Extract notes from a pptx slide and export to markdown or docx

extract_pptx_notes.R

extract_pptx_notes <- function(pp_file, format = c("md", "docx")) {
  if (!requireNamespace("officer", quietly = TRUE))
    stop("pacakge 'officer' required.")
  if (!requireNamespace("xml2", quietly = TRUE))
    stop("pacakge 'xml2' required.")
  if (!requireNamespace("rmarkdown", quietly = TRUE))
    stop("pacakge 'rmarkdown' required.")
  if (!requireNamespace("glue", quietly = TRUE))
    stop("pacakge 'glue' required.")

  format <- match.arg(format)

  pp <- read_pptx(pp_file)

  file_sans_ext <- tools::file_path_sans_ext(pp_file)

  slides_with_notes_meta <- pp$notesSlide$get_metadata()
  slides_with_notes_meta$notes <- as.numeric(gsub("notesSlide(\\d{1,3}).*", "\\1", rownames(slides_with_notes_meta)))
  slides_with_notes_meta <- slides_with_notes_meta[!grepl("notesMaster", slides_with_notes_meta$target), ]
  slides_with_notes_meta$slide <- as.numeric(gsub(".+slide(\\d{1,3})\\.xml", "\\1", slides_with_notes_meta$target))

  if (!nrow(slides_with_notes_meta)) stop("No notes in this presentation")

  slide_nums <- seq(1, max(slides_with_notes_meta$slide))

  notes <- lapply(slide_nums, \(x) {
    notes_slide <- slides_with_notes_meta[slides_with_notes_meta$slide == x, "notes"]
    if (!length(notes_slide)) return(character(0))
    slide <- pp$notesSlide$get_slide(notes_slide)
    xml <- slide$get()
    # xpath search from here: https://robaboukhalil.medium.com/your-slide-deck-is-a-zip-file-in-disguise-36bb14f11c0b
    xpath <- "//*[local-name()='txBody']/*[local-name()='p']/*[local-name()='r']/*[local-name()='t']/text()"
    node <- xml_find_all(xml, xpath)
    as.character(node)
  })

  names(notes) <- paste("Slide", slide_nums)

  out <- paste0(file_sans_ext, "_notes.md")

  if (file.exists(out)) {
    overwrite <- askYesNo(glue("File {out} already exists. Overwrite?"))
    if (!overwrite) stop("Quitting", call. = FALSE)
    file.remove(out)
  }

  cat(glue('---
title: "{basename(file_sans_ext)}"
output: word_document
date: "{Sys.Date()}"
---\n\n\n'), file = out)

  for (n in names(notes)) {
    cat(paste0("## ", n, ":\n\n"), file = out, append = TRUE)
    if (length(notes[[n]])) {
      cat(notes[[n]], file = out, sep = "\n\n", append = TRUE)
      cat("\n", file = out, append = TRUE)
    }
  }

  if (format == "docx") {
    out_docx <- paste0(file_sans_ext, "_notes.docx")
    return(rmarkdown::render(out, output_file = out_docx))
  }

  out
}