Esempio n. 1
0
// Reads all boxes from the given filename.
// Reads a specific target_page number if >= 0, or all pages otherwise.
// Skips blanks if skip_blanks is true.
// The UTF-8 label of the box is put in texts, and the full box definition as
// a string is put in box_texts, with the corresponding page number in pages.
// Each of the output vectors is optional (may be NULL).
// Returns false if no boxes are found.
bool ReadAllBoxes(int target_page, bool skip_blanks, const STRING& filename,
                  GenericVector<TBOX>* boxes,
                  GenericVector<STRING>* texts,
                  GenericVector<STRING>* box_texts,
                  GenericVector<int>* pages) {
  GenericVector<char> box_data;
  if (!tesseract::LoadDataFromFile(BoxFileName(filename), &box_data))
    return false;
  return ReadMemBoxes(target_page, skip_blanks, &box_data[0], boxes, texts,
                      box_texts, pages);
}
Esempio n. 2
0
// Reads all boxes from the given filename.
// Reads a specific target_page number if >= 0, or all pages otherwise.
// Skips blanks if skip_blanks is true.
// The UTF-8 label of the box is put in texts, and the full box definition as
// a string is put in box_texts, with the corresponding page number in pages.
// Each of the output vectors is optional (may be NULL).
// Returns false if no boxes are found.
bool ReadAllBoxes(int target_page, bool skip_blanks, const STRING& filename,
                  GenericVector<TBOX>* boxes,
                  GenericVector<STRING>* texts,
                  GenericVector<STRING>* box_texts,
                  GenericVector<int>* pages) {
  GenericVector<char> box_data;
  if (!tesseract::LoadDataFromFile(BoxFileName(filename), &box_data))
    return false;
  // Convert the array of bytes to a string, so it can be used by the parser.
  box_data.push_back('\0');
  return ReadMemBoxes(target_page, skip_blanks, &box_data[0], boxes, texts,
                      box_texts, pages);
}
Esempio n. 3
0
// Parses the text string as a box file and adds any discovered boxes that
// match the page number. Returns false on error.
bool ImageData::AddBoxes(const char* box_text) {
  if (box_text != NULL && box_text[0] != '\0') {
    GenericVector<TBOX> boxes;
    GenericVector<STRING> texts;
    GenericVector<int> box_pages;
    if (ReadMemBoxes(page_number_, false, box_text, &boxes,
                     &texts, NULL, &box_pages)) {
      AddBoxes(boxes, texts, box_pages);
      return true;
    } else {
      tprintf("Error: No boxes for page %d from image %s!\n",
              page_number_, imagefilename_.string());
    }
  }
  return false;
}