Switch hflights to nycflights.

Closes tidyverse#562

.travis.yml

@@ -21,7 +21,7 @@ install:
 
 before_script:
 - psql -c 'create database lahman;' -U postgres
-- psql -c 'create database hflights;' -U postgres
+- psql -c 'create database nycflights13;' -U postgres
 - psql -c 'create database test;' -U postgres
 # Install dplyr package then cache sqlite and postgres version of Lahman
 - R CMD INSTALL .

DESCRIPTION

@@ -31,7 +31,7 @@ Suggests:
     ggplot2,
     mgcv,
     Lahman,
-    hflights
+    nycflights13
 VignetteBuilder: knitr
 LazyData: yes
 LinkingTo: Rcpp (>= 0.11.1),
@@ -48,9 +48,9 @@ Collate:
     'colwise.R'
     'compute-collect.r'
     'copy-to.r'
-    'data-hflights.r'
     'data-lahman.r'
     'data-nasa.r'
+    'data-nycflights13.r'
     'data-temp.r'
     'data.r'
     'dbi-s3.r'

NAMESPACE

@@ -338,8 +338,6 @@ export(grouped_dt)
 export(groups)
 export(has_cluster)
 export(has_lahman)
-export(hflights_postgres)
-export(hflights_sqlite)
 export(id)
 export(ident)
 export(init_cluster)
@@ -374,6 +372,8 @@ export(n_distinct)
 export(n_groups)
 export(nth)
 export(ntile)
+export(nycflights13_postgres)
+export(nycflights13_sqlite)
 export(order_by)
 export(partial_eval)
 export(percent_rank)

NEWS.md

@@ -1,5 +1,8 @@
 # dplyr 0.2.0.99
 
+* Now use `nycflights13` instead of `hflights` because it the variables have
+  better names and there are a few interlinked tables (#562).
+
 * `group_by()` will rename grouping variables (#410).
 
 * When `mutate()` creates a new variable that uses a window function, 

R/chain.r

@@ -26,24 +26,24 @@
 #' @export
 #' @examples
 #' # If you're performing many operations you can either do step by step
-#' if (require("hflights")) {
-#' a1 <- group_by(hflights, Year, Month, DayofMonth)
-#' a2 <- select(a1, Year:DayofMonth, ArrDelay, DepDelay)
+#' if (require("nycflights13")) {
+#' a1 <- group_by(flights, year, month, day)
+#' a2 <- select(a1, arr_delay, dep_delay)
 #' a3 <- summarise(a2,
-#'   arr = mean(ArrDelay, na.rm = TRUE),
-#'   dep = mean(DepDelay, na.rm = TRUE))
+#'   arr = mean(arr_delay, na.rm = TRUE),
+#'   dep = mean(dep_delay, na.rm = TRUE))
 #' a4 <- filter(a3, arr > 30 | dep > 30)
 #'
 #' # If you don't want to save the intermediate results, you need to
 #' # wrap the functions:
 #' filter(
 #'   summarise(
 #'     select(
-#'       group_by(hflights, Year, Month, DayofMonth),
-#'       Year:DayofMonth, ArrDelay, DepDelay
+#'       group_by(flights, year, month, day),
+#'       arr_delay, dep_delay
 #'     ),
-#'     arr = mean(ArrDelay, na.rm = TRUE),
-#'     dep = mean(DepDelay, na.rm = TRUE)
+#'     arr = mean(arr_delay, na.rm = TRUE),
+#'     dep = mean(dep_delay, na.rm = TRUE)
 #'   ),
 #'   arr > 30 | dep > 30
 #' )
@@ -53,12 +53,12 @@
 #' # Alternatively you can use chain or %>% to sequence the operations
 #' # linearly:
 #'
-#' hflights %>%
-#'   group_by(Year, Month, DayofMonth) %>%
-#'   select(Year:DayofMonth, ArrDelay, DepDelay) %>%
+#' flights %>%
+#'   group_by(year, month, day) %>%
+#'   select(arr_delay, dep_delay) %>%
 #'   summarise(
-#'     arr = mean(ArrDelay, na.rm = TRUE),
-#'     dep = mean(DepDelay, na.rm = TRUE)
+#'     arr = mean(arr_delay, na.rm = TRUE),
+#'     dep = mean(dep_delay, na.rm = TRUE)
 #'   ) %>%
 #'   filter(arr > 30 | dep > 30)
 #' }

