After not using n8n for a few months, today I came back to it and updated to latest version, which immediately broke my scripts.
After debugging I came to realize that “Split in Batches” as well as the newer version of it “Loop over items” is removing certain properties on the array being split.
When i fed it the same properties but with some mock data with similar number of items, it works properly.
Can someone replicate this to possibly fix?
Running on 1.38.1, Docker for Windows, using default SQlite db
Welcome to the community!
You are in the schema view, this can be a bit odd sometimes. As it will analyse multiple items and output a schema.
So if some items do have extra fields in it this might be a bit awkward when the first item is in the output without all the fields.
Hope that makes sense.
Best way to check actual fields is the JSON view, you will see no fields are actually missing.
