batch rename PDF files based on their metadata (uid_doc_title-doc_author.pdf), using R + PDFtk

pdktk.r

# load XML package
library(XML)

# http://stackoverflow.com/a/5060203
html2txt <- function(str) {
  xpathApply(htmlParse(str, asText=TRUE),
             "//body//text()",
             xmlValue)[[1]]
}

f = sample(list.files())
# f[ !grepl("_", f) ]

for (i in rev(f)) {

  cat("[", which(f == i), "]", i, "\n")

  system(paste("pdftk", i, "data_dump output nfo.txt"))
  t = try(readLines("nfo.txt"), silent = TRUE)

  if ("try-error" %in% class(t)) {

    cat(" -- ERROR!\n")
    next

  }

  a = gsub("InfoValue: ", "", t[ which(t == "InfoKey: Author") + 1 ])
  t = gsub("InfoValue: ", "", t[ which(t == "InfoKey: Title") + 1 ])
  n = gsub("\\.pdf", gsub(" ", "_", paste0("_", t, "-", a, ".pdf")), i)
  n = html2txt(n)

  n = gsub("&#40;", "(", n)
  n = gsub("&#41;", ")", n)
  n = gsub("&ndash;", "-", n)
  # n = gsub("B9780080970868", "", n)
  
  cat(" -->", n, "\n")
  # stopifnot(!grepl("&", n))

  file.rename(i, gsub("/|:", ",", n))
  
}