Planetiler (Streets GL Edition)

This is a fork of Michael Barry's Planetiler project, modified to generate vector tiles in a special format compatible with Streets GL.

Introduced modifications include:

• Support for level 16 zoom.
• Features in vector tiles are stored with additional precision.
• Buildings are fully embedded into tiles they intersect.
• Building outlines mostly covered by building parts are discarded.
• OMBBs and poles of inaccessibility are embedded into polygons.

How to generate tiles for Streets GL

Clone this repository with submodules:

git clone --recurse-submodules https://github.com/StrandedKitty/planetiler.git

Install Java 17+ and Maven on your system. It's known to work with OpenJDK 17 and Maven 3.8.3.

Download planet PBF and water polygons:

wget -O data/sources/planet.osm.pbf http://ftp.snt.utwente.nl/pub/misc/openstreetmap/planet-latest.osm.pbf
wget -O data/sources/water.zip https://osmdata.openstreetmap.de/download/water-polygons-split-3857.zip
unzip data/sources/water.zip -d data/sources

Build the project using Maven:

mvn -q -DskipTests --projects planetiler-dist -am clean package

Run the tile generator and save the output to planet.log:

java -Xmx110g -XX:MaxHeapFreeRatio=40 -cp planetiler-dist/target/*-with-deps.jar com.onthegomap.planetiler.examples.StreetsProfile --nodemap-type=array --storage=ram 2>&1 | tee >(cat > planet.log)

Tested on AWS EC2 r6id.4xlarge (16 vCPUs, 128GB RAM, 800GB storage). On this machine the generation process for the whole planet takes around 2.5 hours. The resulting database is around 220GB in size.

For local testing it's advised to generate tiles for small regions. You can download small portions of OSM data using Protomaps.

To run the tile generator on a small machine you can use java -cp planetiler-dist/target/*-with-deps.jar com.onthegomap.planetiler.examples.StreetsProfile.

For local testing you can use tileserver-gl-light to serve tiles:

npm install -g tileserver-gl-light
tileserver-gl-light data/data.mbtiles -p 8081

Tiles served by tileserver-gl-light are available at http://localhost:8081/data/data/{z}/{x}/{y}.pbf.

Original README.md below.

---

Planetiler (pla·nuh·tai·lr, formerly named "Flatmap") is a tool that generates Vector Tiles from geographic data sources like OpenStreetMap. Planetiler aims to be fast and memory-efficient so that you can build a map of the world in a few hours on a single machine without any external tools or database.

Vector tiles contain raw point, line, and polygon geometries that clients like MapLibre can use to render custom maps in the browser, native apps, or on a server. Planetiler packages tiles into an MBTiles (sqlite) or PMTiles file that can be served using tools like TileServer GL or Martin or even queried directly from the browser. See awesome-vector-tiles for more projects that work with data in this format.

Planetiler works by mapping input elements to vector tile features, flattening them into a big list, then sorting by tile ID to group into tiles. See ARCHITECTURE.md for more details or this blog post for more of the backstory.

Demo

See the live demo of vector tiles created by Planetiler and hosted by the OpenStreetMap Americana Project.

© OpenMapTiles © OpenStreetMap contributors

Usage

To generate a map of an area using the OpenMapTiles profile, you will need:

• Java 21+ (see CONTRIBUTING.md) or Docker
• at least 1GB of free disk space plus 5-10x the size of the .osm.pbf file
• at least 0.5x as much free RAM as the input .osm.pbf file size

To build the map:

Using Java, download planetiler.jar from the latest release and run it:

wget https://github.com/onthegomap/planetiler/releases/latest/download/planetiler.jar
java -Xmx1g -jar planetiler.jar --download --area=monaco

Or using Docker:

docker run -e JAVA_TOOL_OPTIONS="-Xmx1g" -v "$(pwd)/data":/data ghcr.io/onthegomap/planetiler:latest --download --area=monaco

⚠️ This starts off by downloading about 1GB of data sources required by the OpenMapTiles profile including ~750MB for ocean polygons and ~240MB for Natural Earth Data.

To download smaller extracts just for Monaco:

Java:

java -Xmx1g -jar planetiler.jar --download --area=monaco \
  --water-polygons-url=https://github.com/onthegomap/planetiler/raw/main/planetiler-core/src/test/resources/water-polygons-split-3857.zip \
  --natural-earth-url=https://github.com/onthegomap/planetiler/raw/main/planetiler-core/src/test/resources/natural_earth_vector.sqlite.zip

Docker:

docker run -e JAVA_TOOL_OPTIONS="-Xmx1g" -v "$(pwd)/data":/data ghcr.io/onthegomap/planetiler:latest --download --area=monaco \
  --water-polygons-url=https://github.com/onthegomap/planetiler/raw/main/planetiler-core/src/test/resources/water-polygons-split-3857.zip \
  --natural-earth-url=https://github.com/onthegomap/planetiler/raw/main/planetiler-core/src/test/resources/natural_earth_vector.sqlite.zip

You will need the full data sources to run anywhere besides Monaco.

