Examples with TesseractOCRParser org.apache.tika.parser.ocr.TesseractOCRParser used on opensource projects

Example 1 with TesseractOCRParser

use of org.apache.tika.parser.ocr.TesseractOCRParser in project tika by apache.

the class TikaResourceTest method testPDFOCRConfig.

//TIKA-2290
@Test
public void testPDFOCRConfig() throws Exception {
    if (!new TesseractOCRParser().hasTesseract(new TesseractOCRConfig())) {
        return;
    }
    Response response = WebClient.create(endPoint + TIKA_PATH).type("application/pdf").accept("text/plain").header(TikaResource.X_TIKA_PDF_HEADER_PREFIX + "OcrStrategy", "no_ocr").put(ClassLoader.getSystemResourceAsStream("testOCR.pdf"));
    String responseMsg = getStringFromInputStream((InputStream) response.getEntity());
    assertTrue(responseMsg.trim().equals(""));
    response = WebClient.create(endPoint + TIKA_PATH).type("application/pdf").accept("text/plain").header(TikaResource.X_TIKA_PDF_HEADER_PREFIX + "OcrStrategy", "ocr_only").put(ClassLoader.getSystemResourceAsStream("testOCR.pdf"));
    responseMsg = getStringFromInputStream((InputStream) response.getEntity());
    assertContains("Happy New Year 2003!", responseMsg);
    //now try a bad value
    response = WebClient.create(endPoint + TIKA_PATH).type("application/pdf").accept("text/plain").header(TikaResource.X_TIKA_PDF_HEADER_PREFIX + "OcrStrategy", "non-sense-value").put(ClassLoader.getSystemResourceAsStream("testOCR.pdf"));
    assertEquals(500, response.getStatus());
}

Example 2 with TesseractOCRParser

use of org.apache.tika.parser.ocr.TesseractOCRParser in project tika by apache.

the class AbstractPDF2XHTML method doOCROnCurrentPage.

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(), config.getOcrImageQuality());
        }
        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();
    }
}

Example 3 with TesseractOCRParser

use of org.apache.tika.parser.ocr.TesseractOCRParser in project tika by apache.

the class BundleIT method testTesseractParser.

@Test
public void testTesseractParser() throws Exception {
    ContentHandler handler = new BodyContentHandler();
    ParseContext context = new ParseContext();
    Parser tesseractParser = new TesseractOCRParser();
    try (InputStream stream = new FileInputStream("src/test/resources/testOCR.jpg")) {
        tesseractParser.parse(stream, handler, new Metadata(), context);
    }
}