In ropensci/bikedata: Download and Aggregate Data from Public Hire Bicycle Systems

Script to assemble the `bike_test_data.rda` file

devtools::load_all (".", export_all = FALSE)
library (magrittr)
data_dir <- tempdir ()
nrows <- 200 # number of rows to read from each file

names (bike_test_data)
head (bike_test_data$la)

# ----- DC -----
dl_bikedata (city = "dc", data_dir = data_dir, dates = 201701)
f <- list.files (tempdir ())
f <- file.path (data_dir, f [grep ("capitalbikeshare", f)])
fi <- unzip (f, list = TRUE)$Name
unzip (f, files = fi [1], exdir = data_dir, junkpaths = TRUE)
dc <- read.csv (file.path (data_dir, fi [1]), header = TRUE, nrows = nrows)

# ----- LO -----
dl_bikedata (city = "lo", data_dir = data_dir, dates = 201601)
f <- list.files (tempdir ())
f <- file.path (data_dir, f [grep ("JourneyDataExtract", f)])
lo <- read.csv (f [1], header = TRUE, nrows = nrows)

# ----- BO -----
# These 3 different time periods have 3 different formats
dl_bikedata (city = "bo", data_dir = data_dir, dates = 2012)
dl_bikedata (city = "bo", data_dir = data_dir, dates = 201701)
dl_bikedata (city = "bo", data_dir = data_dir, dates = 201801)
f <- list.files (tempdir ())
f <- file.path (data_dir, f [grep ("hubway", f)])
# grepping all at once doesn't put them in this order:
f <- c (f [grep ("2012", f)], f [grep ("2017", f)], f [grep ("2018", f)])
bo12 <- read.csv (f [1], header = TRUE, nrows = nrows)
fi <- unzip (f [2], list = TRUE)$Name
unzip (f [2], files = fi, exdir = data_dir, junkpaths = TRUE)
bo17 <- read.csv (file.path (data_dir, fi), header = TRUE, nrows = nrows)
fi <- unzip (f [3], list = TRUE)$Name
unzip (f [3], files = fi, exdir = data_dir, junkpaths = TRUE)
bo18 <- read.csv (file.path (data_dir, fi), header = TRUE, nrows = nrows)
# stations also need to be downloaded
dl_files <- bikedata:::get_bike_files (city = 'bo')
dl_files <- dl_files [which (grepl ('Stations', dl_files))]
for (f in dl_files)
{
    furl <- gsub (" ", "%20", f)
    f <- gsub (" ", "", f)
    destfile <- file.path (data_dir, basename(f))
    resp <- httr::GET (furl, httr::write_disk (destfile, overwrite = TRUE))
}
f <- list.files (tempdir ())
f <- file.path (data_dir, f [grep ("hubway_stations", f, ignore.case = TRUE)])
bo_st1 <- read.csv (f [1], header = TRUE)
bo_st2 <- read.csv (f [2], header = TRUE)

# ----- NY -----
dl_bikedata (city = "ny", data_dir = data_dir, dates = 201612)
f <- list.files (tempdir ())
f <- file.path (data_dir, f [grep ("^201612-citibike", f)])
fi <- unzip (f, list = TRUE)$Name
unzip (f, files = fi [1], exdir = data_dir, junkpaths = TRUE)
ny <- read.csv (file.path (data_dir, fi [1]), header = TRUE, nrows = nrows)

# ----- CH -----
dl_bikedata (city = "ch", data_dir = data_dir, dates = 201612)
f <- list.files (tempdir ())
f <- file.path (data_dir, f [grep ("Divvy", f)])
fi <- unzip (f, list = TRUE)$Name
fitr <- fi [grep ("Trips_2016_Q4", fi)]
fist <- fi [grep ("Stations_2016_Q4", fi)]
unzip (f, files = c (fitr, fist), exdir = data_dir, junkpaths = TRUE)
ch_tr <- read.csv (file.path (data_dir, fitr), header = TRUE, nrows = nrows)
ch_st <- read.csv (file.path (data_dir, fist), header = TRUE)

# ----- LA -----
dl_bikedata (city = "la", data_dir = data_dir, dates = 201701)
f <- list.files (tempdir ())
f <- file.path (data_dir, f [grep ("la_metro", f)])
fi <- unzip (f, list = TRUE)$Name
unzip (f, files = fi, exdir = data_dir, junkpaths = TRUE)
la <- read.csv (file.path (data_dir, fi), header = TRUE, nrows = nrows)

# ----- MN -----
# data have to be pre-downloaded
mn_dir <- "/data/data/bikes/mn"
f <- list.files (mn_dir, full.names = TRUE) [3] # random file for 2012
fi <- unzip (f, list = TRUE)$Name
fitr <- fi [grepl ("trip", fi)]
fist <- fi [grepl ("station", fi)]
unzip (f, files = c (fitr, fist), exdir = data_dir, junkpaths = TRUE)
mn_tr <- read.csv (file.path (data_dir, basename (fitr)),
                   header = TRUE, nrows = nrows)
mn_st <- read.csv (file.path (data_dir, basename (fist)),
                   header = TRUE)

bike_test_data <- list (dc = dc,
                        lo = lo,
                        bo12 = bo12,
                        bo17 = bo17,
                        bo18 = bo18,
                        bo_st1 = bo_st1,
                        bo_st2 = bo_st2,
                        ny = ny,
                        ch_tr = ch_tr,
                        ch_st = ch_st,
                        la = la,
                        mn_tr = mn_tr,
                        mn_st = mn_st)
save (bike_test_data, file = "./data/bike_test_data.rda", compress = "xz")