Initial package for getting and loading the data.

DESCRIPTION

@@ -0,0 +1,16 @@
+Package: geysertimes
+Title: Geyser Data from GeyserTimes.org
+Version: 0.0.0.9000
+Imports: lubridate, rappdirs, readr 
+Authors@R: 
+    person(given = "Stephen",
+           family = "Kaluzny",
+           role = c("aut", "cre"),
+           email = "[email protected]")
+Description: Gets geyser eruption and observation data from the GeyserTimes
+  database and optionally stores it locally.
+License: MIT + file LICENSE
+VignetteBuilder: knitr
+Encoding: UTF-8
+LazyData: true
+NeedsCompilation: no

NAMESPACE

@@ -0,0 +1,6 @@
+export(
+  gt_get_data,
+  gt_load_data,
+  gt_path,
+  gt_version
+)

R/gt_get_data.R

@@ -0,0 +1,33 @@
+gt_get_data <- function(dest_folder = file.path(tempdir(), "GeyserTimes"),
+  overwrite=FALSE, quiet=FALSE, version=lubridate::today()) {
+  if(dest_folder != gt_path()) {
+    if(!quiet) {
+      message("Set dest_folder to GeyserTimes::gt_path() so that data persists between R sessions.\n")
+    }
+  }
+  outpath <- file.path(dest_folder, version, "eruptions_data.rds")
+  if(file.exists(outpath) && !overwrite) {
+    warning("GeyserTimes data for this version already exists on the local machine. Use the 'overwrite' argument to re-download if neccessary.")
+    return(invisible(outpath))
+  }
+  outdir <- dirname(outpath)
+  if(!dir.exists(outdir)) {
+    dir.create(outdir, recursive=TRUE)
+  }
+  base_url <- "https://geysertimes.org/archive/complete/"
+  raw_data_file <- paste0("geysertimes_eruptions_complete_", version, ".tsv.gz")
+  download_data_file_path <- file.path(tempdir(), raw_data_file)
+  data_url <- paste0(base_url, raw_data_file)
+  oldOpt <- options(warn=-1)
+  on.exit(options(oldOpt))
+  trydownload <- try(
+    download.file(data_url, destfile=download_data_file_path, quiet=TRUE), 
+      silent=TRUE)
+  gt_tib <- readr::read_tsv(gzfile(download_data_file_path),
+    col_types=c("dcddddddddddddccccdddc"), quote="", progress=FALSE)
+  gt_tib[["eruption_time_epoch"]] <- lubridate::as_datetime(gt_tib[["eruption_time_epoch"]])
+  gt_tib[["time_updated"]] <- lubridate::as_datetime(gt_tib[["time_updated"]])
+  gt_tib[["time_entered"]] <- lubridate::as_datetime(gt_tib[["time_entered"]])
+  saveRDS(gt_tib, file=outpath)
+  invisible(outpath)
+}

R/gt_load_data.R

@@ -0,0 +1,22 @@
+"gt_load_data" <- function(path=gt_path(), quiet=FALSE, version=NULL) {
+  if(is.null(version)) {
+    version <- gt_version(path, quiet=TRUE)
+  }
+  if(is.null(version)) {
+    if(!quiet) {
+      message("Cannot find any GeyserTimes data under ", path)
+    }
+    # Look in Rtmp
+    path <- file.path(tempdir(), "GeyserTimes")
+    version <- gt_version(path, quiet=TRUE)
+    if(is.null(version)) {
+      return(NULL)
+    } else {
+      if(!quiet) {
+        message("Loading data from ", path)
+      }
+    }
+  }
+  full_path <- file.path(path, version, "eruptions_data.rds")
+  readRDS(full_path)
+}

R/gt_path.R

@@ -0,0 +1,7 @@
+"gt_path" <- function(temp=FALSE) {
+  if(temp) {
+    file.path(tempdir(), "GeyserTimes")
+  } else {
+    rappdirs::user_data_dir(appname = "GeyserTimes", appauthor = "GeyserTimes")
+  }
+}

