CODEWITHSUNDEEP
pythongraphgremlin
Aureon
Aureon

Reputation: 141

Load csv on Gremlin error "Mapping for code not found"

I am trying to load a csv on Gremlin console. The file just has two columns [code,desc] and almost 6K records. The code (slide 9) used to load the csv is provided as a solution in a similar question.

:install org.apache.commons commons-csv 1.5
import org.apache.commons.csv.CSVFormat
g = TinkerGraph.open().traversal()
fileReader = new FileReader('C:/airports.csv')
records = CSVFormat.RFC4180.withFirstRecordAsHeader().parse(fileReader);[]
records.each{
        code=it.get('code');
        desc=it.get('desc');
        g.V().has('code', code).fold().coalesce(
            unfold(),
            addV('airport').property('code', code).property('desc',desc)
        ).iterate()
}
g.V().count()

And the error that I got is:

Mapping for code not found, expected one of [´╗┐code, desc]

Finally, I want to know if by loading the CSV file on the Gremlin console I will be able to read it in Python, provided I set up the remote connection.

Thank you

Upvotes: 0

Views: 155

Answers (1)

Kelvin Lawrence
Kelvin Lawrence

Reputation: 14391

The error you are seeing implies the CSV file does not have a header row. Using this test file:

$ cat airports.csv 
code,desc
LHR,"London Heathrow"
LGW,"London Gatwick"
DFW,"Dallas Fort Worth"

I was able to read it using

gremlin> fileReader = new FileReader('airports.csv')
==>java.io.FileReader@19c1820d
gremlin> records = CSVFormat.RFC4180.withFirstRecordAsHeader().parse(fileReader);[]
gremlin> records.each{
......1>         c = it.get('code')
......2>         d = it.get('desc')
......3>         println(c + " : " + d)
......4> }  

LHR : London Heathrow
LGW : London Gatwick
DFW : Dallas Fort Worth

Note that I did not use desc as a variable name. Inside the Gremlin console that will collide with the Order.desc enum.

Given these building blocks...

gremlin> fileReader = new FileReader('airports.csv')
==>java.io.FileReader@10850d17
gremlin> records = CSVFormat.RFC4180.withFirstRecordAsHeader().parse(fileReader);[]
gremlin> records.each{
......1>         c=it.get('code');
......2>         d=it.get('desc');
......3>         println(it)
......4>         g.V().has('code', c).fold().coalesce(
......5>             unfold(),
......6>             addV('airport').property('code',c).property('desc',d)
......7>         ).iterate()
......8> }

CSVRecord [comment=null, mapping={code=0, desc=1}, recordNumber=1, values=[LHR, London Heathrow]]
CSVRecord [comment=null, mapping={code=0, desc=1}, recordNumber=2, values=[LGW, London Gatwick]]
CSVRecord [comment=null, mapping={code=0, desc=1}, recordNumber=3, values=[DFW, Dallas Fort Worth]]

gremlin> g.V().count()
==>3

gremlin> g.V().valueMap()
==>[code:[LHR],desc:[London Heathrow]]
==>[code:[LGW],desc:[London Gatwick]]
==>[code:[DFW],desc:[Dallas Fort Worth]]

As to your last question, to read the Graph using Python you will need to run a Gremlin Server and connect to it from the Gremlin Console. After loading the file you can then write a Python script that uses the Gremlin Python client to connect to the Gremlin Server.

You could also just do it all in Python. Either way you will need a Gremlin Server if you want to use Python to access the graph.

Upvotes: 1

Related Questions