This project provides the source code for WoolNet (see demo): a visual system to help explore connections and thematic sub-graphs for Wikidata.
- Node: v16.17.1
- Java: openjdk 11
- In
/frontend
callnpm i
to install dependenciaes. - Load
/rdf-entity-path
into Eclipse IDE. - To create data to populate the graph, use graph-compress library, the data must be in Compressed Graph format.
- To create test data, see the README in the
python
folder.
- Set in
/frontend/.env
the following environment variables as correspond to your setting.
BASE_URL = "https://urldeejemplo.ejemplo" # Url of the application
WEB_SOCKET_URL = "wss://urldeejemplo.ejemplo" # Url of the WebSocket
- Run
build.ps1
. - The file that runs the application exports to
/export
. - Run the file with the following parameters:
-Xmx59g
: RAM available.-Dgraph-data=
: Pass the parameter the path to the file with graph data.
For example:
java -jar -Xmx59g -Dgraph-data="latest-truthy_small" rdf-entity-path-0.0.1-SNAPSHOT.jar
The data format used should be Adjacency Cache format (see the
python
folder).
- Set in
/frontend/.env
the following environment variables.
BASE_URL = "http://localhost:8080"
WEB_SOCKET_URL = "ws://localhost:8080"
- Run the following commands:
cd frontend
npm run dev
-
The file
frontend/public/index.html
will show the live changes. -
Run the following commands:
npm run build
- Use
RunServer.launch
to launch
The classes found in rdf-entity-path/src/main/experiments
can be used for experiments/testing and should be compiled to a .jar
.
NeighborsTest.java
: Loads a graph measuring memory and time usage. Later it measures the time needed to obtain the neighbours of a particular node.NodesNeighborsTest.launch
contains the parameters.PathFindTest.java
: Loads a graph and measures the memory and time usage for finding paths that connect two or more nodes.PathFindTest.launch
contains the parameters.