R/gt_version.R

@@ -0,0 +1,19 @@
+"gt_version" <- function(path=gt_path(), quiet=FALSE, all=FALSE) {
+  gt_files <- list.files(path, pattern="eruptions_data\\.rds$", recursive=TRUE)
+  versions <- as.Date(dirname(gt_files), format="%Y-%m-%d")
+  # only directories of form yyyy-mm-dd are allowed:
+  versions <- sort(versions[as.character(versions) == dirname(gt_files)],
+    decreasing=TRUE)
+  if(length(gt_files) < 1 || all(is.na(versions))) {
+    if(!quiet) {
+      message("Cannot find any GeyserTimes data under", path)
+    }
+    return(NULL)
+  }
+  version <- if(all) {
+    versions[!is.na(versions)]
+  } else {
+    versions[1]
+  }
+  version
+}

man/gt_get_data.Rd

@@ -0,0 +1,59 @@
+\name{gt_get_data}
+\alias{gt_get_data}
+\title{
+Download Geyser Times Data
+}
+\description{
+Downloads the data from geysertimes.org.
+Reads the data and creates a tibble object in `dest_dir`.
+}
+\usage{
+gt_get_data(dest_dir = file.path(tempdir(), "GeyserTimes"),
+  overwrite = FALSE, quiet = FALSE, version = lubridate::today())
+}
+\arguments{
+  \item{dest_dir}{
+the location where the binary tibble object should be written.
+The default is under the current R session's temp directory
+which will disappear when the session ends.
+}
+  \item{overwrite}{
+a logical value,
+if\code{FALSE}, the data will not be downloaded again if copy of the
+data, with \code{version}, already exists in \code{dest_dir}.
+}
+  \item{quiet}{
+a logical value, if \code{TRUE}, no messages are displayed.
+}
+  \item{version}{
+a character string giving the version of the data to download.
+This should a date in the form \code{yyyy-mm-dd}.
+Typically, only the version with today's date is available.
+}
+}
+\details{
+The data is downloaded from the GeyserTimes archive web site
+\url{https://geysertimes.org/archive/} to the \code{tempdir()} directory.
+The data is then read with \code{readr::read_tsv} wtih appropriate
+column types.
+The resulting \code{tibble} object is then saved as an binary (\code{.rds})
+in \code{dest_dir}.
+}
+\value{
+a character string giving the full path to GeyserTimes data object.
+}
+\author{
+Stephen Kaluzny <[email protected]>.
+}
+\note{
+Users are encouraged to set \code{dest_dir} to \code{gt_path()} to save
+a persistent copy of the data.
+}
+\seealso{
+gt_load_data.
+}
+\examples{
+dpath0 <- gt_get_data()  # data saved under tempdir()
+dpath1 <- gt_get_data(dest_dir=gt_path())  # data saved under gt_path()
+}
+\keyword{geysertimes}

man/gt_load_data.Rd

@@ -0,0 +1,49 @@
+\name{gt_load_data}
+\alias{gt_load_data}
+\title{
+Load the Geyser Times Data
+}
+\description{
+Loads the Geyser Times data that was previously downloaded by a call
+to \code{gt_get_data}.
+}
+\usage{
+gt_load_data(path = gt_path(), quiet = FALSE, version = NULL)
+}
+\arguments{
+  \item{path}{
+a character string, the local location where the Geyser Times data
+has been written.
+The default is the local permanent location given by \code{gt_path()}.
+If no appropriate data is found at that location,
+the function will look in the temporary location given
+\code{by gt_path(temp=TRUE)}.
+}
+  \item{quiet}{
+a logical value, if \code{TRUE}, no messages are displayed.
+}
+  \item{version}{
+a character string giving the version of the Geyser Times data to load.
+Calling \code{gt_version(path, all=TRUE)} will list all versions
+available under \code{path}.
+}
+}
+\details{
+Typically, a user would download the data once, with a call to
+`gt_get_data(destdir=gt_path())`.
+Subsequent R sessions can the load this downloaded data with a
+call to `gt_load_data()`.
+}
+\value{
+a tibble containing the Geyser Times data with names:
+(need to decide on appropriate names)
+}
+\author{
+Stephen Kaluzny <[email protected]>
+}
+\seealso{
+\code{gt_get_data}.
+}
+\examples{
+}
+\keyword{geysertimes}

