public virtual void ComparePdfA3uRGBSpanishJPG()
{
String testName = "comparePdfA3uRGBSpanishJPG";
String filename = "spanish_01";
String expectedPdfPath = TEST_DOCUMENTS_DIRECTORY + filename + "_a3u.pdf";
String resultPdfPath = GetTargetDirectory() + filename + "_" + testName + "_a3u.pdf";
Tesseract4OcrEngineProperties properties = new Tesseract4OcrEngineProperties(tesseractReader.GetTesseract4OcrEngineProperties
());
properties.SetPathToTessData(GetTessDataDirectory());
properties.SetLanguages(JavaCollectionsUtil.SingletonList <String>("spa"));
tesseractReader.SetTesseract4OcrEngineProperties(properties);
OcrPdfCreatorProperties ocrPdfCreatorProperties = new OcrPdfCreatorProperties();
ocrPdfCreatorProperties.SetPdfLang("en-US");
ocrPdfCreatorProperties.SetTitle("");
ocrPdfCreatorProperties.SetTextColor(DeviceRgb.BLACK);
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(tesseractReader, ocrPdfCreatorProperties);
PdfDocument doc = ocrPdfCreator.CreatePdfA(JavaCollectionsUtil.SingletonList <FileInfo>(new FileInfo(TEST_IMAGES_DIRECTORY
+ filename + ".jpg")), GetPdfWriter(resultPdfPath), GetRGBPdfOutputIntent());
NUnit.Framework.Assert.IsNotNull(doc);
doc.Close();
NUnit.Framework.Assert.IsNull(new CompareTool().CompareByContent(resultPdfPath, expectedPdfPath, GetTargetDirectory
(), "diff_"));
}
public virtual void InitTesseractProperties()
{
Tesseract4OcrEngineProperties ocrEngineProperties = new Tesseract4OcrEngineProperties();
ocrEngineProperties.SetPathToTessData(GetTessDataDirectory());
tesseractReader.SetTesseract4OcrEngineProperties(ocrEngineProperties);
}
public virtual void InitTesseractProperties()
{
Tesseract4OcrEngineProperties ocrEngineProperties = new Tesseract4OcrEngineProperties();
ocrEngineProperties.SetPathToTessData(new FileInfo(sourceFolder + "../../tessdata"));
tesseractReader.SetTesseract4OcrEngineProperties(ocrEngineProperties);
}
public virtual void CompareMultiLangImage()
{
String testName = "compareMultiLangImage";
String filename = "multilang";
String expectedPdfPath = TEST_DOCUMENTS_DIRECTORY + filename + "_" + testFileTypeName + ".pdf";
String resultPdfPath = GetTargetDirectory() + filename + "_" + testName + "_" + testFileTypeName + ".pdf";
try {
Tesseract4OcrEngineProperties properties = tesseractReader.GetTesseract4OcrEngineProperties();
properties.SetTextPositioning(TextPositioning.BY_WORDS);
properties.SetPathToTessData(GetTessDataDirectory());
properties.SetPageSegMode(3);
tesseractReader.SetTesseract4OcrEngineProperties(properties);
DoOcrAndSavePdfToPath(tesseractReader, TEST_IMAGES_DIRECTORY + filename + ".jpg", resultPdfPath, JavaUtil.ArraysAsList
("eng", "deu", "spa"), DeviceCmyk.BLACK);
NUnit.Framework.Assert.IsNull(new CompareTool().CompareByContent(resultPdfPath, expectedPdfPath, TEST_DOCUMENTS_DIRECTORY
, "diff_"));
}
finally {
NUnit.Framework.Assert.AreEqual(TextPositioning.BY_WORDS, tesseractReader.GetTesseract4OcrEngineProperties
().GetTextPositioning());
NUnit.Framework.Assert.AreEqual(3, tesseractReader.GetTesseract4OcrEngineProperties().GetPageSegMode().Value
