Made file creating a bit more bulletproof and added 'priming'

DESCRIPTION

@@ -1,19 +1,20 @@
 Package: decapitated
 Type: Package
 Title: Headless 'Chrome' Orchestration
-Version: 0.1.0
-Date: 2017-05-02
-Author: Bob Rudis (bob@rud.is)
+Version: 0.2.0
+Authors@R: c(
+    person("Bob", "Rudis", email = "bob@rud.is", role = c("aut", "cre"), 
+           comment = c(ORCID = "0000-0001-5670-2640")),
+    person("Chanyub", "Park", email="mrchypark@gmail.com", role=c("ctb"))
+    )
 Maintainer: Bob Rudis <bob@rud.is>
-Contributor: Chanyub Park <mrchypark@gmail.com>
 Description: The 'Chrome' browser <https://www.google.com/chrome/> has a headless mode
     which can be instrumented programmatically. Tools are provided to perform headless
-    'Chrome' instrumentation on the command-line and will eventually provide support
-    for the 'DevTools' instrumentation 'API' or the forthcoming 'phantomjs'-like higher-level
-    'API' being promised by the development team.
+    'Chrome' instrumentation, including retrieving the 
+    'javascript'-executed 'DOM' contents, 'PDF' output for, or screen shot of a 'URL'.
 URL: https://github.com/hrbrmstr/decapitated
 BugReports: https://github.com/hrbrmstr/decapitated/issues
-License: AGPL
+License: MIT + file LICENSE
 SystemRequirements: Chrome 59+ on macOS/Linux, 60+ on Windows
 Suggests:
     testthat,

LICENSE

@@ -0,0 +1,2 @@
+YEAR: 2017
+COPYRIGHT HOLDER: Bob Rudis

R/chrome-pdf.r

@@ -8,6 +8,10 @@
 #'        placed in the current working directory by incrementing trailing numbers before
 #'        the end of it.
 #' @param overwrite overwrite existing file? Default: `TRUE`
+#' @param prime if `TRUE` preliminary URL retrieval requests will be sent to "prime" the
+#'        headless Chrome cache. This seems to be necessary primarily on recent versions of macOS.
+#'        If numeric, that number of "prime" requests will be sent ahead of the capture request.
+#'        If `FALSE` no priming requests will be sent.
 #' @param chrome_bin the path to Chrome (auto-set from `HEADLESS_CHROME` environment variable)
 #' @return output fileame (invisibly)
 #' @export
@@ -18,9 +22,9 @@ chrome_dump_pdf <- function(url, path=NULL, overwrite=TRUE, chrome_bin=Sys.geten
   curwd <- getwd()
   on.exit(setwd(curwd), add = TRUE)
 
-  if (is.null(path)) path <- "."
+  path <- if (is.null(path)) "." else path[1]
 
-  path <- normalizePath(path.expand(path[1]))
+  path <- suppressWarnings(normalizePath(path.expand(path)))
 
   if (!grepl("\\.pdf$", path)) {
     fil_nam <- "output.pdf"
@@ -46,6 +50,11 @@ chrome_dump_pdf <- function(url, path=NULL, overwrite=TRUE, chrome_bin=Sys.geten
   args <- c(args, sprintf("--utility-allowed-dir=%s", .get_app_dir()))
   args <- c(args, "--print-to-pdf", url)
 
+  vers <- chrome_version(quiet=TRUE)
+
+  if (is.logical(prime) & prime) .prime_url(url, 1, chrome_bin)
+  if (is.numeric(prime) & (prime>0)) .prime_url(url, prime, chrome_bin)
+
   processx::run(
     command = chrome_bin,
     args = args,

R/chrome-shot.r

@@ -14,6 +14,10 @@
 #'        placed in the current working directory by incrementing trailing numbers before
 #'        the end of it.
 #' @param overwrite overwrite existing file? Default: `TRUE`
+#' @param prime if `TRUE` preliminary URL retrieval requests will be sent to "prime" the
+#'        headless Chrome cache. This seems to be necessary primarily on recent versions of macOS.
+#'        If numeric, that number of "prime" requests will be sent ahead of the capture request.
+#'        If `FALSE` no priming requests will be sent.
 #' @param chrome_bin the path to Chrome (auto-set from `HEADLESS_CHROME` environment variable)
 #' @return `magick`
 #' @export
@@ -25,9 +29,9 @@ chrome_shot <- function(url, width=NULL, height=NULL, path=NULL, overwrite=TRUE,
   curwd <- getwd()
   on.exit(setwd(curwd), add = TRUE)
 
