README.md

Learning environment

Create a docker image with Dgraph zero and alpha. Auto bulk load the data and schemas present in the import folder on start up if no data is present (no p directory)

bulk load is the fastest way to do an initial load of data in Dgraph. This docker image simplifies and speed up the time to get a Dgraph cluster up and running with data.

Build the docker learning image using

make

Start an instance

docker run --name image-name -d -p "8080:8080" -p "9080:9080" -v .:/dgraph dgraph/learning:latest

If a p directory exists, the instance will start the zero and alpha. If the directory has no data i.e no p directory, the images checks the import folder. If the import folder contains a rdf or rdf.gz file, it will be loaded using dgraph bulk Optionally place a <basename>.schema and/or <basename>.graphql file in the import folder with the basename of the rdf file to load a DQL and GraphQL schema.

For example to start an image with the donors data set from dgrap-io/benchmarks repository

mkdir import

curl -L -o donors.rdf.gz https://github.com/hypermodeinc/dgraph-benchmarks/blob/main/donors/donorsCA/donors-CA.rdf.gz

curl -LJO https://github.com/hypermodeinc/dgraph-benchmarks/blob/main/donors/donorsCA/donors.graphql

curl -LJO https://github.com/hypermodeinc/dgraph-benchmarks/blob/main/donors/donorsCA/donors.schema


docker run --name learning-donors -d -p "8080:8080" -p "9080:9080" -v .:/dgraph dgraph/learning:latest

To re-start the image with you initial data, simply stop the instance, delete the p, z, w, zw directories and restart the instance.

You can see the dgraph bulk logs in the docker image logs.