To view tiles locally:

Using Node.js:

npm install -g tileserver-gl-light
tileserver-gl-light data/output.mbtiles

Or using Docker:

docker run --rm -it -v "$(pwd)/data":/data -p 8080:8080 maptiler/tileserver-gl -p 8080

Then open http://localhost:8080 to view tiles.

Some common arguments:

• --output tells planetiler where to write output to, and what format to write it in. For example --output=australia.pmtiles creates a pmtiles archive named australia.pmtiles.
• --download downloads input sources automatically and --only-download exits after downloading
• --area=monaco downloads a .osm.pbf extract from Geofabrik
• --osm-path=path/to/file.osm.pbf points Planetiler at an existing OSM extract on disk
• -Xmx1g controls how much RAM to give the JVM (recommended: 0.5x the input .osm.pbf file size to leave room for memory-mapped files)
• --force overwrites the output file
• --help shows all of the options and exits

Git submodules

Planetiler has a submodule dependency on planetiler-openmaptiles. Add --recurse-submodules to git clone, git pull, or git checkout commands to also update submodule dependencies.

To clone the repo with submodules:

git clone --recurse-submodules https://github.com/onthegomap/planetiler.git

If you already pulled the repo, you can initialize submodules with:

git submodule update --init

To force git to always update submodules (recommended), run this command in your local repo:

git config --local submodule.recurse true

Learn more about working with submodules here.

Generating a Map of the World

See PLANET.md.

Generating Custom Vector Tiles

If you want to customize the OpenMapTiles schema or generate an mbtiles file with OpenMapTiles + extra layers, then fork https://github.com/openmaptiles/planetiler-openmaptiles make changes there, and run directly from that repo. It is a standalone Java project with a dependency on Planetiler.

If you want to generate a separate mbtiles file with overlay layers or a full custom basemap, then:

• For simple schemas, run a recent planetiler jar or docker image with a custom schema defined in a yaml configuration file. See planetiler-custommap for details.
• For complex schemas (or if you prefer working in Java), create a new Java project that depends on Planetiler. See the planetiler-examples project for a working example.

If you want to customize how planetiler works internally, then fork this project, build from source, and consider contributing your change back for others to use!

Benchmarks

Some example runtimes for the OpenMapTiles profile (excluding downloading resources):

Input	Version	Machine	Time	output size	Logs
s3://osm-pds/2024/planet-240115.osm.pbf (69GB)	0.7.0	c3d-standard-180 (180cpu/720GB)	22m cpu:44h34m avg:120	69GB pmtiles	logs
s3://osm-pds/2024/planet-240108.osm.pbf (73GB)	0.7.0	c7gd.16xlarge (64cpu/128GB)	42m cpu:42m28s avg:52	69GB pmtiles	logs
s3://osm-pds/2022/planet-220530.osm.pbf (69GB)	0.5.0	c6gd.16xlarge (64cpu/128GB)	53m cpu:41h58m avg:47.1	79GB mbtiles	logs, VisualVM Profile
s3://osm-pds/2022/planet-220530.osm.pbf (69GB)	0.5.0	c6gd.8xlarge (32cpu/64GB)	1h27m cpu:37h55m avg:26.1	79GB mbtiles	logs
s3://osm-pds/2022/planet-220530.osm.pbf (69GB)	0.5.0	c6gd.4xlarge (16cpu/32GB)	2h38m cpu:34h3m avg:12.9	79GB mbtiles	logs

Merging nearby buildings at z13 is very expensive, when run with --building-merge-z13=false:

Input	Version	Machine	Time	output size	Logs
s3://osm-pds/2024/planet-240115.osm.pbf (69GB)	0.7.0	c3d-standard-180 (180cpu/720GB)	16m cpu:27h45m avg:104	69GB pmtiles	logs
s3://osm-pds/2024/planet-240108.osm.pbf (73GB)	0.7.0	c7gd.16xlarge (64cpu/128GB)	29m cpu:23h57 avg:50	69GB pmtiles	logs
s3://osm-pds/2024/planet-240108.osm.pbf (73GB)	0.7.0	c7gd.2xlarge (8cpu/16GB)	3h35m cpu:19h45 avg:5.5	69GB pmtiles	logs
s3://osm-pds/2024/planet-240108.osm.pbf (73GB)	0.7.0	im4gn.large (2cpu/8GB)	18h18m cpu:28h6m avg:1.5	69GB pmtiles	logs
s3://osm-pds/2022/planet-220530.osm.pbf (69GB)	0.5.0	c6gd.16xlarge (64cpu/128GB)	39m cpu:27h4m avg:42.1	79GB mbtiles	logs, VisualVM Profile

Alternatives

Some other tools that generate vector tiles from OpenStreetMap data:

• OpenMapTiles is the reference implementation of the OpenMapTiles schema that the OpenMapTiles profile is based on. It uses an intermediate postgres database and operates in two modes:
  1. Import data into database (~1 day) then serve vector tiles directly from the database. Tile serving is slower and requires bigger machines, but lets you easily incorporate realtime updates
  2. Import data into database (~1 day) then pregenerate every tile for the planet into an mbtiles file which takes over 100 days or a cluster of machines, but then tiles can be served faster on smaller machines
• Tilemaker uses a similar approach to Planetiler (no intermediate database), is more mature, and has a convenient lua API for building custom profiles without recompiling the tool, but takes about a day to generate a map of the world

Some companies that generate and host tiles for you:

• Mapbox - data from the pioneer of vector tile technologies
• Maptiler - data from the creator of OpenMapTiles schema
• Stadia Maps - what onthegomap.com uses in production

If you want to host tiles yourself but have someone else generate them for you, those companies also offer plans to download regularly-updated tilesets.

