File indexing completed on 2024-05-12 15:55:08

 /************************************************************************
  *                                  *
  *  This file is part of Kooka, a scanning/OCR application using    *
  *  Qt <http://www.qt.io> and KDE Frameworks <http://www.kde.org>.  *
  *                                  *
  *  Copyright (C) 2020      Jonathan Marten <jjm@keelhaul.me.uk>    *
  *                                  *
  *  Kooka is free software; you can redistribute it and/or modify it    *
  *  under the terms of the GNU Library General Public License as    *
  *  published by the Free Software Foundation and appearing in the  *
  *  file COPYING included in the packaging of this file;  either    *
  *  version 2 of the License, or (at your option) any later version.    *
  *                                  *
  *  As a special exception, permission is given to link this program    *
  *  with any version of the KADMOS OCR/ICR engine (a product of     *
  *  reRecognition GmbH, Kreuzlingen), and distribute the resulting  *
  *  executable without including the source code for KADMOS in the  *
  *  source distribution.                        *
  *                                  *
  *  This program is distributed in the hope that it will be useful, *
  *  but WITHOUT ANY WARRANTY; without even the implied warranty of  *
  *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the   *
  *  GNU General Public License for more details.            *
  *                                  *
  *  You should have received a copy of the GNU General Public       *
  *  License along with this program;  see the file COPYING.  If     *
  *  not, see <http://www.gnu.org/licenses/>.                *
  *                                  *
  ************************************************************************/
 
 #include "ocrtesseractengine.h"
 
 #include <qregexp.h>
 #include <qfile.h>
 #include <qfileinfo.h>
 
 #include <QXmlStreamReader>
 
 #include <klocalizedstring.h>
 #include <kpluginfactory.h>
 
 #include "imageformat.h"
 #include "kookasettings.h"
 #include "ocrtesseractdialog.h"
 #include "executablepathdialogue.h"
 #include "ocr_logging.h"
 
 
 K_PLUGIN_FACTORY_WITH_JSON(OcrTesseractEngineFactory, "kookaocr-tesseract.json", registerPlugin<OcrTesseractEngine>();)
 #include "ocrtesseractengine.moc"
 
 
 OcrTesseractEngine::OcrTesseractEngine(QObject *pnt, const QVariantList &args)
     : AbstractOcrEngine(pnt, "OcrTesseractEngine")
 {
     m_ocrImageIn = QString();
     m_tempHOCROut = QString();
     m_tesseractVersion = 0;
 }
 
 
 AbstractOcrDialogue *OcrTesseractEngine::createOcrDialogue(AbstractOcrEngine *plugin, QWidget *pnt)
 {
     return (new OcrTesseractDialog(plugin, pnt));
 }
 
 
 bool OcrTesseractEngine::createOcrProcess(AbstractOcrDialogue *dia, ScanImage::Ptr img)
 {
     OcrTesseractDialog *parentDialog = static_cast<OcrTesseractDialog *>(dia);
     m_tesseractVersion = parentDialog->getNumVersion();
 
     const QString cmd = parentDialog->getOCRCmd();
 
     const QString ocrResultFile = tempSaveImage(img, ImageFormat("PGM"), 8);
     setResultImage(ocrResultFile);
     // TODO: if the input file is local and is readable by Tesseract,
     // can use it directly (but don't delete it afterwards!)
     m_ocrImageIn = tempSaveImage(img, ImageFormat("PNG"), 8);
 
     QProcess *proc = initOcrProcess();          // start process for OCR
     QStringList args;                   // arguments for process
 
     // Input file
     args << QFile::encodeName(m_ocrImageIn);        // file with the input image
 
     // Output base name
     m_tempHOCROut = tempFileName("");           // Tesseract just wants base name
     args << QFile::encodeName(m_tempHOCROut);       // the HOCR result file
     m_tempHOCROut += ".hocr";               // suffix that it will have
 
     // Language
     QString s = KookaSettings::ocrTesseractLanguage();
     if (!s.isEmpty()) args << "-l" << s;
 
     // User words
     QUrl u = KookaSettings::ocrTesseractUserWords();
     if (u.isValid()) args << "--user-words" << u.toLocalFile();
 
     // User patterns
     u = KookaSettings::ocrTesseractUserPatterns();
     if (u.isValid()) args << "--user-patterns" << u.toLocalFile();
 
     // Page segmentation mode
     s = KookaSettings::ocrTesseractSegmentationMode();
     if (!s.isEmpty()) args << "--psm" << s;
 
     // OCR engine mode
     s = KookaSettings::ocrTesseractEngineMode();
     if (!s.isEmpty()) args << "--oem" << s;
 
     //if (verboseDebug()) args << "-v";
 
     s = KookaSettings::ocrTesseractExtraArguments();
     if (!s.isEmpty()) args << s;
 