);
}
}
public IntegrationTestHelper()
{
Tesseract4OcrEngineProperties ocrEngineProperties = new Tesseract4OcrEngineProperties();
ocrEngineProperties.SetPathToTessData(GetTessDataDirectory());
tesseractLibReader = new Tesseract4LibOcrEngine(ocrEngineProperties);
tesseractExecutableReader = new Tesseract4ExecutableOcrEngine(GetTesseractDirectory(), ocrEngineProperties
);
}
public virtual void TestEmptyPathToTessData()
{
NUnit.Framework.Assert.That(() => {
FileInfo file = new FileInfo(TEST_IMAGES_DIRECTORY + "spanish_01.jpg");
Tesseract4OcrEngineProperties properties = tesseractReader.GetTesseract4OcrEngineProperties().SetPathToTessData
(new FileInfo("."));
tesseractReader.SetTesseract4OcrEngineProperties(properties);
GetTextFromPdf(tesseractReader, file);
NUnit.Framework.Assert.AreEqual(new FileInfo("").FullName, tesseractReader.GetTesseract4OcrEngineProperties
().GetPathToTessData().FullName);
}
, NUnit.Framework.Throws.InstanceOf <Tesseract4OcrException>().With.Message.EqualTo(MessageFormatUtil.Format(Tesseract4OcrException.INCORRECT_LANGUAGE, "eng.traineddata", new FileInfo(".").FullName)))
;
}
/// <summary>
/// Perform OCR using provided path to image (imgPath)
/// and save result PDF document to "pdfPath".
/// </summary>
/// <remarks>
/// Perform OCR using provided path to image (imgPath)
/// and save result PDF document to "pdfPath".
/// (Method is used for compare tool)
/// </remarks>
protected internal virtual void DoOcrAndSavePdfToPath(AbstractTesseract4OcrEngine tesseractReader, String
imgPath, String pdfPath, IList <String> languages, IList <String> fonts, Color color)
{
if (languages != null)
{
Tesseract4OcrEngineProperties properties = tesseractReader.GetTesseract4OcrEngineProperties();
properties.SetLanguages(languages);
tesseractReader.SetTesseract4OcrEngineProperties(properties);
}
OcrPdfCreatorProperties properties_1 = new OcrPdfCreatorProperties();
properties_1.SetPdfLang("en-US");
properties_1.SetTitle("");
if (fonts != null && fonts.Count > 0)
{
FontProvider fontProvider = new FontProvider();
foreach (String fontPath in fonts)
{
String name = FONT_PATH_TO_FONT_NAME_MAP.Get(fontPath);
fontProvider.GetFontSet().AddFont(fontPath, PdfEncodings.IDENTITY_H, name);
}
properties_1.SetFontProvider(fontProvider);
}
if (color != null)
{
properties_1.SetTextColor(color);
}
if (languages != null)
{
NUnit.Framework.Assert.AreEqual(languages.Count, tesseractReader.GetTesseract4OcrEngineProperties().GetLanguages
().Count);
}
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(tesseractReader, properties_1);
try {
using (PdfWriter pdfWriter = GetPdfWriter(pdfPath)) {
PdfDocument doc = ocrPdfCreator.CreatePdf(JavaCollectionsUtil.SingletonList <FileInfo>(new FileInfo(imgPath
)), pdfWriter);
NUnit.Framework.Assert.IsNotNull(doc);
doc.Close();
}
}
catch (System.IO.IOException e) {
LOGGER.Error(e.Message);
}
}
/// <summary>
/// Perform OCR using provided path to image (imgPath)
/// and save result to text file.