Features

• Supports Natural Earth, OpenStreetMap .osm.pbf, geopackage, and Esri Shapefiles data sources
• Writes to MBTiles or or PMTiles output.
• Java-based Profile API to customize how source elements map to vector tile features, and post-process generated tiles using JTS geometry utilities
• YAML config file format that lets you create custom schemas without writing Java code
• Merge nearby lines or polygons with the same tags before emitting vector tiles
• Automatically fixes self-intersecting polygons
• Built-in OpenMapTiles profile based on OpenMapTiles v3.13.1
• Optionally download additional name translations for elements from Wikidata
• Export real-time stats to a prometheus push gateway using --pushgateway=http://user:password@ip argument (and a grafana dashboard for viewing)
• Automatically downloads region extracts from Geofabrik using geofabrik:australia shortcut as a source URL
• Unit-test profiles to verify mapping logic, or integration-test to verify the actual contents of a generated mbtiles file (example)

Limitations

• It is harder to join and group data than when using database. Planetiler automatically groups features into tiles, so you can easily post-process nearby features in the same tile before emitting, but if you want to group or join across features in different tiles, then you must explicitly store data when processing a feature to use with later features or store features and defer processing until an input source is finished (boundary layer example)
• Planetiler only does full imports from .osm.pbf snapshots, there is no way to incorporate real-time updates.

Use as a library

Planetiler can be used as a maven-style dependency in a Java project using the settings below:

Maven

Add this repository block to your pom.xml:

<repositories>
  <repository>
    <id>osgeo</id>
    <name>OSGeo Release Repository</name>
    <url>https://repo.osgeo.org/repository/release/</url>
    <snapshots>
      <enabled>false</enabled>
    </snapshots>
    <releases>
      <enabled>true</enabled>
    </releases>
  </repository>
</repositories>

Then add the following dependency:

<dependency>
  <groupId>com.onthegomap.planetiler</groupId>
  <artifactId>planetiler-core</artifactId>
  <version>${planetiler.version}</version>
</dependency>

Gradle

Set up your repositories block::

mavenCentral()
maven {
    url "https://repo.osgeo.org/repository/release/"
}

Set up your dependencies block:

implementation 'com.onthegomap.planetiler:planetiler-core:<version>'

Contributing

Pull requests are welcome! See CONTRIBUTING.md for details.

Support

For general questions, check out the #planetiler channel on OSM-US Slack (get an invite here), or start a GitHub discussion.

Found a bug or have a feature request? Open a GitHub issue to report.

This is a side project, so support is limited. If you have the time and ability, feel free to open a pull request to fix issues or implement new features.

Acknowledgement

Planetiler is made possible by these awesome open source projects:

• OpenMapTiles for the schema and reference implementation that the openmaptiles profile is based on
• Graphhopper for basis of utilities to process OpenStreetMap data in Java
• JTS Topology Suite for working with vector geometries
• Geotools for shapefile processing
• SQLite JDBC Driver for reading Natural Earth data and writing MBTiles files
• MessagePack for compact binary encoding of intermediate map features
• geojson-vt for the basis of the stripe clipping algorithm that planetiler uses to slice geometries into tiles
• java-vector-tile for the basis of the vector tile encoder
• imposm3 for the basis of OSM multipolygon processing and tag parsing utilities
• HPPC for high-performance primitive Java collections
• Osmosis for Java utilities to parse OpenStreetMap data
• JNR-FFI for utilities to access low-level system utilities to improve memory-mapped file performance.
• cel-java for the Java implementation of Google's Common Expression Language that powers dynamic expressions embedded in schema config files.
• PMTiles optimized tile storage format
• Apache Parquet to support reading geoparquet files in java (with dependencies minimized by parquet-floor)

See NOTICE.md for a full list and license details.

Author

Planetiler was created by Michael Barry for future use generating custom basemaps or overlays for On The Go Map.

License and Attribution

Planetiler source code is licensed under the Apache 2.0 License, so it can be used and modified in commercial or other open source projects according to the license guidelines.

Maps built using planetiler do not require any special attribution, but the data or schema used might. Any maps generated from OpenStreetMap data must visibly credit OpenStreetMap contributors. Any map generated with the profile based on OpenMapTiles or a derivative must visibly credit OpenMapTiles as well.