R/data-nycflights13.r

@@ -0,0 +1,53 @@
+#' Database versions of the nycflights13 data
+#'
+#' These functions cache the data from the \code{nycflights13} database in
+#' a local database, for use in examples and vignettes. Indexes are created
+#' to making joining tables on natural keys efficient.
+#'
+#' @keywords internal
+#' @name nycflights13
+NULL
+
+#' @export
+#' @rdname nycflights13
+#' @param path location of sqlite database file
+nycflights13_sqlite <- function(path = NULL) {
+  cache_computation("nycflights_sqlite", {
+    path <- db_location(path, "nycflights13.sqlite")
+    message("Caching nycflights db at ", path)
+    src <- src_sqlite(path, create = TRUE)
+    cache_nycflights13(src)
+  })
+}
+
+#' @export
+#' @rdname nycflights13
+#' @param dbname,... Arguments passed on to \code{\link{src_postgres}}
+nycflights13_postgres <- function(dbname = "nycflights13", ...) {
+  cache_computation("nycflights_postgres", {
+    message("Caching nycflights db in postgresql db ", dbname)
+    cache_nycflights13(src_postgres(dbname, ...))
+  })
+}
+
+cache_nycflights13 <- function(src, ...) {
+  all <- data(package = "nycflights13")$results[, 3]
+  index <- list(
+    airlines = list("carrier"),
+    airports = list("faa"),
+    flights =  list(c("year", "month", "day"), "carrier", "tailnum", "origin", "dest"),
+    planes =   list("tailnum"),
+    weather =  list(c("year", "month", "day"), "origin")
+  )
+
+  tables <- setdiff(all, src_tbls(src))
+
+  # Create missing tables
+  for(table in tables) {
+    df <- getExportedValue("nycflights13", table)
+    message("Creating table: ", table)
+
+    copy_to(src, df, table, indexes = index[[table]], temporary = FALSE)
+  }
+  src
+}

R/data.r

@@ -12,6 +12,16 @@ get_cache <- function(name) {
   get(name, envir = cache)
 }
 
+cache_computation <- function(name, computation) {
+  if (is_cached(name)) {
+    get_cache(name)
+  } else {
+    res <- force(computation)
+    set_cache(name, res)
+    res
+  }
+}
+
 load_srcs <- function(f, src_names, quiet = NULL) {
   if (is.null(quiet)) {
     quiet <- !identical(Sys.getenv("NOT_CRAN"), "true")

R/do.r

@@ -58,21 +58,21 @@
 #' compare <- models %>% do(aov = anova(.$mod_linear, .$mod_quad))
 #' # compare %>% summarise(p.value = aov$`Pr(>F)`)
 #'
-#' if (require("hflights")) {
+#' if (require("nycflights13")) {
 #' # You can use it to do any arbitrary computation, like fitting a linear
 #' # model. Let's explore how carrier departure delays vary over the time
-#' carriers <- group_by(hflights, UniqueCarrier)
+#' carriers <- group_by(flights, carrier)
 #' group_size(carriers)
 #'
-#' mods <- do(carriers, mod = lm(ArrDelay ~ DepTime, data = .))
+#' mods <- do(carriers, mod = lm(arr_delay ~ dep_time, data = .))
 #' mods %>% do(as.data.frame(coef(.$mod)))
 #' mods %>% summarise(rsq = summary(mod)$r.squared)
 #'
 #' \dontrun{
 #' # This longer example shows the progress bar in action
-#' by_dest <- hflights %>% group_by(Dest) %>% filter(n() > 100)
+#' by_dest <- flights %>% group_by(dest) %>% filter(n() > 100)
 #' library(mgcv)
-#' by_dest %>% do(smooth = gam(ArrDelay ~ s(DepTime) + Month, data = .))
+#' by_dest %>% do(smooth = gam(arr_delay ~ s(dep_time) + month, data = .))
 #' }
 #' }
 do <- function(.data, ...) UseMethod("do")

R/group-size.r

@@ -3,13 +3,13 @@
 #' @param x a grouped tbl
 #' @export
 #' @examples
-#' if (require("hflights")) {
+#' if (require("nycflights13")) {
 #'
-#' by_day <- hflights %>% group_by(Year, Month, DayofMonth)
+#' by_day <- flights %>% group_by(year, month, day)
 #' n_groups(by_day)
 #' group_size(by_day)
 #'
-#' by_dest <- hflights %>% group_by(Dest)
+#' by_dest <- flights %>% group_by(dest)
 #' n_groups(by_dest)
 #' group_size(by_dest)
 #' }

R/grouped-dt.r

@@ -8,13 +8,13 @@
 #' @param vars a list of quoted variables.
 #' @export
 #' @examples
-#' if (require("data.table") && require("hflights")) {
-#' hflights_dt <- tbl_dt(hflights)
-#' group_size(group_by(hflights_dt, Year, Month, DayofMonth))
-#' group_size(group_by(hflights_dt, Dest))
+#' if (require("data.table") && require("nycflights")) {
+#' flights_dt <- tbl_dt(flights)
+#' group_size(group_by(flights_dt, year, month, day))
+#' group_size(group_by(flights_dt, dest))
 #'
-#' monthly <- group_by(hflights_dt, Month)
-#' summarise(monthly, n = n(), delay = mean(ArrDelay))
+#' monthly <- group_by(flights_dt, month)
+#' summarise(monthly, n = n(), delay = mean(arr_delay))
 #' }
 grouped_dt <- function(data, vars) {
   stopifnot(is.data.table(data))