/// </summary>
protected internal virtual void DoOcrAndSaveToTextFile(AbstractTesseract4OcrEngine tesseractReader, String
imgPath, String txtPath, IList <String> languages)
{
if (languages != null)
{
Tesseract4OcrEngineProperties properties = tesseractReader.GetTesseract4OcrEngineProperties();
properties.SetLanguages(languages);
tesseractReader.SetTesseract4OcrEngineProperties(properties);
}
tesseractReader.CreateTxtFile(JavaCollectionsUtil.SingletonList <FileInfo>(new FileInfo(imgPath)), new FileInfo
(txtPath));
if (languages != null)
{
NUnit.Framework.Assert.AreEqual(languages.Count, tesseractReader.GetTesseract4OcrEngineProperties().GetLanguages
().Count);
}
}
public virtual void CompareInvoiceFrontThaiImage()
{
String testName = "compareInvoiceFrontThaiImage";
String filename = "invoice_front_thai";
//Tesseract for Java and Tesseract for .NET give different output
//So we cannot use one reference pdf file for them
String expectedPdfPathJava = TEST_DOCUMENTS_DIRECTORY + filename + "_" + testFileTypeName + "_java.pdf";
String expectedPdfPathDotNet = TEST_DOCUMENTS_DIRECTORY + filename + "_" + testFileTypeName + "_dotnet.pdf";
String resultPdfPath = GetTargetDirectory() + filename + "_" + testName + "_" + testFileTypeName + ".pdf";
Tesseract4OcrEngineProperties properties = tesseractReader.GetTesseract4OcrEngineProperties();
properties.SetTextPositioning(TextPositioning.BY_WORDS_AND_LINES);
properties.SetPathToTessData(GetTessDataDirectory());
properties.SetLanguages(JavaUtil.ArraysAsList("tha", "eng"));
tesseractReader.SetTesseract4OcrEngineProperties(properties);
DoOcrAndSavePdfToPath(tesseractReader, TEST_IMAGES_DIRECTORY + filename + ".jpg", resultPdfPath, JavaUtil.ArraysAsList
("tha", "eng"), JavaUtil.ArraysAsList(NOTO_SANS_THAI_FONT_PATH, NOTO_SANS_FONT_PATH), DeviceRgb.RED);
bool javaTest = new CompareTool().CompareByContent(resultPdfPath, expectedPdfPathJava, TEST_DOCUMENTS_DIRECTORY
, "diff_") == null;
bool dotNetTest = new CompareTool().CompareByContent(resultPdfPath, expectedPdfPathDotNet, TEST_DOCUMENTS_DIRECTORY
, "diff_") == null;
NUnit.Framework.Assert.IsTrue(javaTest || dotNetTest);
}
public virtual void TestSpanishPNG()
{
String testName = "compareSpanishPNG";
String filename = "scanned_spa_01";
String expectedText1 = "¿Y SI ENSAYARA COMO ACTUAR?";
String expectedText2 = "¿Y SI ENSAYARA ACTUAR?";
String resultPdfPath = GetTargetDirectory() + filename + "_" + testName + "_" + testFileTypeName + ".pdf";
IList <String> languages = JavaUtil.ArraysAsList("spa", "spa_old");
Tesseract4OcrEngineProperties properties = tesseractReader.GetTesseract4OcrEngineProperties();
if (isExecutableReaderType)
{
properties.SetPreprocessingImages(false);
}
// locate text by words
properties.SetTextPositioning(TextPositioning.BY_WORDS);
properties.SetLanguages(languages);
tesseractReader.SetTesseract4OcrEngineProperties(properties);
OcrPdfCreatorProperties ocrPdfCreatorProperties = new OcrPdfCreatorProperties();
ocrPdfCreatorProperties.SetTextColor(DeviceCmyk.BLACK);
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(tesseractReader, ocrPdfCreatorProperties);
using (PdfWriter pdfWriter = GetPdfWriter(resultPdfPath)) {
ocrPdfCreator.CreatePdf(JavaCollectionsUtil.SingletonList <FileInfo>(new FileInfo(TEST_IMAGES_DIRECTORY + filename
+ ".png")), pdfWriter).Close();
}
try {
String result = GetTextFromPdfLayer(resultPdfPath, null, 1).Replace("\n", " ");
NUnit.Framework.Assert.IsTrue(result.Contains(expectedText1) || result.Contains(expectedText2));
}
finally {
NUnit.Framework.Assert.AreEqual(TextPositioning.BY_WORDS, tesseractReader.GetTesseract4OcrEngineProperties
().GetTextPositioning());
}
}
