Transitdata-tripupdate-processor

This project is part of the Transitdata Pulsar-pipeline.

Description

This application creates GTFS-RT Trip Update messages by combining individual stop estimates. Stop cancellation messages are also applied to create cancellations in GTFS-RT format. This application reads multiple Pulsar topics and publishes messages to one topic.

Building

Dependencies

This project depends on transitdata-common project.

Locally

• mvn compile
• mvn package

Docker image

• Run this script to build the Docker image

Tests

We're separating our unit & integration tests using this pattern.

Unit tests:

• add test classes under ./src/test with suffix *Test.java
• mvn clean test -P unit-test

Integration tests:

• add test classes under ./src/integration-test with prefix IT*.java
• mvn clean verify -P integration-test

Running

Dependencies

• Pulsar

Environment variables

• TRIPUPDATE_MAX_AGE_IN_SECS: maximum age of the trip update. If data is older than this, no trip update is created
• TRIPUPDATE_MIN_TIME_BEFORE_DEPARTURE_IN_SECS: maximum amount of time before the scheduled departure time of the trip
• TRIPUPDATE_MAX_MISSING_ESTIMATES: maximum number of missing estimates. If higher than this, no trip update is created
  • The main use case of this environment variable is to filter bad metro estimates
• TRIPUPDATE_TIMEZONE: timezone to use in the trip update
• FILTER_TRAIN_DATA: whether to filter data for trains. If true, no trip updates are created for trains
• PUBLISHER_DEBOUNCE_DELAY: debounce delay period when publishing trip updates
  • This environment variable is used to limit the amount of trip updates published because stop estimates tend to arrive in bursts