R/manip-dt.r

@@ -7,22 +7,22 @@
 #' @param .env The environment in which to evaluate arguments not included
 #'   in the data. The default should suffice for ordinary usage.
 #' @examples
-#' if (require("data.table") && require("hflights")) {
+#' if (require("data.table") && require("nycflights13")) {
 #' # If you start with a data table, you end up with a data table
-#' hflights <- as.data.table(hflights)
-#' filter(hflights, Month == 1, DayofMonth == 1, Dest == "DFW")
-#' head(select(hflights, Year:DayOfWeek))
-#' summarise(hflights, delay = mean(ArrDelay, na.rm = TRUE), n = length(ArrDelay))
-#' head(mutate(hflights, gained = ArrDelay - DepDelay))
-#' head(arrange(hflights, Dest, desc(ArrDelay)))
+#' flights <- as.data.table(flights)
+#' filter(flights, month == 1, day == 1, dest == "DFW")
+#' head(select(flights, year:day))
+#' summarise(flights, delay = mean(arr_delay, na.rm = TRUE), n = length(arr_delay))
+#' head(mutate(flights, gained = arr_delay - dep_delay))
+#' head(arrange(flights, dest, desc(arr_delay)))
 #'
 #' # If you start with a tbl, you end up with a tbl
-#' hflights2 <- as.tbl(hflights)
-#' filter(hflights2, Month == 1, DayofMonth == 1, Dest == "DFW")
-#' head(select(hflights2, Year:DayOfWeek))
-#' summarise(hflights2, delay = mean(ArrDelay, na.rm = TRUE), n = length(ArrDelay))
-#' head(mutate(hflights2, gained = ArrDelay - DepDelay))
-#' head(arrange(hflights2, Dest, desc(ArrDelay)))
+#' flights2 <- as.tbl(flights)
+#' filter(flights2, month == 1, day == 1, dest == "DFW")
+#' head(select(flights2, year:day))
+#' summarise(flights2, delay = mean(arr_delay, na.rm = TRUE), n = length(arr_delay))
+#' head(mutate(flights2, gained = arr_delay - dep_delay))
+#' head(arrange(flights2, dest, desc(arr_delay)))
 #' }
 #' @name manip_dt
 NULL

R/manip-grouped-dt.r

@@ -5,24 +5,24 @@
 #' @param inplace if \code{FALSE} (the default) the data frame will be copied
 #'   prior to modification to avoid changes propagating via reference.
 #' @examples
-#' if (require("data.table") && require("hflights")) {
-#' hflights2 <- tbl_dt(hflights)
-#' by_dest <- group_by(hflights2, Dest)
+#' if (require("data.table") && require("nycflights13")) {
+#' flights2 <- tbl_dt(flights)
+#' by_dest <- group_by(flights2, dest)
 #'
-#' filter(by_dest, ArrDelay == max(ArrDelay, na.rm = TRUE))
-#' summarise(by_dest, arr = mean(ArrDelay, na.rm = TRUE))
+#' filter(by_dest, arr_delay == max(arr_delay, na.rm = TRUE))
+#' summarise(by_dest, arr = mean(arr_delay, na.rm = TRUE))
 #'
 #' # Normalise arrival and departure delays by airport
-#' scaled <- mutate(by_dest, arr_z = scale(ArrDelay), dep_z = scale(DepDelay))
-#' select(scaled, Year:DayOfWeek, Dest, arr_z:dep_z)
+#' scaled <- mutate(by_dest, arr_z = scale(arr_delay), dep_z = scale(dep_delay))
+#' select(scaled, year:day, dest, arr_z:dep_z)
 #'
-#' arrange(by_dest, desc(ArrDelay))
-#' select(by_dest, -(DayOfWeek:TailNum))
+#' arrange(by_dest, desc(arr_delay))
+#' select(by_dest, -(day:tailnum))
 #'
 #' # All manip functions preserve grouping structure, except for summarise
 #' # which removes a grouping level
-#' by_day <- group_by(hflights, Year, Month, DayofMonth)
-#' by_month <- summarise(by_day, delayed = sum(ArrDelay > 0, na.rm = TRUE))
+#' by_day <- group_by(flights, year, month, day)
+#' by_month <- summarise(by_day, delayed = sum(arr_delay > 0, na.rm = TRUE))
 #' by_month
 #' summarise(by_month, delayed = sum(delayed))
 #'

R/manip.r

@@ -172,11 +172,12 @@ rename <- function(.data, ...) UseMethod("rename")
 #'
 #' @export
 #' @examples
-#' data("hflights", package = "hflights")
-#' carriers <- group_by(hflights, UniqueCarrier)
+#' if (require("nycflights13")) {
+#' carriers <- group_by(flights, carrier)
 #' summarise(carriers, n())
 #' mutate(carriers, n = n())
-#' filter(carriers, n() == 79)
+#' filter(carriers, n() < 100)
+#' }
 n <- function() {
   stop("This function should not be called directly")
 }