     // Output format.  This option generates HOCR (HTML with OCR markup)
     // as specificied at http://kba.cloud/hocr-spec/1.2/
     args << "hocr";
 
     proc->setProgram(cmd);
     proc->setArguments(args);
 
     proc->setProcessChannelMode(QProcess::SeparateChannels);
     m_tempStdoutLog = tempFileName("stdout.log");
     proc->setStandardOutputFile(m_tempStdoutLog);
 
     return (runOcrProcess());
 }
 
 
 QStringList OcrTesseractEngine::tempFiles(bool retain)
 {
     QStringList result;
     result << m_ocrImageIn;
     result << m_tempHOCROut;
     result << m_tempStdoutLog;
     return (result);
 }
 
 
 bool OcrTesseractEngine::finishedOcrProcess(QProcess *proc)
 {
     qCDebug(OCR_LOG);
     QString errStr = readHOCR(m_tempHOCROut);       // parse the OCR results
     if (errStr.isEmpty()) return (true);        // parsed successfully
 
     setErrorText(errStr);               // record the parse error
     return (false);                 // parsing failed
 }
 
 
 QString OcrTesseractEngine::readHOCR(const QString &fileName)
 {
     // some basic checks on the file
     QFileInfo fi(fileName);
     if (!fi.exists()) return (xi18nc("@info", "File <filename>%1</filename> does not exist", fileName));
     if (!fi.isReadable()) return (xi18nc("@info", "File <filename>%1</filename> unreadable", fileName));
 
     qCDebug(OCR_LOG) << "Starting to analyse HOCR" << fileName;
 
     QFile file(fileName);
     if (!file.open(QIODevice::ReadOnly))
     {
         return (xi18nc("@info", "Cannot open file <filename>%1</filename>", fileName));
     }
 
     startResultDocument();              // start document to receive results
 
     QXmlStreamReader reader(&file);
     while (!reader.atEnd())
     {
         reader.readNext();              // get next XML token
         if (!reader.isStartElement()) continue;     // only interested in element start
 
         // We only take note of elements defining new paragraphs, new lines and words.
         // Examples of these are:
         //
         //   <p class='ocr_par' id='par_1_1' lang='eng' title="bbox 52 35 1469 84">
         //   <span class='ocr_line' id='line_1_1' title="bbox 52 35 1469 84; baseline 0.001 -10; x_size 41; x_descenders 9; x_ascenders 8">
         //   <span class='ocrx_word' id='word_1_1' title='bbox 52 42 123 74; x_wconf 92'>The</span> 
         //
         // These same HTML elements may specify other page layout data which are
         // not supported, but the class ID allows them to be distinguished.
 
         const QStringRef name = reader.name();
         if (name=="span" || name=="p")          // may be either SPAN or P element
         {
             const QStringRef cls = reader.attributes().value("class");
             if (cls=="ocr_par" || cls=="ocr_line")  // paragraph or line start
             {
                 // The start of a paragraph is always followed by the start of a line.
                 // Generating a new output line for both means that paragraphs are
                 // separated by blank lines, as intended.
                 startLine();
             }
             else if (cls=="ocrx_word")
             {
                 OcrWordData wd;
 
                 // The TITLE attribute of the SPAN element indicates the word bounding box.
                 const QString ttl = reader.attributes().value("title").toString();
 
                 QRegExp rx("bbox\\s+(\\d+)\\s+(\\d+)\\s+(\\d+)\\s+(\\d+);");
                 if (!ttl.isEmpty() && ttl.contains(rx))
                 {
                     QRect wordRect;
                     wordRect.setLeft(rx.cap(1).toInt());
                     wordRect.setTop(rx.cap(2).toInt());
                     wordRect.setRight(rx.cap(3).toInt());
                     wordRect.setBottom(rx.cap(4).toInt());
                     wd.setProperty(OcrWordData::Rectangle, wordRect);
                 }
 
                 // The contained text is the recognised word.  No child HTML elements
                 // are expected inside a word SPAN.
                 QString text = reader.readElementText(QXmlStreamReader::SkipChildElements);
                 addWord(text, wd);
             }
         }
     }
 
     if (reader.hasError())
     {
         qCDebug(OCR_LOG) << "XML reader error, line" << reader.lineNumber() << reader.error();
         return (i18n("HOCR parsing error, %1", reader.errorString()));
     }
 
     finishResultDocument();             // finished with output document
     file.close();                   // finished with HOCR file
 
     qCDebug(OCR_LOG) << "Finished analysing HOCR";
 
     return (QString());                 // no error detected
 }
 
 
 void OcrTesseractEngine::openAdvancedSettings()
 {
     ExecutablePathDialogue d(nullptr);
 
     QString exec = KookaSettings::ocrTesseractBinary();
     if (exec.isEmpty())
     {
         KConfigSkeletonItem *ski = KookaSettings::self()->ocrTesseractBinaryItem();
         ski->setDefault();
         exec = KookaSettings::ocrTesseractBinary();
     }
 
     d.setPath(exec);
     d.setLabel(i18n("Name or path of the Tesseract executable:"));
     if (!d.exec()) return;
 
     KookaSettings::setOcrTesseractBinary(d.path());
 }