
author: ac
tags: 
 - vector tile
categories: 
 - Database
---

![马丁](./images/logo.png)

### Martin - 基于PostGIS的矢量瓦片服务器

### 1. 简介

目前流行的矢量瓦片的切图方案：

- `mapbox gl` + [tippecanoe](https://github.com/mapbox/tippecanoe) ：v2收费，tippecanoe是mapbox官方推荐的矢量瓦片静态生成工具 ，适用于大数据量场景，且不频繁更新的空间数据；
- `openlayers` + `geoserver` ：开源，使用geoserver的矢量瓦片扩展，增加矢量瓦片的输出格式；
- `maplibre`+ `Martin` +[postgis](https://www.postgis.net/docs/manual-dev/en/ST_AsMVT.html) ：开源，martin是矢量切片服务器，通过postgis的函数动态生成矢量瓦片。
- `maptiler`：收费

> mapbox v2 必须使用 access token 才能初始化 Map 对象。进行token计算，每月50000免费次数。

### 2. Martin

 `Martin`是一个开源的PostGIS矢量切片服务器，可以从任何PostGIS表或视图中创建`MVT`矢量切片，也可以从 PMTile 和MBTile文件中动态生成矢量瓦片，是使用 `Rust`编写，针对切片速度和大流量进行了优化，是极快且轻量级的切片服务器。

#### 2.1 安装

> 如果将Martin和PostgreSQL一起使用，PostGIS版本必须为v3.0+

Martin支持Linux、macOS、windows平台和docker环境，本例采用windows环境。先从github上下载[martin-Windows-x86_64](https://github.com/maplibre/martin/releases/latest/download/martin-Windows-x86_64.zip) 。解压后会发现这两个exe:

```yaml
─martin-Windows-x86_64
  ├─martin.exe
  └─mbtiles.exe
```

windows平台可以直接使用`martin.exe`来启动切片服务器，可以先在命令行查看一下参数：

```shell
D:\tools\vectorTileTool\martin-Windows-x86_64>martin.exe --help
Blazing fast and lightweight tile server with PostGIS, MBTiles, and PMTiles support

Usage: martin.exe [OPTIONS] [CONNECTION]...

Arguments:
  [CONNECTION]...  Connection strings, e.g. postgres://... or /path/to/files

Options:
  -c, --config <CONFIG>
          Path to config file. If set, no tile source-related parameters are allowed
      --save-config <SAVE_CONFIG>
          Save resulting config to a file or use "-" to print to stdout. By default, only print if sources are auto-detected
  -s, --sprite <SPRITE>
          Export a directory with SVG files as a sprite source. Can be specified multiple times
  -k, --keep-alive <KEEP_ALIVE>
          Connection keep alive timeout. [DEFAULT: 75]
  -l, --listen-addresses <LISTEN_ADDRESSES>
          The socket address to bind. [DEFAULT: 0.0.0.0:3000]
  -W, --workers <WORKERS>
          Number of web server workers
  -b, --disable-bounds
          Disable the automatic generation of bounds for spatial PG tables
      --ca-root-file <CA_ROOT_FILE>
          Loads trusted root certificates from a file. The file should contain a sequence of PEM-formatted CA certificates
  -d, --default-srid <DEFAULT_SRID>
          If a spatial PG table has SRID 0, then this default SRID will be used as a fallback
  -p, --pool-size <POOL_SIZE>
          Maximum connections pool size [DEFAULT: 20]
  -m, --max-feature-count <MAX_FEATURE_COUNT>
          Limit the number of features in a tile from a PG table source
  -h, --help
          Print help
  -V, --version
          Print version
```

