A record is the logical unit of a tabular data resource (ie a tabular data resource is composed of one or more records).
| Name | Definition |
|---|---|
| table | the records are rows |
| directory | the records are contained files and/or directories |
| csv | the records are lines |
Each data resource has its own logical definition of what a record is.
If the data resource is a content resource, the count attribute gives back the number of record that a data resource contains.
In a pipeline, you can change the granularity from resource to record with the split operation
We use the term record and not row because it is more generic and refer to data structures in various contexts (files, documents, etc.). A row is the record of a table.