man/gt_path.Rd

@@ -0,0 +1,37 @@
+\name{gt_path}
+\alias{gt_path}
+\title{
+Path to GeyserTimes Local Data
+}
+\description{
+Returns the path where local GeyserTimes data is stored.
+}
+\usage{
+gt_path(temp = FALSE)
+}
+\arguments{
+  \item{temp}{
+a logical value, if \code{TRUE}, the temporary path is returned.
+This location will disappear when the R session ends.
+}
+}
+\details{
+%%  ~~ If necessary, more details than the description above ~~
+}
+\value{
+a character string giving the full path where local GeyserTimes data is stored.
+}
+\author{
+Stephen Kaluzny <[email protected]>
+}
+\note{
+%%  ~~further notes~~
+}
+\seealso{
+%% ~~objects to See Also as \code{\link{help}}, ~~~
+}
+\examples{
+# The default location for the GeyserTimes data:
+gt_path()
+}
+\keyword{geysertimes}

man/gt_version.Rd

@@ -0,0 +1,42 @@
+\name{gt_version}
+\alias{gt_version}
+\title{Version of GeyserTimes Data}
+\description{
+Returns the version of the current GeyserTimes data.
+This is a character string date in year-mm-dy format.
+}
+\usage{
+gt_version(path = gt_path(), quiet = FALSE, all = FALSE)
+}
+\arguments{
+  \item{path}{
+the path to the GeyserTimes data.
+The default is the suggested location used by `gt_get_data`.
+}
+  \item{quiet}{
+a logical value, if \code{TRUE}, no messages are printed.
+}
+  \item{all}{
+list all versions of the GeyserTimes data found,
+not just the newest.
+}
+}
+\details{
+%%  ~~ If necessary, more details than the description above ~~
+}
+\value{
+a character string listing the version(s) of GeyserTimes data
+stored under `path`.
+}
+\author{
+Stephen Kaluzny <[email protected]>
+}
+\note{
+}
+\seealso{
+\code{gt_path}
+}
+\examples{
+gt_version()
+}
+\keyword{geysertimes}

vignettes/geysertimes.Rmd

@@ -0,0 +1,85 @@
+---
+title: "Introduction to geysertimes"
+output: rmarkdown::html_vignette
+vignette: >
+  %\VignetteIndexEntry{Introduction to geysertimes}
+  %\VignetteEngine{knitr::rmarkdown}
+---
+
+```{r, echo = FALSE, message = FALSE}
+knitr::opts_chunk$set(collapse = T, comment = "#>")
+options(tibble.print_min = 4L, tibble.print_max = 4L)
+```
+
+# Basic Use
+
+Load the package
+```{r library}
+library("geysertimes")
+```
+
+## Get the Data
+
+The `gt_get_data` function downloads the compressed eruptions data
+from `https://geysertimes.org/archive/`,
+reads the data compressed data into R
+and saves version of the R object
+in the location specified
+in the `dest_folder` argument to the function.
+The default location for `dest_folder` is
+`file.path(tempdir(), "GeyserTimes"))`.
+This default location is used to meet the CRAN requirement of
+not writing files by default to any location other than under `tempdir()`.
+
+```{r default_get}
+defpath <- gt_get_data()
+defpath
+```
+
+Users are encouraged to set `dest_folder` to the value given by
+`gt_path()` which is a permanent location appropriate for the
+user on the particular platform.
+
+```{r gt_path}
+gt_path()
+```
+
+If a permanent location is used, the user only needs to get the
+data once.
+Using the suggested value for `dest_folder`:
+```{r recommend_path}
+recpath <- gt_get_data(dest_folder=gt_path())
+recpath
+```
+
+## Load the Data
+
+The `gt_load_data` is used to load the saved R object.
+
+```{r load01}
+gtdata <- gt_load_data()
+```
+
+A quick look at the data:
+```{r look}
+dim(gtdata)
+names(gtdata)
+```
+
+### Data Version
+The data that is downloaded is versioned.
+The version id is the date when the data was downloaded.
+
+The `gt_version()` lists the latest version of the data that
+has been downloaded.
+Setting `all=TRUE` will list all versions of the data that have been
+downloaded.
+
+```{r version}
+gt_version()
+```
+
+```{r version_all}
+gt_version(all=TRUE)
+```
+