pedestrian-topologies-extraction-geojson

A Compiler to Extract Pedestrian Topologies into GeoJSON

This Data Processing Library Java example shows how to use the HERE Data SDK to build a compiler pipeline that extracts all pedestrian topologies for a particular input catalog.

The compiler in this example is an incremental RefTreeCompiler which takes HERE Map Content input data, finds all pedestrian topologies, and produces output tiles in GeoJSON Format. The output of this compiler is the pedestriansegments layer GeoJSON HEREtile format.

Get Your Credentials

To run this example, you need two sets of credentials:

• Platform credentials: To get access to the platform data and resources, including HERE Map Content data for your pipeline input.
• Repository credentials: To download HERE Data SDK for Java & Scala libraries and Maven archetypes to your environment.

For more details on how to set up your credentials, see Identity & Access Management Developer Guide.

For more details on how to verify that your platform credentials are configured correctly, see the Verify Your Credentials tutorial.

Configure a Project

To follow this example, you will need a project. A project is a collection of platform resources (catalogs, pipelines, and schemas) with controlled access. You can create a project through the HERE platform portal / HERE platform portal in China.

Alternatively, use the OLP CLI olp project create command to create the project:

olp project create $PROJECT_ID $PROJECT_NAME

The command returns the HERE Resource Name (HRN) of your new project. Note down this HRN as you will need it later in this tutorial.

Note

You do not have to provide a --scope parameter if your app has a default scope. For details on how to set a default project scope for an app, see the Specify a default Project for Apps chapter of the Identity & Access Management Developer Guide.

For more information on how to work with projects, see the Organize your work in projects tutorial.

Create the Pedestrian Topologies Catalog

The catalog you need to create is used to store the pedestrian topologies generated by a compiler.

Use the HERE platform portal / HERE platform portal in China to create the output catalog in your project and add the following layers:

Layer ID	Layer Type	Partitioning	Zoom Level	Content Type	Content Encoding
pedestriansegments	Versioned	HEREtile	14	application/vnd.geo+json	uncompressed
state	Versioned	Generic	N.A.	application/octet-stream	uncompressed

Alternatively, you can use the OLP CLI to create the catalog and the corresponding layers.

In the commands that follow, replace the variable placeholders with the following values:

• $CATALOG_ID is your output catalog's ID.
• $CATALOG_HRN is your output catalog's HRN (returned by olp catalog create).
• $PROJECT_HRN is your project's HRN (returned by olp project create command).
• $CATALOG_RIB is the HRN of the public HERE Map Content catalog in your pipeline configuration (HERE environment or HERE environment in China).

Note

We recommend to set values to variables so that you can easily copy and execute the following commands.

1. Use the olp catalog create command to create the catalog. Make sure to note down the HRN returned by the following command for later use:

olp catalog create $CATALOG_ID $CATALOG_ID \
            --summary "Pedestrian topologies extracted from HERE Map Content" \
            --description "Pedestrian topologies extracted from HERE Map Content" \
            --scope $PROJECT_HRN

2. Use the olp catalog layer add command to add two versioned layers to your catalog:

olp catalog layer add $CATALOG_HRN pedestriansegments pedestriansegments --versioned --summary "pedestrian segments" \
            --description "pedestrian segments" --partitioning heretile:14 \
            --content-type application/vnd.geo+json --scope $PROJECT_HRN
olp catalog layer add $CATALOG_HRN state state --versioned --summary "state" \
            --description "state" --partitioning Generic \
            --content-type application/octet-stream --scope $PROJECT_HRN

3. Use the olp project resources link command to link the HERE Map Content catalog to your project:

olp project resources link $PROJECT_HRN $CATALOG_RIB

• For more details on catalog commands, see Catalog Commands.
• For more details on layer commands, see Layer Commands.
• For more details on project commands, see Project Commands.
• For instructions on how to link a resource to a project, see Project Resources Link command.

Configure the Compiler

From the SDK examples directory, open the data-processing/java/pedestrian-topologies-extraction-geojson project in your Integrated Development Environment (IDE).

The compiler/config/here/pipeline-config.conf (for the HERE platform environment) and compiler/config/here-china/pipeline-config.conf (for the HERE platform environment in China) files contain the permanent configuration of the data sources for the compiler.

Pick the file that corresponds to your platform environment. For example, the pipeline configuration for the HERE platform environment looks like:

pipeline.config {
  output-catalog {hrn = "YOUR_OUTPUT_CATALOG_HRN"}
  input-catalogs {
    rib {hrn = "hrn:here:data::olp-here:rib-2"}
  }
}

Replace YOUR_OUTPUT_CATALOG_HRN with the HRN of your pedestrian topologies catalog.

To find the HRN, in the HERE platform portal or the HERE platform portal in China, navigate to your catalog. The HRN is displayed in the Catalog info section.

Alternatively, you can use the catalog's HRN from the OLP CLI output from Create the Pedestrian Topologies Catalog.

