Skip to content


Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

HERE is stopping its engagement on starting 03/01/2022 in favour of a fully productised and production-grade integration of the rendering engine into HERE Maps API for JavaScript ( All 3D features and many more will be offered via HERE Maps API for JavaScript moving forward.

CI codecov twitter is an experimental and work in progress open-source 3D map rendering engine written in TypeScript. site Slack channel Registration available here.


You can use this engine to:

  • Develop visually appealing 3D maps
  • Create highly animated and dynamic map visualization with WebGL, using the popular three.js library.
  • Create themeable maps, with themes that can change on the fly.
  • Create a smooth map experience with highly performant map rendering and decoding. Web workers parallelize the CPU intensive tasks, for optimal responsiveness.
  • Design your maps modularly, where you can swap out modules and data providers as required.

With that in mind, we have included some modules that let's you get started with some simple web applications that can display a map using our default style. You can get results like the one shown below:

New York City rendered with our default style

Getting started with

There are three methods to get up and running with quickly, in order of difficulty:

  1. Using the yeoman generator.
  2. linking a simple bundle as a <script> tag in your html
  3. installing a set of node modules from npm

If you want to learn more about the applications you can create, please check the Getting Started Guide.


Regardless of how you structure your project, you need some means to authenticate, please see the following guide to generate a token.

About This Repository

This repository is a monorepo containing the core components of, organized in a yarn workspace.

All components can be used stand-alone and are in the @here subdirectory.


In Node.js

All modules are installable via yarn (or npm):

yarn add @here/harp-mapview
npm install @here/harp-mapview

In Browser

Since consists of a set of modules, there are no ready-made bundles available. Take a look at the examples on information on how to use tools like webpack to create a bundle for the browser.



  • Node.js - Please see for installation instructions
  • Yarn - Please see for installation instructions.

Download dependencies


yarn install

to download and install all required packages and set up the yarn workspace.

Launch development server for examples


yarn start

To launch webpack-dev-server. Open http://localhost:8080/ in your favorite browser.

Launch development server for unit tests


yarn start-tests

Open http://localhost:8080/ in your favorite browser to run the tests.

Run unit & integration tests in Browser environment


yarn run start-tests
>: Project is running at http://localhost:8080/

Note the URL and invoke tests using mocha-webdriver-runner. Example:

npx mocha-webdriver-runner http://localhost:8081/ --chrome
npx mocha-webdriver-runner http://localhost:8081/ --headless-firefox

Run performance tests in Node.js environment

As for now, there is no baseline for performance tests results, so before examining performance one have to establish baseline:

Performance test steps

  1. Establish baseline results.
$ git checkout master
PROFILEHELPER_COMMAND=baseline yarn performance-test-node # create baseline of measurements for your particular platform

Note, that performance test suite is very limited, so it is highly possible that you have to write new dedicated performance test for code that is about to be optimized. See tests/performance for examples.

  1. Go back to your branch, change stuff and

  2. Rerun tests with your changes

yarn performance-test-node --grep lines # assuming you're playing with lines
  1. Examine output:

performance createLineGeometry segments=2
  min=0.0014ms (-2.44% vs 0.0014ms) sum=999.16ms (0% vs 999.12ms) repeats=499568.00 (-6.47% vs 534131.00) throughput=499988.43/s (-6.47% vs 534600.13/s)
  avg=0.002ms (6.92% vs 0.0019ms) med=0.0015ms (0.2% vs 0.0015ms) med95=0.0031ms (17.6% vs 0.0026ms)
  gcTime=39.6195ms (-3.39% vs 41.011ms) sumNoGc=959.54ms (0.15% vs 958.11ms) throughputNoGc=520633.00/s (-6.61% vs 557461.83/s)

Debug Tests with VSCode


   yarn test-debug

Add and start this configuration to your launch.json :

            "type": "chrome",
            "request": "attach",
            "name": "Karma Tests",
            "sourceMaps": true,
            "webRoot": "${workspaceRoot}/test",
            "address": "localhost",
            "port": 9333,
            "pathMapping": {
                "/": "${workspaceRoot}",
                "/base/": "${workspaceRoot}/"
            "sourceMapPathOverrides": {
                "webpack:///./*": "${webRoot}/*",
                "webpack:///src/*": "${webRoot}/*",
                "webpack:///*": "*",
                "webpack:///./~/*": "${webRoot}/node_modules/*",
                "meteor://app/*": "${webRoot}/*"

Generate documentation


yarn run typedoc

It will output all documentation under /dist/doc.


Copyright (C) 2017-2021 HERE Europe B.V.

See the LICENSE file in the root of this project for license details about using

In addition, please note that the fonts are under a different set of licenses.

For other use cases not listed in the license terms, please contact us.