Apply Function Row-Wise With Null Protection

Applies a function to each row of a data frame, with automatic handling of NULL or empty inputs. This is a safer alternative to base::by that returns NULL instead of throwing errors when given invalid or empty data frames.

Usage

byNotEmpty(df, func)

Arguments

df: Data frame to iterate over. If NULL, not a data frame, or has zero rows, the function returns NULL without attempting iteration.
func: Function to apply to each row. The function receives one argument: a single-row data frame representing the current row. Should return any value.

Value

A by object containing the results from applying func to each row, or NULL if the input data frame is empty, NULL, or invalid. The by object can be converted to a list using as.list() or further processed.

Details

This function wraps base::by() with null-safety checks, making it suitable for use in pipelines where data frames may be empty or NULL. It's commonly used throughout improveR for iterating over resource lists, child resources, and query results where the number of results may be zero.

The function is applied to each row independently, with rows identified by seq_len(nrow(df)). If you need the results automatically converted back to a data frame, use byNotEmptyAsDf instead.

Examples

if (FALSE) { # \dontrun{
# Safely iterate over resources
resources <- loadChildResources("/Data")$data[[1]]
byNotEmpty(resources, function(row) {
  cat("Processing:", row$name, "\n")
  # Returns NULL for empty result sets
})

# Extract specific fields from each row
result <- byNotEmpty(resources, function(row) {
  list(name = row$name, type = row$nodeType)
})
} # }