# compare formats
x <- iris[1:3,]
toJSON(x)
stream_out(x)
# Trivial example
mydata <- stream_in(url("http://httpbin.org/stream/100"))
## Not run: stream large dataset to file and back
# library(nycflights13)
# stream_out(flights, file(tmp <- tempfile()))
# flights2 <- stream_in(file(tmp))
# unlink(tmp)
# all.equal(flights2, as.data.frame(flights))
#
# # stream over HTTP
# diamonds2 <- stream_in(url("http://jeroenooms.github.io/data/diamonds.json"))
#
# # stream over HTTP with gzip compression
# flights3 <- stream_in(gzcon(url("http://jeroenooms.github.io/data/nycflights13.json.gz")))
# all.equal(flights3, as.data.frame(flights))
#
# # stream over HTTPS (HTTP+SSL) via curl
# library(curl)
# flights4 <- stream_in(gzcon(curl("https://jeroenooms.github.io/data/nycflights13.json.gz")))
# all.equal(flights4, as.data.frame(flights))
#
# # or alternatively:
# flights5 <- stream_in(gzcon(pipe("curl https://jeroenooms.github.io/data/nycflights13.json.gz")))
# all.equal(flights5, as.data.frame(flights))
#
# # Full JSON IO stream from URL to file connection.
# # Calculate delays for flights over 1000 miles in batches of 5k
# library(dplyr)
# con_in <- gzcon(url("http://jeroenooms.github.io/data/nycflights13.json.gz"))
# con_out <- file(tmp <- tempfile(), open = "wb")
# stream_in(con_in, handler = function(df){
# df <- dplyr::filter(df, distance > 1000)
# df <- dplyr::mutate(df, delta = dep_delay - arr_delay)
# stream_out(df, con_out, pagesize = 1000)
# }, pagesize = 5000)
# close(con_out)
#
# # stream it back in
# mydata <- stream_in(file(tmp))
# nrow(mydata)
# unlink(tmp)
#
# # Data from http://openweathermap.org/current#bulk
# # Each row contains a nested data frame.
# daily14 <- stream_in(gzcon(url("http://78.46.48.103/sample/daily_14.json.gz")), pagesize=50)
# subset(daily14, city$name == "Berlin")$data[[1]]
#
# # Or with dplyr:
# library(dplyr)
# daily14f <- flatten(daily14)
# filter(daily14f, city.name == "Berlin")$data[[1]]
#
# # Stream import large data from zip file
# tmp <- tempfile()
# download.file("http://jsonstudio.com/wp-content/uploads/2014/02/companies.zip", tmp)
# companies <- stream_in(unz(tmp, "companies.json"))
# ## End(Not run)
Run the code above in your browser using DataLab