Multi field merge etl

Hey there, I’ve been trying to create edges among vertices that have already been ingested into Orientdb. My indexes are spread across two properties so I’m looking for a way to use merge to find this index. here is my etl json example.

{
  "source": { "file": { "path": "/orientdb/datasets/tech_to_server_edges.csv" } },
  "extractor": { "csv": {
      "columnsOnFirstLine": true,
      "nullValue": "NULL"
  } },
  "transformers": [
    { "merge": [
      { "joinFieldName": "name", "lookup": "software.name" },
      { "joinFieldName": "version", "lookup": "software.version" }
      ]},
    { "vertex": {
      "class": "software", "skipDuplicates": true  }
    },
    {
      "edge": {
        "class": "contains",
        "joinFieldName": "subcategory",
        "lookup": "software_subcategory.name",
        "edgeFields": {
          "segmentId": "PADU",
          "modifiedByProcess": "PADU",
          "domain": "PADU"
        },
        "direction": "in"
      }
    }
  ],
  "loader": {
    "orientdb": {
       "dbURL": "remote:localhost:/databases/test-padu",
       "dbType": "graph",
        "serverUser": "root",
        "serverPassword": "root",
        "standardElementConstraints" : true,
        "dbAutoCreateProperties": true,
        "tx" : false,
        "wal" : false,
        "batchCommit" : 1000
    }
  }
}

Is this even possible?

Hi,
what version are you using?
Can you also post the csv file?

Thanks

Regards,
Michela

name	version	modifiedByProcess	segmentId	domain	subcategory
Snow Inventory Client	3.7	PADU	PADU	PADU	Uncategorized

This is an example row of data. The first 5 columns are already ingested as a vertex. I need to update it with it’s edges.

OrientDB version : 3.0.18

also my index is this:
software.name_version UNIQUE ["name","version"] CELL_BTREE

Hi,
is

name version modifiedByProcess segmentId domain subcategory
Snow Inventory Client 3.7 PADU PADU PADU Uncategorized

the content of tech_to_server_edges.csv?

Thanks

Yes That is the contents