void doOCROnCurrentPage() throws IOException, TikaException, SAXException {
if (config.getOCRStrategy().equals(NO_OCR)) {
return;
}
TesseractOCRConfig tesseractConfig =
context.get(TesseractOCRConfig.class, DEFAULT_TESSERACT_CONFIG);
TesseractOCRParser tesseractOCRParser = new TesseractOCRParser();
if (!tesseractOCRParser.hasTesseract(tesseractConfig)) {
throw new TikaException(
"Tesseract is not available. "
+ "Please set the OCR_STRATEGY to NO_OCR or configure Tesseract correctly");
}
PDFRenderer renderer = new PDFRenderer(pdDocument);
TemporaryResources tmp = new TemporaryResources();
try {
BufferedImage image = renderer.renderImage(pageIndex, 2.0f, config.getOCRImageType());
Path tmpFile = tmp.createTempFile();
try (OutputStream os = Files.newOutputStream(tmpFile)) {
// TODO: get output format from TesseractConfig
ImageIOUtil.writeImage(image, config.getOCRImageFormatName(), os, config.getOCRDPI());
}
try (InputStream is = TikaInputStream.get(tmpFile)) {
tesseractOCRParser.parseInline(is, xhtml, tesseractConfig);
}
} catch (IOException e) {
handleCatchableIOE(e);
} catch (SAXException e) {
throw new IOExceptionWithCause("error writing OCR content from PDF", e);
} finally {
tmp.dispose();
}
}
