Dr. Test

Installation

Pull the image from DockerHub

To pull the latest Docker image from DockerHub use the following command:

docker pull drtesthelps/drtest:latest

Local Docker build

You can create the Docker image locally by using the following command:

git clone git@github.com:otto-de/dr-test.git
cd dr-test
docker build -t drtest .

Usage

Using run.sh

run.sh is meant to simplify the usage of dr-test. After building the docker image using the step described above, call the script with the schema files you want to serve data for.

./run.sh drtest:latest my-schema1.avsc my-schema2.avsc

The script will then call the docker image with the schema files mounted into the expected directory.

Using docker run

If you would like to use docker run directly, feel free to do so.

You can either call docker run and pass your schema files as mounted files:

docker run -v /path/to/my-schema1.avsc:/opt/schemafiles/my-schema1.avsc drtest:latest

Or to pass multiple schema files:

docker run -v /path/to/my-schema1.avsc:/opt/schemafiles/my-schema1.avsc -v /other/path/to/my-schema2.avsc:/opt/schemafiles/my-schema2.avsc drtest:latest

You can also mount an entire local directory. dr-test will use all files with an .avsc extension.

The webserver starts on container port 8080 by default. You can overwrite it using docker's -p flag.

Given your record's name is Foo, your test samples can be fetched via /foo or /foo/{amount}.

curl localhost:8080/foo

curl localhost:8080/foo/100

Without Docker

You can generate a test sample generator from multiple schemas as once.
All you have to do is run go run gen/gen.go gen/avro.go --target-dir=./generated $PATH_TO_SCHEMA to generate the necessary methods for the webserver. Starting the webserver go run webserver/cmd/app.go will expose one endpoint per record name.

ToDos

Randomizer Config

• Implemented

Make the randomizer configurable, such that

• Ranges
• Valid values
• Defaults
• etc can be configured to be used for each field name present in the schema.

Multiple schemas in docker container

• Implemented

Either in docker build or docker run it should be possible to load multiple avro schemas at once.

Note: see ["Without Docker"](Without Docker)) on how to use multiple schemas during generation, alleviating this problem.

Deterministic samples

• Implemented

Between two consecutive test sample fetches, data should be consistent given the schema, nor the requested quantity did not change.

Generation of edge-cases

• Implemented

Edge cases for the defined entity (nullable fields, empty lists and maps, etc.) should be obtainable by request.

Additional APIs

• Implemented

Test samples can also be consumed via kafka, sqs, amq, etc.

Multiple schema technologies

• Implemented

Add protobuf, json-schema, etc. to the list of supported schema technologies.