Examples with JAXBPageTranscript eu.transkribus.core.model.beans.JAXBPageTranscript used on opensource projects

Example 6 with JAXBPageTranscript

use of eu.transkribus.core.model.beans.JAXBPageTranscript in project TranskribusCore by Transkribus.

the class ExportCache method storePageTranscripts4Export.

public void storePageTranscripts4Export(TrpDoc doc, Set<Integer> pageIndices, IProgressMonitor monitor, String versionStatus, int pageIdx, TrpTranscriptMetadata loadedTranscript) throws Exception {
    pageTranscripts = new ArrayList<JAXBPageTranscript>();
    List<TrpPage> pages = doc.getPages();
    int totalPages = pages.size();
    int c = 0;
    for (int i = 0; i < totalPages; ++i) {
        if (pageIndices != null && !pageIndices.contains(i)) {
            // fill up with null to have the proper index of each page later on
            // logger.debug(" add null to transcripts " + i);
            pageTranscripts.add(null);
            continue;
        }
        if (monitor != null && monitor.isCanceled()) {
            throw new Exception("User canceled the export");
        }
        TrpPage page = pages.get(i);
        TrpTranscriptMetadata md = null;
        if (versionStatus.contains("Latest")) {
            // current transcript
            md = page.getCurrentTranscript();
        } else if (versionStatus.contains("Loaded")) {
            // if loaded page idx == i than we can export the loaded version and for all other pages the latest
            if (i == pageIdx && loadedTranscript != null) {
                md = loadedTranscript;
            // String loadedStatus = loadedTranscript.getStatus().getStr();
            // md = page.getTranscriptWithStatus(loadedStatus);
            } else {
                md = page.getCurrentTranscript();
            }
        } else {
            // logger.debug("We want to export pages with status: " + versionStatus);
            md = page.getTranscriptWithStatusOrNull(versionStatus);
        }
        /*
			 * for pages where we have not found versions with the defined status -> remove from the page list so that
			 * they will not exported
			 */
        if (md == null) {
            // pageIndices contained but has not the desired status -> add transcript 0;
            if (pageIndices != null && pageIndices.contains(i)) {
                // logger.debug("remove page index " + i);
                pageIndices.remove(new Integer(i));
                pageTranscripts.add(null);
            }
            continue;
        }
        JAXBPageTranscript tr = new JAXBPageTranscript(md);
        tr.build();
        pageTranscripts.add(tr);
        logger.debug("Loaded Transcript from page " + (i + 1));
        if (monitor != null) {
            monitor.setTaskName("Loaded Transcript from page " + (i + 1));
            monitor.worked(++c);
        }
    }
}

Example 7 with JAXBPageTranscript

use of eu.transkribus.core.model.beans.JAXBPageTranscript in project TranskribusCore by Transkribus.

the class PageXmlDaoTest method main.

public static void main(String[] args) {
    TrpDoc doc = FakeDocProvider.create(false);
    TrpTranscriptMetadata md = doc.getPages().get(0).getTranscripts().get(0);
    try {
        JAXBPageTranscript transcript = new JAXBPageTranscript(md);
        transcript.build();
        // JAXBPageTranscript transcript = TrpPageTranscriptBuilder.build(md);
        // get Source Document as String
        // DOMSource domSource = new DOMSource(transcript.getSourceDoc());
        // StringWriter writer = new StringWriter();
        // StreamResult result = new StreamResult(writer);
        // TransformerFactory tf = TransformerFactory.newInstance();
        // Transformer transformer = tf.newTransformer();
        // transformer.transform(domSource, result);
        // System.out.println("XML IN String format is: \n" + writer.toString());
        // check JaxB Element
        PcGtsType page = transcript.getPageData();
        if (page == null) {
            System.out.println("page XML is null");
            System.exit(0);
        }
        PageType pageType = page.getPage();
        System.out.println(page.getMetadata());
        if (pageType == null) {
            System.out.println("pagetype element is null");
            System.exit(0);
        }
        if (pageType.getTextRegionOrImageRegionOrLineDrawingRegion() == null) {
            System.out.println("Region list is null");
            System.exit(0);
        }
        int i = 0;
        for (TextRegionType tr : transcript.getPage().getTextRegions(true)) {
            tr.setId("" + i++);
        }
        List<TrpRegionType> regions = pageType.getTextRegionOrImageRegionOrLineDrawingRegion();
        for (RegionType r : regions) {
            if (r instanceof TextRegionType) {
                TextRegionType t = (TextRegionType) r;
                System.out.println(t.getId());
            }
        }
    } catch (IllegalArgumentException | IOException e) {
        e.printStackTrace();
    }
// try {
// PrimaPageTranscript ppt = PageXmlDao.getPrimaPageTranscript(md);
// Page page = ppt.getPageData();
// 
// System.out.println(page.getImageFilename());
// Region r = page.getLayout().getRegion("tempReg357564684568544579089");
// System.out.println(r.getType().getName());
// //			System.out.println(page.getLayout().getParentChildRelation(r.getType(), r.getId().toString()).getRelationType());
// 
// IdRegister idr = r.getIdRegister();
// //			idr.
// System.out.println(idr);
// } catch (IllegalArgumentException e) {
// 
// e.printStackTrace();
// } catch (MalformedURLException e) {
// 
// e.printStackTrace();
// } catch (UnsupportedFormatVersionException e) {
// 
// e.printStackTrace();
// }
}

