Search code examples
scalagatling

Load more records from Gatling feeder

I would like to inject n-rows from my csv file to Gatling feeder. The default approach of Gatling is to read and inject one row at a time. However, I cannot find anywhere, how to take and inject an eg. Array into a template.
I came up with creating a JSON template with Gatling Expressions as some of the fields. The issue is I have a JSON array with N-elements:

[
  {"myKey": ${value}, "mySecondKey": ${value2}, ...}, 
  {"myKey": ${value}, "mySecondKey": ${value2}, ...},
  {"myKey": ${value}, "mySecondKey": ${value2}, ...},
  {"myKey": ${value}, "mySecondKey": ${value2}, ...}
]

And my csv:

value,value2,... 
value,value2,... 
value,value2,... 
value,value2,... 
...

I would like to make it as efficient as possible. My data is in CSV file, so I would like to use csv feeder. Also, the size is large, so readRecords is not possible, since I'm getting out of memory.

Is there a way I can put N-records into the request body using Gatling?

Solution

  • From the documentation:

    feed(feeder, 2)

    Old Gatling versions:

    Attribute names, will be suffixed. For example, if the columns are name “foo” and “bar” and you’re feeding 2 records at once, you’ll get “foo1”, “bar1”, “foo2” and “bar2” session attributes.

    Modern Gatling versions:

    values will be arrays containing all the values of the same key.

    In this latter case, you can access a value at a given index with Gatling EL: #{foo(0)}, #{foo(1)}, #{bar(0)} and #{bar(1)}