-  if (is.null(path)) path <- "."
+  path <- if (is.null(path)) "." else path[1]
 
-  path <- normalizePath(path.expand(path[1]))
+  path <- suppressWarnings(normalizePath(path.expand(path)))
 
   if (!grepl("\\.pdf$", path)) {
     fil_nam <- "screenshot.png"
@@ -57,6 +61,11 @@ chrome_shot <- function(url, width=NULL, height=NULL, path=NULL, overwrite=TRUE,
     args <-  c(args, sprintf("--window-size=%s,%s", height, width))
   }
 
+  vers <- chrome_version(quiet=TRUE)
+
+  if (is.logical(prime) & prime) .prime_url(url, 1, chrome_bin)
+  if (is.numeric(prime) & (prime>0)) .prime_url(url, prime, chrome_bin)
+
   processx::run(
     command = chrome_bin,
     args = args,
@@ -65,6 +74,7 @@ chrome_shot <- function(url, width=NULL, height=NULL, path=NULL, overwrite=TRUE,
     echo = FALSE
   ) -> res
 
+  message(res$stderr)
 
   first_fil <- file.path(dir_nam, sprintf("%s.%s", fil_pre, fil_ext))
   out_fil <- first_fil

R/prime-url.r

@@ -0,0 +1,22 @@
+.prime_url <- function(url, prime_ct = 1, chrome_bin=Sys.getenv("HEADLESS_CHROME")) {
+
+  args <- c("--headless")
+  args <- c(args, "--disable-gpu")
+  args <- c(args, "--no-sandbox")
+  args <- c(args, "--allow-no-sandbox-job")
+  args <- c(args, sprintf("--user-data-dir=%s", .get_app_dir()))
+  args <- c(args, sprintf("--crash-dumps-dir=%s", .get_app_dir()))
+  args <- c(args, sprintf("--utility-allowed-dir=%s", .get_app_dir()))
+  args <- c(args, "--dump-dom", url)
+
+  for (i in 1:prime_ct) {
+    processx::run(
+      command = chrome_bin,
+      args = args,
+      error_on_status = FALSE,
+      echo_cmd = FALSE,
+      echo = FALSE
+    ) -> res
+  }
+
+}

R/read-html.r

@@ -4,11 +4,15 @@
 #' @note This only grabs the `<body>` `innerHTML` contents
 #' @param url URL to read from
 #' @param render if `TRUE` then return an `xml_document`, else the raw HTML (invisibly)
+#' @param prime if `TRUE` preliminary URL retrieval requests will be sent to "prime" the
+#'        headless Chrome cache. This seems to be necessary primarily on recent versions of macOS.
+#'        If numeric, that number of "prime" requests will be sent ahead of the capture request.
+#'        If `FALSE` no priming requests will be sent.
 #' @param chrome_bin the path to Chrome (auto-set from `HEADLESS_CHROME` environment variable)
 #' @export
 #' @examples
 #' chrome_read_html("https://www.r-project.org/")
-chrome_read_html <- function(url, render=TRUE, chrome_bin=Sys.getenv("HEADLESS_CHROME")) {
+chrome_read_html <- function(url, render=TRUE, prime=TRUE, chrome_bin=Sys.getenv("HEADLESS_CHROME")) {
 
   args <- c("--headless")
   args <- c(args, "--disable-gpu")
@@ -19,6 +23,11 @@ chrome_read_html <- function(url, render=TRUE, chrome_bin=Sys.getenv("HEADLESS_C
   args <- c(args, sprintf("--utility-allowed-dir=%s", .get_app_dir()))
   args <- c(args, "--dump-dom", url)
 