#### 2.2 使用

> 本例pg安装postgis是v3.3.3.1

`martin`连接pg可以用`connection_string`（命令行参数）和配置文件两种形式。

PostgreSQL连接字符串的形式：

```shell
# martin.exe postgresql://user:password@host/db
D:\tools\vectorTileTool\martin-Windows-x86_64>martin.exe postgresql://postgres:123@127.0.0.1:5433/postgres
[2023-08-08T03:50:11Z INFO  martin] Starting Martin v0.8.7
[2023-08-08T03:50:11Z INFO  martin] Config file is not specified, auto-detecting sources
[2023-08-08T03:50:11Z INFO  martin::pg::pool] Connecting to postgresql://postgres:123@127.0.0.1:5433/postgres
[2023-08-08T03:50:11Z INFO  martin::pg::configurator] Discovered source zhujiang_river from table public.zhujiang_river with geom column (MULTIPOLYGON, SRID=4326)
[2023-08-08T03:50:11Z INFO  martin] Use --save-config to save or print Martin configuration.
[2023-08-08T03:50:11Z INFO  martin] Martin has been started on 0.0.0.0:3000.
[2023-08-08T03:50:11Z INFO  martin] Use http://0.0.0.0:3000/catalog to get the list of available sources.

```

配置文件的形式 config.yaml：

```yaml
# Connection keep alive timeout [default: 75]
keep_alive: 75

# The socket address to bind [default: 0.0.0.0:3000]
listen_addresses: '127.0.0.1:3000'

# Number of web server workers
worker_processes: 8
# Database configuration. This can also be a list of PG configs.
postgres:
  # Database connection string. You can use env vars too, for example:
  #   $DATABASE_URL
  #   ${DATABASE_URL:-postgresql://postgres@localhost/db}
  connection_string: 'postgresql://postgres@localhost:5433/postgres?sslmode=disable&user=postgres&password=123'
  
  # Same as PGSSLCERT for psql
  #ssl_cert: './postgresql.crt'
  # Same as PGSSLKEY for psql
  #ssl_key: './postgresql.key'
  # Same as PGSSLROOTCERT for psql
  #ssl_root_cert: './root.crt'

  #  If a spatial table has SRID 0, then this SRID will be used as a fallback
  default_srid: 4326

  # Maximum connections pool size [default: 20]
  pool_size: 20

  # Limit the number of table geo features included in a tile. Unlimited by default.
  max_feature_count: 1000

  # Control the automatic generation of bounds for spatial tables [default: false]
  # If enabled, it will spend some time on startup to compute geometry bounds.
  disable_bounds: false
```

```shell
D:\tools\vectorTileTool\martin-Windows-x86_64>martin.exe -c config.yaml
[2023-08-08T03:54:43Z INFO  martin] Starting Martin v0.8.7
[2023-08-08T03:54:43Z INFO  martin] Using config.yaml
[2023-08-08T03:54:43Z INFO  martin::pg::pool] Connecting to postgresql://postgres@localhost:5433/postgres?sslmode=disable&user=postgres&password=123
[2023-08-08T03:54:43Z INFO  martin::pg::configurator] Discovered source zhujiang_river from table public.zhujiang_river with geom column (MULTIPOLYGON, SRID=4326)
[2023-08-08T03:54:44Z INFO  martin] Use --save-config to save or print Martin configuration.
[2023-08-08T03:54:44Z INFO  martin] Martin has been started on 127.0.0.1:3000.
[2023-08-08T03:54:44Z INFO  martin] Use http://127.0.0.1:3000/catalog to get the list of available sources.
```

本例采用的是配置文件的形式启动服务器，服务地址是`http://127.0.0.1:3000`。

从启动的日志可以看到`Martin`将所连接的pg库中的`zhujiang_river`表发布为数据源。

> Martin 会将所有表发布为数据源（如果它们至少有一个几何列）。如果 SRID 为 0，则必须设置默认 SRID，否则该地理列/表将被忽略。所有非几何表列都将发布为矢量切片要素标签（属性）。

![image-20230808134430612](./images/image-20230808134430612.png)

打开目录可以看到已发布的数据源：

![image-20230808135027306](./images/image-20230808135027306.png)

![image-20230809094825559](./images/image-20230809094825559.png)

