A platform to build knowledge graphs.
This is an open source implementation of the LinkeData.Center SDaaS™ product. See documentation in SDaaS wiki.
The SDaaS requires docker
This implementation embeds a sdaas-rdfstore based on blazegraph.
Try a short on-line demo of the platform.
This command will start a sdaas platform attached to an internal rdfstore with a micro memory foorprint
docker run --rm -ti -p 8080:8080 linkeddatacenter/sdaas-ce --reboot
browse local reasoner at http://localhost:8080/sdaas type exit
to leave the platform.
This command is the same as the previous but does not expose the workbench and uses a small memory foorprint
docker run --rm -ti -e SDAAS_SIZE=small linkeddatacenter/sdaas-ce
To run sdaas platform withouth the local rdfstore
docker run --rm -ti -e SD_NOWARMUP=1 linkeddatacenter/sdaas-ce
Use this command to start and stop by hand a local micro rdfstore:
SD_START_LOCAL_REASONING_ENGINE # you can specify the required memory footprint, default=micro
SD_STOP_LOCAL_REASONING_ENGINE
** build local image:**
docker build -t sdaas .
Smoke tests:
Manually start sdaas cli without the local reasoner
docker run --name sdmp --rm -ti -v ${PWD}:/workspace --entrypoint bash sdaas
git --version
jq --version
yq --version
gettext --version
command -v csvtool
command -v shaclvalidate.sh
scripts/sdaas
exit
Unit tests:
In order to run unit tests bats is needed (see https://github.com/bats-core/bats-core ):
bats tests/unit/
Functional tests:
To run functional and system tests you will need the local instance of blazegraph. By default, test scripts expect blazegraph endpoint at http://localhost:8080/sdaas but you can configure a different address exporting the the SD_SPARQL_ENDPOINT. The instance of blazegraph must share /workspace volume with sdaas.
For functional test execute:
/sdaas-start -d #start embedded graph engine in background
bats tests/functional
System tests:
For system test, verify that the host is able to access Internet then execute
bats tests/system/platform
scripts/sdaas
SD_SPARQL_QUERY csv "SELECT (COUNT(?s) AS ?edges) WHERE{?s?p?o}"
curl -d ESTCARD http://localhost:8080/sdaas/sparql
# in both case you should > 31K triples
SD_SPARQL_UPDATE "DROP ALL"
exit
To free the docker resources:
exit
Have a look also to the developer wiki
To push a new docker image to docker hub:
docker login
# input the docker hub credentials...
docker build -t linkeddatacenter/sdaas-ce .
docker tag linkeddatacenter/sdaas-ce linkeddatacenter/sdaas-ce:3.3.1
docker push linkeddatacenter/sdaas-ce
docker push linkeddatacenter/sdaas-ce:3.3.1
The sdaas community edition platform is derived from LinkedData.Center SDaas Product and licensed with MIT by LinkedData.Center
Copyright (C) 2018-2021 LinkedData.Center SRL
- All Rights Reserved Permission to copy and modify is granted under the MIT license