Merge pull request #21 from muschellij2/master

ENH: Function to Convert PPTX to PDF

.travis.yml

@@ -1,6 +1,10 @@
 language: R
-sudo: false
+sudo: required
 cache: packages
+
+before_install:
+  - if [[ "${TRAVIS_OS_NAME}" = "osx" ]]; then brew cask install libreoffice ; fi  
+  - if [[ "${TRAVIS_OS_NAME}" = "linux" ]]; then sudo apt-get update -y && sudo apt-get install libreoffice; fi
 
 after_success:
   - Rscript -e 'covr::codecov()'

NAMESPACE

@@ -3,6 +3,7 @@
 S3method(print,docx)
 export("%>%")
 export(assign_colnames)
+export(convert_to_pdf)
 export(docx_cmnt_count)
 export(docx_describe_cmnts)
 export(docx_describe_tbls)

R/convert_pptx_to_pdf.R

@@ -0,0 +1,41 @@
+#' Convert a Document (usually PowerPoint) to a PDF
+#'
+#' @md
+#' @param path path to the document, can be PowerPoint or DOCX
+#' @param pdf_file output PDF file name.  By default, creates a PDF in the
+#'        same directory as the `path` file.
+#'        This functionality requires the use of
+#'        LibreOffice and the `soffice` binary it contains.  See
+#'        [set_libreoffice_path] for more information.  Note,
+#' @export
+#' @examples
+#' \dontrun{
+#' path = system.file("examples/ex.pptx", package="docxtractr")
+#' pdf <- convert_to_pdf(path, pdf_file = tempfile(fileext = ".pdf"))
+#' path = system.file("examples/data.docx", package="docxtractr")
+#' pdf_doc <- convert_to_pdf(path, pdf_file = tempfile(fileext = ".pdf"))
+#' }
+convert_to_pdf <- function(path, pdf_file = sub("[.]pptx", ".pdf", path)) {
+  stopifnot(is_pptx(path) | is_doc(path) | is_docx(path))
+
+  lo_assert()
+  lo_path <- getOption("path_to_libreoffice")
+
+  # making temporary file because by default soffice
+  # will make sub("[.]pptx", ".pdf", path) output
+  # and don't want to do that in case pdf_file in other location
+  cp_path = tempfile(fileext = ".pptx")
+  cp_pdf = sub("[.](pptx|docx|doc)$", ".pdf", cp_path)
+  file.copy(path, cp_path)
+
+  if (Sys.info()["sysname"] == "Windows") {
+    convert_win(lo_path, dirname(cp_path), cp_path, convert_to = "pdf")
+  } else {
+    convert_osx(lo_path, dirname(cp_path), cp_path, convert_to = "pdf")
+  }
+  if (!file.exists(cp_pdf)) {
+    stop("Conversion from PPTX to PDF did not succeed")
+  }
+  file.copy(cp_pdf, pdf_file)
+  return(pdf_file)
+}

R/read_docs.r

@@ -28,8 +28,8 @@
 #'
 #' \dontrun{
 #' # from a URL
-# budget <- read_docx(
-# "http://rud.is/dl/1.DOCX")
+#' budget <- read_docx(
+#' "http://rud.is/dl/1.DOCX")
 #' }
 read_docx <- function(path, track_changes=NULL) {
 

R/utils.r

@@ -33,6 +33,8 @@ is_url <- function(path) { grepl("^(http|ftp)s?://", path) }
 
 is_docx <- function(path) { tolower(tools::file_ext(path)) == "docx" }
 
+is_pptx <- function(path) { tolower(tools::file_ext(path)) == "pptx" }
+
 is_doc <- function(path) { tolower(tools::file_ext(path)) == "doc" }
 
 # Copy a file to a new location, throw an error if the copy fails.
@@ -56,18 +58,22 @@ convert_doc_to_docx <- function(docx_dir, doc_file) {
 }
 
 # .docx to .doc convertion for Windows
-convert_win <- function(lo_path, docx_dir, doc_file) {
-  cmd <- sprintf('"%s" -convert-to docx:"MS Word 2007 XML" -headless -outdir "%s" "%s"',
+convert_win <- function(lo_path, docx_dir, doc_file,
+                        convert_to = 'docx:"MS Word 2007 XML"') {
+  cmd <- sprintf('"%s" --convert-to %s -headless -outdir "%s" "%s"',
                  lo_path,
+                 convert_to,
                  docx_dir,
                  doc_file)
   system(cmd, show.output.on.console = FALSE)
 }
 
 # .docx to .doc convertion for OSX
-convert_osx <- function(lo_path, docx_dir, doc_file) {
-  cmd <- sprintf('"%s" --convert-to docx:"MS Word 2007 XML" --headless --outdir "%s" "%s"',
+convert_osx <- function(lo_path, docx_dir, doc_file,
+                        convert_to = 'docx:"MS Word 2007 XML"') {
+  cmd <- sprintf('"%s" --convert-to %s --headless --outdir "%s" "%s"',
                  lo_path,
+                 convert_to,
                  docx_dir,
                  doc_file)
   res <- system(cmd, intern = TRUE)

appveyor.yml

@@ -16,6 +16,7 @@ cache:
 # Adapt as necessary starting from here
 
 build_script:
+  - choco install libreoffice-fresh & exit 0
   - travis-tool.sh install_deps
 
 test_script:

tests/testthat/test-doc-conversion.R

@@ -0,0 +1,11 @@
+context("DOC conversion works")
+test_that("we can convert a DOC to DOCX if LibreOffice Installed", {
+  lp = try({
+    docxtractr:::lo_find()
+  }, silent = TRUE)
+  if (!inherits(lp, "try-error")) {
+    path <- system.file("examples/preserve.doc", package = "docxtractr")
+    doc = read_docx(path)
+    expect_that(doc, is_a("docx"))
+  }
+})

tests/testthat/test-pptx-conversion.R

@@ -0,0 +1,23 @@
+context("PPTX conversion works")
+test_that("we can convert a PPTX if LibreOffice Installed", {
+  lp = try({
+    docxtractr:::lo_find()
+  }, silent = TRUE)
+  if (!inherits(lp, "try-error")) {
+    path <- system.file("examples/ex.pptx", package = "docxtractr")
+    pdf <- convert_to_pdf(path, pdf_file = tempfile(fileext = ".pdf"))
+    expect_true(file.size(pdf) > 0)
+  }
+})
+
+test_that("we can convert a DOCX to PDF if LibreOffice Installed", {
+  lp = try({
+    docxtractr:::lo_find()
+  }, silent = TRUE)
+  if (!inherits(lp, "try-error")) {
+    path <- system.file("examples/data.docx", package = "docxtractr")
+    pdf <- convert_to_pdf(path, pdf_file = tempfile(fileext = ".pdf"))
+    expect_true(file.size(pdf) > 0)
+  }
+})
+