数据表的元数据是符合`TileJSON 3.0.0`标准的JSON文件。

`Martin`为每张要素表都提供一个`TileJSON` endpoint。

> TileJSON是表示地图元数据的开放标准，endpoint这个词以前经常被用来描述进程间通信。



#### 2.3 mapbox

发布完数据源后，使用mapbox gl 来调用一下：

```html
<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta http-equiv="X-UA-Compatible" content="IE=edge">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Document</title>
    <script src="https://unpkg.com/mapbox-gl@1.9.1/dist/mapbox-gl.js"></script>
    <link href="https://unpkg.com/mapbox-gl@1.9.1/dist/mapbox-gl.css"/>
    <style>
        #map{
            height: 100vh;
            width: 100%;
        }
    </style>
</head>
<body>
    <div id="map"></div>
    <script>
        
        const TIANDITU_URL = 'http://t0.tianditu.gov.cn';
        const tdtKey = "your_token";
        const tdtConfig = {
            code: 'tdt_img',
            name: '天地图影像',
            source: {
                type: 'raster',
                tiles: [
                    `${TIANDITU_URL}/DataServer?T=img_w&x={x}&y={y}&l={z}&tk=${tdtKey}`
                ],
                tileSize: 256,
                minzoom: 0,
                maxzoom: 22
            },
            layer: {
                id: 'tdt-img-tiles',
                type: 'raster',
                minzoom: 0,
                maxzoom: 22,
                layout: { visibility: 'visible' }
            }
        }
        let map = new mapboxgl.Map({
            container:'map',
            style:{
                "version":8,
                "sources":{},
                "layers":[],
                glyphs: "mapbox://fonts/mapbox/{fontstack}/{range}.pbf",
            },
            center: [112.39747, 22.908823], // 广东
            zoom: 8, // starting zoom 地图初始的拉伸比例
            pitch: 0, // 地图的角度，不写默认是0，取值是0-85度，一般在3D中使用
            bearing: 0, // 地图的初始方向，值是北的逆时针度数，默认是0，即是正北
            antialias: true, // 抗锯齿，通过false关闭提升性能
            minZoom: 3,
            maxZoom: 17.36
        });
        map.on('load', () => {
            // 天地图底图加载
            addLayerConfig(tdtConfig);
            addMvt();
        });
        function addLayerConfig(layerConfig){
            const {code, source,layer} = layerConfig;
            map.addSource(code, source);
            const layerParams = {
                ...layer,
                source: code
            };
            // 添加图层
            map.addLayer(layerParams);
        }
        function addMvt(){
            map.addLayer({
                id: 'lines',
                type: 'fill',
                source: {
                    type: 'vector',
                    url: 'http://localhost:3000/zhujiang_river'
                },
                'source-layer': 'zhujiang_river',
                paint: {
                	'fill-color': 'red'
                }
            });
        }
    </script>
</body>
</html>
```

加载效果：

![image-20230808141517572](./images/image-20230808141517572.png)



#### 2.4 openlayers

> **由于postgis 切片生成的是mvt格式图片,所以只能使用web墨卡托坐标系**



### 3.应用

#### 3.1 docker安装

```shell
[root@localhost martin]# pwd
/home/martin
# 下载martin镜像
[root@localhost martin]# docker pull ghcr.io/maplibre/martin:main
main: Pulling from maplibre/martin
7264a8db6415: Pull complete 
8b9d286dcf48: Pull complete 
ada398a88aa0: Pull complete 
Digest: sha256:13b38133bc15fd1d8b9b9444ca28e0380146799f1faa4c1f418090f067d48474
Status: Downloaded newer image for ghcr.io/maplibre/martin:main
ghcr.io/maplibre/martin:main


```





### 4. 遇到的问题



局限性：只能连接一个pg库。

对于有分库的系统，可以使用Nginx来转发。

#### 4.1 表没有空间索引

`Table public.table_name has no spatial index on column geom`

```sql
CREATE INDEX table_name_geom_idx
ON table_name
USING GIST (geom);
```







### 参考文章

[1] Martin https://martin.maplibre.org/

[2] Martin Tile Server Documentation https://maplibre.org/martin/installation.html