This repository is a fork of the original OpenMetroMaps project used for educational purposes. The project was intentionally simplified by removing all web related subprojects.
Table of Contents
- Requirements
- Command line interface (CLI)
- File Format
- Desktop Tools
- Hacking
- Data Sources
- Station Data
- Research
- Other Ideas
In order to run the software from the development tree you need a Java Development Kit (JDK), Version 21 or later. The project uses Maven as a build tool, but you should use the included Maven Wrapper for building the project.
On Debian-based systems such as Ubuntu or Mint, you can install the JDK like this:
sudo apt-get install openjdk-21-jdk
Run the Maven package
task to build the CLI:
./mvnw package
This project has a main executable that can be executed like this:
./scripts/openmetromaps-cli <task>
Alternatively, add the scripts
directory to your PATH
environment
variable in order to run openmetromaps-cli
without specifying its location
each time. The following examples assume you have done that:
export PATH=$PATH:$(readlink -f scripts)
Then invoke the main executable like this:
openmetromaps-cli <task>
Where <task>
can be any of the following:
ui-selector
osm-filter
osm-extract
osm-query
osm-import
osm-inspect
map-editor
map-viewer
map-morpher
simple-map-viewer
gtfs-import
graphml-import
create-markdown-view
util
export
Each task accepts its own set of command line parameters. To run the Map Viewer you would type:
openmetromaps-cli map-viewer --input test-data/src/main/resources/berlin.xml
To run the Map Editor, type:
openmetromaps-cli map-editor --input test-data/src/main/resources/berlin.xml
The osm-import
task imports data from OpenStreetMap and offers more sub-tasks:
openmetromaps-cli osm-import <sub-task>
where <sub-task>
may be one of the following:
file
overpass
The util
task works on map model files and offers more sub-tasks:
openmetromaps-cli util <sub-task>
where <sub-task>
may be one of the following:
info
list-change-stations
list-lines-with-change-stations
purge-stations
The export
task works on map model files and offers more sub-tasks:
openmetromaps-cli export <sub-task>
where <sub-task>
may be one of the following:
svg
png
Examples:
openmetromaps-cli export png --input test-data/src/main/resources/berlin.xml
--output berlin.png --zoom 2
openmetromaps-cli export svg --input test-data/src/main/resources/berlin.xml
--output berlin.svg --zoom 3
A major goal of this project is to develop a file format for storing schematic maps for public transport networks. There's no formal specification of the file format yet and features of the format are still under construction. See an example file or the Berlin testing file to get an idea of how it's going to look. See the specification draft.
We're developing a set of desktop tools for working with the map files. Those tools are written in Java and user interfaces are based on Swing with DockingFrames for dockable dialogs.
One core component is the Map Editor that allows you to create new maps based on OpenStreetMap data or from scratch and lets you manipulate existing maps.
The Map Editor is the main interface for creating and manipulating maps. There's a separate manual that explains the features in some detail.
Have a look at the To-Do list.
To start hacking on the project, you should use an IDE. We're using IntelliJ here.
Once you've set up your working environment, you can start running the editor
from within the IDE. Navigate to the class TestMapEditor
and run this class
as a Java application.
The Map Editor provides an infrastructure for implementing algorithms for optimizing maps. When you run the editor, you can access the available optimization algorithms via the menu (Edit → Algorithms). Currently there's only two algorithms available:
- Dummy Optimization: This is a placeholder algorithm that does nothing and only
exists in order to show how to add an algorithm to the menu. Have a look at
class
DummyOptimizationAction
. - StraightenAxisParallelLines: This is a very basic optimization algorithm that
strives to detect subway lines with almost axis-parallel sections. Sections
that are classified as quasi axis-parallel will be modified so that they are
really axis-parallel afterwards. Have a look at
StraightenAxisParallelLinesAction
to see how the menu action can be set up and atStraightenAxisParallelLinesOptimization
to see the actual optimization code.
To write your own optimization algorithm, we recommend to copy and rename
the classes StraightenAxisParallelLinesAction
and
StraightenAxisParallelLinesOptimization
and start modifying the existing code.
See this list of papers for possible implementations that have been discussed in literature.
We currently support data imports from the following sources:
- OpenStreetMap (OSM)
- General Transit Feed Specification (GTFS)
Both types of import can be done using the Command Line
Interface.
See the commands osm-import
for importing OSM data and gtfs-import
for
importing GTFS data.
- If you're not familiar with the OpenStreetMap project, start by browsing through the pages listed on the Use OpenStreetMap page on the OSM Wiki.
- See the Downloading data page on the OSM Wiki on how to obtain suitable OSM data.
- Probably you want to use smaller extracts such as those available from Geofabrik downloads instead of downloading the whole planet as a file. Using the Overpass API is also a good way for obtaining relevant data sets.
- TransitFeeds collects links to official GTFS data worldwide (GitHub page)
The file format is text-based and pretty simple, so you can create a data file with a normal text editor. When you want to use existing data, you can write an import algorithm of your own.
Wikidata also stores information about stations, lines and the like.
In addition to the main map file format, we're also working on an additional file format and corresponding tools to collect data about stations and their tracks. In particular, files in this format store locations of things on a station's track as a relative position on the train (front to back / tail). This information can be used to compute efficient micro-routing within line networks, i.e. optimize on which car to board a train to reach something most quickly on the destination station such as a specific exit or a stairway to your connecting train. See an example file or the Berlin testing file to get an idea of how this file works. Also see the specification draft and a longer discussion on the requirements of that format.
We're also gathering material about transit maps in general on the Research page.
- It would be nice to be able to create morphing animations from two views
of the same network. Inspired by
this article about
a number of posts on reddit
(Berlin).
Done via the
maps-morpher
module and the CLI'smap-morpher
task. - Build something like this interactive route planner using the Javascript viewer component.
- Integrate the Android component into Transportr.
- Implement different optimization algorithms to transform geographic maps into schematic maps automatically. Some hints on how to get started with that are already available. Also, there is an implementation available which could possibly be built upon (although the license changed from MIT to GPL, which makes it impossible to integrate easily). We collect a list of papers about possible algorithms.
Possible resources: