focus on pdf documents

Author: pachadotdev

NAMESPACE

@@ -1,8 +1,6 @@
 # Generated by roxygen2: do not edit by hand
 
-S3method(print,tesseract)
 export(ocr)
-export(ocr_data)
 export(tesseract)
 export(tesseract_contributed_download)
 export(tesseract_download)

R/cpp11.R

@@ -32,22 +32,10 @@ get_param_values <- function(api, params) {
   .Call(`_cpp11tesseract_get_param_values`, api, params)
 }
 
-ocr_raw <- function(input, ptr, HOCR) {
-  .Call(`_cpp11tesseract_ocr_raw`, input, ptr, HOCR)
-}
-
 ocr_file <- function(file, ptr, HOCR) {
   .Call(`_cpp11tesseract_ocr_file`, file, ptr, HOCR)
 }
 
-ocr_raw_data <- function(input, ptr) {
-  .Call(`_cpp11tesseract_ocr_raw_data`, input, ptr)
-}
-
-ocr_file_data <- function(file, ptr) {
-  .Call(`_cpp11tesseract_ocr_file_data`, file, ptr)
-}
-
 n_pages <- function(file_path, opw, upw) {
   .Call(`_cpp11tesseract_n_pages`, file_path, opw, upw)
 }

R/images.R

@@ -6,15 +6,15 @@
 #' the package vignette for image preprocessing tips.
 #'
 #' The `ocr()` function returns plain text by default, or hOCR text if hOCR is set to `TRUE`.
-#' The `ocr_data()` function returns a data frame with a confidence rate and bounding box for
-#' each word in the text.
 #'
 #' @export
 #' @return character vector of text extracted from the file. If the file
 #'  is has TIFF or PDF extension, it will be a vector of length equal to the
 #'  number of pages.
 #' @family tesseract
 #' @param file file path or raw vector (png, tiff, jpeg, etc).
+#' @param pages a numeric vector of pages to extract text from. If `NULL` all
+#'  pages will be extracted.
 #' @param engine a tesseract engine created with [tesseract()]. Alternatively a
 #' language string which will be passed to [tesseract()].
 #' @param HOCR if `TRUE` return results as HOCR xml instead of plain text
@@ -26,63 +26,57 @@
 #' @references [Tesseract: Improving Quality](https://github.com/tesseract-ocr/tesseract/wiki/ImproveQuality)
 #' @examples
 #' # Simple example
-#' file <- system.file("examples", "testocr.png", package = "cpp11tesseract")
+#' file <- system.file("examples", "oscarwilde.pdf", package = "cpp11tesseract")
 #' text <- ocr(file)
 #' cat(text)
-ocr <- function(file, engine = tesseract("eng"), HOCR = FALSE, opw = "", upw = "") {
-  if (is.character(engine)) {
-    engine <- tesseract(engine)
-  }
+ocr <- function(file, pages = NULL, engine = tesseract("eng"), HOCR = FALSE, opw = "", upw = "") {
+  if (is.character(engine)) { engine <- tesseract(engine) }
+  if (is.numeric(pages)) { pages <- as.integer(pages) }
   stopifnot(inherits(engine, "externalptr"))
-  if (isTRUE(inherits(file, "magick-image"))) {
-    vapply(file, function(x) {
-      tmp <- tempfile(fileext = ".png")
-      on.exit(unlink(tmp))
-      magick::image_write(x, tmp, format = "PNG", density = "300x300")
-      ocr(tmp, engine = engine, HOCR = HOCR)
-    }, character(1))
-  } else if (isTRUE(is.character(file)) && isFALSE(is.pdf(file))) {
-    if (isFALSE(is.tiff(file))) {
-      vapply(file, ocr_file, character(1), ptr = engine, HOCR = HOCR, USE.NAMES = FALSE)
-    } else {
-      ocr(tiff_convert(file), engine, HOCR = HOCR)
-    }
-  } else if (isTRUE(is.raw(file))) {
-    ocr_raw(file, engine, HOCR = HOCR)
-  } else if (isTRUE(is.pdf(file))) {
-    n <- n_pages(file, opw = opw, upw = upw)
-    fout <- pdf_convert(file, format = "png", pages = 1:n, opw = opw, upw = upw)
-    out <- vapply(fout, function(x) ocr(x, engine = engine, HOCR = HOCR), character(1))
-    unlink(fout)
-    names(out) <- NULL
-    out
-  } else {
-    stop("Argument 'file' must be file-path, url or raw vector")
-  }
+  stopifnot(file.exists(file))
+  stopifnot(is.pdf(file))
+
+  if (is.null(pages)) { pages <- seq_len(n_pages(file, opw = opw, upw = upw)) }
+  fout <- pdf_convert(file, format = "png", pages = pages, opw = opw, upw = upw)
+  out <- vapply(fout, ocr_file, character(1), ptr = engine, HOCR = HOCR, USE.NAMES = FALSE)
+  unlink(fout)
+  out
 }
 
-#' @rdname ocr
-#' @export
-ocr_data <- function(file, engine = tesseract("eng")) {
-  if (is.character(engine)) {
-    engine <- tesseract(engine)
+is.pdf <- function(x) {
+  grepl("\\.pdf$", tolower(x))
+}
+
+pdf_convert <- function(pdf, format = "png", pages = NULL, dpi = 72,
+                        antialias = TRUE, opw = "", upw = "") {
+  config <- get_poppler_config()
+
+  if (isFALSE(config$render) || isFALSE(length(config$format) > 0)) {
+    stop("You version of libppoppler does not support rendering")
   }
-  stopifnot(inherits(engine, "externalptr"))
-  df_list <- if (inherits(file, "magick-image")) {
-    lapply(file, function(x) {
-      tmp <- tempfile(fileext = ".png")
-      on.exit(unlink(tmp))
-      magick::image_write(x, tmp, format = "PNG", density = "300x300")
-      ocr_data(tmp, engine = engine)
-    })
-  } else if (is.character(file)) {
-    lapply(file, function(im) {
-      ocr_file_data(im, ptr = engine)
-    })
-  } else if (is.raw(file)) {
-    list(ocr_raw_data(file, engine))
-  } else {
-    stop("Argument 'file' must be file-path, url or raw vector")
+
+  format <- match.arg(format, config$format)
+
+  if (is.null(pages)) {
+    pages <- seq_len(n_pages(file, opw = opw, upw = upw))
   }
-  df_as_tibble(do.call(rbind.data.frame, unname(df_list)))
+
+  if (isFALSE(is.numeric(pages)) || isFALSE(length(pages) > 0)) {
+    stop("Argument 'pages' must be a one-indexed vector of page numbers")
+  }
+
+  antialiasing <- isTRUE(antialias) || isTRUE(antialias == "draw")
+
+  text_antialiasing <- isTRUE(antialias) || isTRUE(antialias == "text")
+
+  dout <- tempdir()
+  suppressWarnings(try(dir.create(dout)))
+
+  filenames <- file.path(dout, sprintf(
+    "%s-%03d.%s",
+    tools::file_path_sans_ext(basename(pdf)),
+    pages, format
+  ))
+
+  poppler_convert(pdf, format, pages, filenames, dpi, opw, upw, antialiasing, text_antialiasing)
 }

R/ocr.R

@@ -35,14 +35,14 @@
 #'
 #' if (any("fra" %in% tesseract_info()$available)) {
 #'   french <- tesseract("fra")
-#'   file <- system.file("examples", "french.png", package = "cpp11tesseract")
+#'   file <- system.file("examples", "french.pdf", package = "cpp11tesseract")
 #'   text <- ocr(file, engine = french)
 #'   cat(text)
 #' }
 tesseract_download <- function(lang, datapath = NULL, model = c("fast", "best"), progress = interactive()) {
   stopifnot(is.character(lang))
   model <- match.arg(model)
-  if (!length(datapath)) {
+  if (is.null(length(datapath))) {
     warn_on_linux()
     datapath <- tesseract_info()$datapath
   }
@@ -85,7 +85,7 @@ tesseract_download <- function(lang, datapath = NULL, model = c("fast", "best"),
 #'
 #' if (any("grc_hist" %in% tesseract_info()$available)) {
 #'   greek <- tesseract("grc_hist")
-#'   file <- system.file("examples", "polytonicgreek.png", package = "cpp11tesseract")
+#'   file <- system.file("examples", "polytonicgreek.pdf", package = "cpp11tesseract")
 #'   text <- ocr(file, engine = greek)
 #'   cat(text)
 #' }

R/tessdata.R

@@ -1,7 +1,7 @@
 #' Tesseract Engine
 #'
 #' Create an OCR engine for a given language and control parameters. This can be used by
-#' the [ocr] and [ocr_data] functions to recognize text.
+#' the [ocr] function to recognize text.
 #'
 #' Tesseract control parameters can be set either via a named list in the
 #' `options` parameter, or in a `config` file text file which contains the parameter name
@@ -99,16 +99,6 @@ tesseract_engine <- function(datapath, language, configs, options) {
   tesseract_engine_internal(datapath, language, configs, opt_names, opt_values)
 }
 
-#' @export
-#' @noRd
-"print.tesseract" <- function(x, ...) {
-  info <- engine_info_internal(x)
-  cat("<tesseract engine>\n")
-  cat(" loaded:", info$loaded, "\n")
-  cat(" datapath:", info$datapath, "\n")
-  cat(" available:", info$available, "\n")
-}
-
 bail <- function(...) {
   stop(sprintf(...), call. = FALSE)
 }