+  vers <- chrome_version(quiet=TRUE)
+
+  if (is.logical(prime) & prime) .prime_url(url, 1, chrome_bin)
+  if (is.numeric(prime) & (prime>0)) .prime_url(url, prime, chrome_bin)
+
   processx::run(
     command = chrome_bin,
     args = args,

R/version.r

@@ -1,10 +1,14 @@
 #' Get Chrome version
 #'
 #' @md
+#' @param quiet if `TRUE`, no messages are displayed
 #' @param chrome_bin the path to Chrome (auto-set from `HEADLESS_CHROME` environment variable)
+#' @return the Chrome version string (invisibly)
 #' @export
 #' @examples
 #' chrome_version()
-chrome_version <- function(chrome_bin=Sys.getenv("HEADLESS_CHROME")) {
-  system2(chrome_bin, "--version")
+chrome_version <- function(quiet = FALSE, chrome_bin=Sys.getenv("HEADLESS_CHROME")) {
+  res <- processx::run(chrome_bin, "--version")
+  if (!quiet) message(res$stdout)
+  return(invisible(trimws(res$stdout)))
 }

man/chrome_dump_pdf.Rd

@@ -18,6 +18,11 @@ the end of it.}
 \item{overwrite}{overwrite existing file? Default: \code{TRUE}}
 
 \item{chrome_bin}{the path to Chrome (auto-set from \code{HEADLESS_CHROME} environment variable)}
+
+\item{prime}{if \code{TRUE} preliminary URL retrieval requests will be sent to "prime" the
+headless Chrome cache. This seems to be necessary primarily on recent versions of macOS.
+If numeric, that number of "prime" requests will be sent ahead of the capture request.
+If \code{FALSE} no priming requests will be sent.}
 }
 \value{
 output fileame (invisibly)

man/chrome_read_html.Rd

@@ -4,7 +4,7 @@
 \alias{chrome_read_html}
 \title{Read a URL via headless Chrome and return the raw or rendered \code{<body>} \code{innerHTML} DOM elements}
 \usage{
-chrome_read_html(url, render = TRUE,
+chrome_read_html(url, render = TRUE, prime = TRUE,
   chrome_bin = Sys.getenv("HEADLESS_CHROME"))
 }
 \arguments{
@@ -12,6 +12,11 @@ chrome_read_html(url, render = TRUE,
 
 \item{render}{if \code{TRUE} then return an \code{xml_document}, else the raw HTML (invisibly)}
 
+\item{prime}{if \code{TRUE} preliminary URL retrieval requests will be sent to "prime" the
+headless Chrome cache. This seems to be necessary primarily on recent versions of macOS.
+If numeric, that number of "prime" requests will be sent ahead of the capture request.
+If \code{FALSE} no priming requests will be sent.}
+
 \item{chrome_bin}{the path to Chrome (auto-set from \code{HEADLESS_CHROME} environment variable)}
 }
 \description{

man/chrome_shot.Rd

@@ -20,6 +20,11 @@ the end of it.}
 \item{overwrite}{overwrite existing file? Default: \code{TRUE}}
 
 \item{chrome_bin}{the path to Chrome (auto-set from \code{HEADLESS_CHROME} environment variable)}
+
+\item{prime}{if \code{TRUE} preliminary URL retrieval requests will be sent to "prime" the
+headless Chrome cache. This seems to be necessary primarily on recent versions of macOS.
+If numeric, that number of "prime" requests will be sent ahead of the capture request.
+If \code{FALSE} no priming requests will be sent.}
 }
 \value{
 \code{magick}

man/chrome_version.Rd

@@ -4,11 +4,16 @@
 \alias{chrome_version}
 \title{Get Chrome version}
 \usage{
-chrome_version(chrome_bin = Sys.getenv("HEADLESS_CHROME"))
+chrome_version(quiet = FALSE, chrome_bin = Sys.getenv("HEADLESS_CHROME"))
 }
 \arguments{
+\item{quiet}{if \code{TRUE}, no messages are displayed}
+
 \item{chrome_bin}{the path to Chrome (auto-set from \code{HEADLESS_CHROME} environment variable)}
 }
+\value{
+the Chrome version string (invisibly)
+}
 \description{
 Get Chrome version
 }