Creating neo4j graph database from csv file using py2neo

Question

I am currently working in a doctoral program and i am interested about Py2neo, so I am using it in order to perform some experiments using social graphs. However I got into newbie troubles. Excuse me for asking these simple questions.

I got a xml dataset containing data about publications of a jornal, I have converted it into a csv table, there are about 700 records and each record is composed by four fiels: date, title, keywords, author. So my first question is how to create a graph from this table programatically. I considered writing a python script which loops the csv table, reads for each row and columns fields and writes into nodes. +++++++++++++++++++++++++++++++++++++++++++++ Code +++++++++++++++++++++++++++++++++++++++++++

   #!/usr/bin/env python
   #
   import csv
   from py2neo import neo4j, cypher
   from py2neo import node,  rel

   # calls database service of Neo4j
   #
   graph_db = neo4j.GraphDatabaseService("http://localhost:7474/db/data/")
   #
   # Create nodes and relationships from a csv table
   # since it's a csv table, a reader must be invoked


   ifile  = open('testeout5_cp.csv', "rb")
   reader = csv.reader(ifile)

   # clear database
   graph_db.clear()

   rownum = 0
   for row in reader:
        colnum = 0
        for col in row:
            titulo, autor, rel = graph_db.create(
            {"titulo": col[1]}, {"autor": col[3]}, (1, "eh_autor_de", 0)
            )
            print(titulo,  autor)  
   rownum += 1

   ifile.close()

================ I got this output (Fragment): Python 2.7.5 (default, Aug 22 2013, 09:31:58) [GCC 4.8.1 20130603 (Red Hat 4.8.1-1)] on aires2, Standard

    (Node('http://localhost:7474/db/data/node/10392'), Node('http://localhost:7474/db/data /node/10393'))
    (Node('http://localhost:7474/db/data/node/10394'), Node('http://localhost:7474/db/data/node/10395'))
    (Node('http://localhost:7474/db/data/node/10396'), Node('http://localhost:7474/db/data/node/10397'))
    (Node('http://localhost:7474/db/data/node/10398'), Node('http://localhost:7474/db/data/node/10399'))
    (Node('http://localhost:7474/db/data/node/10400'), Node('http://localhost:7474/db/data/node/10401'))
    (Node('http://localhost:7474/db/data/node/10402'), Node('http://localhost:7474/db/data/node/10403'))
    (Node('http://localhost:7474/db/data/node/10404'), Node('http://localhost:7474/db/data/node/10405'))

========= What is wrong?

What are you expecting in the output?

satoru
– satoru

2014-03-05 04:11:54 +00:00
Commented Mar 5, 2014 at 4:11 — satoru
– satoru, Commented Mar 5, 2014 at 4:11

Rik Van Bruggen · Accepted Answer · 2013-09-15 05:46:55Z

0

I am not a py2neo expert, so can't help with that. However, have you tried using a different mechanism to create your graph? Since it is not very big, I would consider using a spreadsheet (I use that a lot) - it's dead easy.

See http://blog.neo4j.org/2013/03/importing-data-into-neo4j-spreadsheet.html for some more info.

Hope it makes sense.

Rik

answered Sep 15, 2013 at 5:46

Rik Van Bruggen

3991 silver badge3 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

hjmnzs Over a year ago

Thank you, Rik for your suggestion. I will follow your suggestion. However, yesterday night,I did converted to Cypher using Nigel Small site link, so I will try both options.

Martin Preusse · Accepted Answer · 2013-09-16 12:55:44Z

0

I think there is nothing wrong, your code looks good.

You print the nodes and get proper py2neo node instances. Try print(titulo, autor, rel) to see if your relationship is also created.

Just check with the webinterface at http://localhost:7474/webadmin/ if your data is there. Since you don't have too many nodes, you could try a simple cypher query to get all nodes and check if everything is ok.

START n=node(*) RETURN n;

answered Sep 16, 2013 at 12:55

Martin Preusse

9,36912 gold badges51 silver badges82 bronze badges

Collectives™ on Stack Overflow

Creating neo4j graph database from csv file using py2neo

2 Answers 2

1 Comment

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related