optimise factors_as_string

[SymbolixAU]

N <- 2e4
data <- data.frame(
	longitude = runif(N, -180, 180),
	latitude = runif(N, -180, 180),
	tooltip = sample(LETTERS, N, replace = TRUE)
)

microbenchmark::microbenchmark(
  fact = {
    js_str <- jsonify::to_json(data, factors_as_string = FALSE )
  },
  str = {
    js_fac <- jsonify::to_json(data, factors_as_string = TRUE )
  },
  times = 5
)

# Unit: milliseconds
# expr        min         lq       mean     median         uq       max neval
# fact   27.08428   27.31386   28.13172   28.05677   28.92288   29.2808     5
#  str 4653.83055 4678.51414 4688.84328 4683.43302 4692.95960 4735.4791     5

related: SymbolixAU/mapdeck#235

[dcooley]

may have a solution

N <- 2e4
data <- data.frame(
  longitude = runif(N, -180, 180),
  latitude = runif(N, -180, 180),
  tooltip = sample(LETTERS, N, replace = TRUE)
  , stringsAsFactors = F
)

js_fac <- jsonify::to_json(data, factors_as_string = TRUE )

microbenchmark::microbenchmark(
  fact = {
    data$tooltip <- as.factor( data$tooltip )
    js_str <- jsonify::to_json(data, factors_as_string = FALSE )
  },
  str = {
    data$tooltip <- as.factor( data$tooltip )
    js_fac <- jsonify::to_json(data, factors_as_string = TRUE )
  },
  times = 5
)
# Unit: milliseconds
# expr      min       lq     mean   median       uq       max neval
# fact 72.17386 77.94805 81.27065 78.44088 82.10928  95.68116     5
#  str 55.44182 81.18584 90.39643 85.23738 88.08029 142.03680     5

[SymbolixAU]

test speed for vectors & lists too.