The Arachne Graph Database server
go get github.com/bmeg/arachne
If you have defined $GOPATH
the application will be installed at
$GOPATH
/bin/arachne otherwise it will be $HOME/go/bin/arachne
arachne server
Turn on local arachne server
Download test data
curl -O http://snap.stanford.edu/data/bigdata/amazon/amazon-meta.txt.gz
Convert the data
python $GOPATH/src/github.com/bmeg/arachne/test/amazon_convert.py amazon-meta.txt.gz test.data
Create Amazon Graph
arachne create amazon
List the Graphs
arachne list
Load data
arachne load --edge test.data.edge --vertex test.data.vertex --graph amazon
Some example queries
import aql
import json
conn = aql.Connection("http://localhost:8000")
O = conn.graph("amazon")
#Count the Vertices
print list(O.query().V().count().execute())
#Count the Edges
print list(O.query().E().count().execute())
#Try simple traveral
#print O.query().V("B00000I06U").outEdge().execute()
#Do a group count of the different 'group's in the graph
print O.query().V().groupCount("group").execute()
#use graph to find every Book that is similar to a DVD
for a in O.query().V().has("group", "Book").mark("a").outgoing("similar").has("group", "DVD").mark("b").select(["a", "b"]).execute():
print a