The config/here/pipeline-job.conf and config/here-china/pipeline-job.conf files contain the compiler's run configuration.

In this file, modify version = 1 to reflect the version of the HERE Map Content catalog you want to process. To find the version of the HERE Map Content catalog, in the HERE platform portal or the HERE platform portal in China, navigate to the HERE Map Content catalog, and view the current catalog's version in the Catalog info section.

The remainder of the configuration is specified in the application.conf file that can be found in the src/main/resources directory of the compiler project. However, you do not have to modify it unless you want to change the behavior of the compiler.

Build the Compiler

To build the compiler, run mvn install in the pedestrian-topologies-extraction-geojson directory.

Run the Compiler Locally

To run the compiler locally, you will need to run the entry point to the compiler:

• com.here.platform.data.processing.example.java.pedestrian.geojson.Main

As arguments, you must provide the -Dspark.master parameter with the address of the Spark server master to connect to, and any configuration parameters you want to override. Alternatively, you can add those parameters to the application.conf file.

Additionally, you also need to specify the -Dpipeline-config.file and -Dpipeline-job.file parameters to specify the location of a configuration file that contains the catalogs as well as job-specific versions of the catalogs, to read and write to.

For local runs, a bounding box filter is provided in the compile/config/here/local-application.conf and compile/config/here-china/local-application.conf to limit the number of partitions to be processed. This speeds up the compilation process. In this example, we use a bounding box around the cities of Berlin and Beijing for the HERE platform and HERE platform in China environments respectively. You can edit the bounding box coordinates to compile a different partition of HERE Map Content. Make sure you update the layer coverage to reflect the different geographical region. In order to use this configuration file, you need to use the -Dconfig.file parameter.

Additionally, you also need to specify the -Dhere.platform.data-client.request-signer.credentials.here-account.here-token-scope=$PROJECT_HRN parameter to provide a scope.

Setup the environment variable $PATH_TO_CONFIG_FOLDER to ./config/here, for the HERE platform environment in China, use the files in the ./config/here-china directory.

Run the Compiler from the Command Line

Finally run the following command line in the pedestrian-topologies-extraction-geojson directory to run the Pedestrian Topologies Compiler.

For the HERE platform environment:

mvn exec:java \
-Dexec.mainClass=com.here.platform.data.processing.example.java.pedestrian.geojson.Main \
-Dpipeline-config.file=$PATH_TO_CONFIG_FOLDER/pipeline-config.conf \
-Dpipeline-job.file=./config/here/pipeline-job.conf \
-Dconfig.file=./config/here/local-application.conf \
-Dspark.master=local[*] \
-Dhere.platform.data-client.request-signer.credentials.here-account.here-token-scope=$PROJECT_HRN

Once the processing is finished, a new version to your pedestriansegments layer will be added.

You can now go to the Verify the Output section to inspect the results.

Run this Compiler as the HERE Platform Pipeline

Generate a Fat JAR file

Run the mvn -Pplatform package command in the pedestrian-topologies-extraction-geojson directory to generate a fat JAR file to deploy the compiler to a Pipeline.

mvn -Pplatform package

Deploy the Compiler to a Pipeline

Once the previous command is finished, your JAR is then available at the target directory, and you can upload it using the HERE pipelines UI (the HERE pipelines UI in China) or the OLP CLI.

You can use the OLP CLI to create pipeline components and activate the pipeline version with the following commands:

1. Create pipeline components:

olp pipeline create $COMPONENT_NAME_Pipeline --scope $PROJECT_HRN
olp pipeline template create $COMPONENT_NAME_Template batch-2.1.0 $PATH_TO_JAR \
                com.here.platform.data.processing.example.java.pedestrian.geojson.Main \
                --workers=4 --worker-units=3 --supervisor-units=2 --input-catalog-ids=rib \
                --scope $PROJECT_HRN
olp pipeline version create $COMPONENT_NAME_version $PIPELINE_ID $PIPELINE_TEMPLATE_ID \
                "$PATH_TO_CONFIG_FOLDER/pipeline-config.conf" \
                --scope $PROJECT_HRN

2. Activate the pipeline version:

olp pipeline version activate $PIPELINE_ID $PIPELINE_VERSION_ID \
                --input-catalogs "$PATH_TO_CONFIG_FOLDER/pipeline-job.conf" \
                --scope $PROJECT_HRN

You do not have to specify the input catalog's version, unless you want to. The latest version will be automatically used.

In the HERE platform portal / HERE platform portal in China, navigate to your pipeline to see its status.

Once the processing is finished, a new version to your pedestriansegments layer will be added.

You can now go to the Verify the Output section to inspect the results.

Verify the Output

In the HERE platform portal / HERE platform portal in China, select the Data tab and find your catalog.

1. Open the pedestriansegments layer and select the Inspect tab.
2. On the map, navigate to the location of your bounding box and set the zoom to level 10.
3. Finally, select any highlighted partition to view the results displayed on the map.

Results should be drawn on the map.