Example 8 with JAXBPageTranscript

use of eu.transkribus.core.model.beans.JAXBPageTranscript in project TranskribusCore by Transkribus.

the class TrpXlsxBuilder method writeXlsxForDoc.

public static void writeXlsxForDoc(TrpDoc doc, boolean wordBased, File exportFile, Set<Integer> pageIndices, IProgressMonitor monitor, ExportCache cache) throws NoTagsException, Exception {
    if (cache == null) {
        throw new IllegalArgumentException("ExportCache must not be null.");
    }
    if (cache.getCustomTagMapForDoc().isEmpty()) {
        logger.info("No tags to store -> Xlsx export cancelled");
        throw new NoTagsException("No tags available to store into Xlsx");
    }
    List<TrpPage> pages = doc.getPages();
    String exportPath = exportFile.getPath();
    Set<String> selectedTags = cache.getOnlySelectedTagnames(ExportUtils.getOnlyWantedTagnames(CustomTagFactory.getRegisteredTagNames()));
    int totalPages = pageIndices == null ? pages.size() : pageIndices.size();
    if (monitor != null) {
        monitor.beginTask("Exporting to Excel", totalPages);
    }
    wb = new XSSFWorkbook();
    int c = 0;
    for (int i = 0; i < pages.size(); ++i) {
        if (pageIndices != null && !pageIndices.contains(i))
            continue;
        if (monitor != null) {
            if (monitor.isCanceled()) {
                throw new InterruptedException("Export was canceled by user");
            // logger.debug("Xlsx export cancelled!");
            // return;
            }
            monitor.subTask("Processing page " + (c + 1));
        }
        TrpPage page = pages.get(i);
        // try to get previously loaded JAXB transcript
        JAXBPageTranscript tr = null;
        if (cache != null) {
            tr = cache.getPageTranscriptAtIndex(i);
        }
        if (tr == null) {
            TrpTranscriptMetadata md = page.getCurrentTranscript();
            tr = new JAXBPageTranscript(md);
            tr.build();
        }
        // old version
        // TrpPage page = pages.get(i);
        // TrpTranscriptMetadata md = page.getCurrentTranscript();
        // JAXBPageTranscript tr = new JAXBPageTranscript(md);
        // tr.build();
        TrpPageType trpPage = tr.getPage();
        logger.debug("writing xlsx for page " + (i + 1) + "/" + doc.getNPages());
        List<TrpTextRegionType> textRegions = trpPage.getTextRegions(true);
        for (int j = 0; j < textRegions.size(); ++j) {
            TrpTextRegionType r = textRegions.get(j);
            List<TextLineType> lines = r.getTextLine();
            for (int k = 0; k < lines.size(); ++k) {
                TrpTextLineType trpL = (TrpTextLineType) lines.get(k);
                List<WordType> words = trpL.getWord();
                if (wordBased) {
                    for (int l = 0; l < words.size(); ++l) {
                        TrpWordType w = (TrpWordType) words.get(l);
                        writeTagsForShapeElement(w, trpL.getUnicodeText(), String.valueOf(doc.getId()), String.valueOf(page.getPageNr()), r.getId(), trpL.getId(), w.getId(), selectedTags);
                    }
                } else {
                    writeTagsForShapeElement(trpL, trpL.getUnicodeText(), String.valueOf(doc.getId()), String.valueOf(page.getPageNr()), r.getId(), trpL.getId(), "", selectedTags);
                }
            }
        }
        ++c;
        if (monitor != null) {
            monitor.worked(c);
        }
    }
    /*
		 * auto size the columns
		 */
    for (int i = 0; i < wb.getNumberOfSheets(); i++) {
        int numberOfCells = 0;
        Iterator rowIterator = wb.getSheetAt(i).rowIterator();
        /**
         * Escape the header row *
         */
        if (rowIterator.hasNext()) {
            Row headerRow = (Row) rowIterator.next();
            // get the number of cells in the header row
            numberOfCells = headerRow.getPhysicalNumberOfCells();
            for (int j = 0; j < numberOfCells; j++) {
                wb.getSheetAt(i).autoSizeColumn(j);
            }
        }
    }
    FileOutputStream fOut;
    try {
        // means no tags at all
        if (wb.getNumberOfSheets() == 0) {
            throw new IOException("Sorry - No tags available for export");
        }
        fOut = new FileOutputStream(exportPath);
        wb.write(fOut);
        fOut.close();
    } catch (IOException e) {
        // TODO Auto-generated catch block
        e.printStackTrace();
        throw e;
    }
    logger.info("wrote xlsx to: " + exportPath);
}

Example 9 with JAXBPageTranscript

use of eu.transkribus.core.model.beans.JAXBPageTranscript in project TranskribusCore by Transkribus.

the class DocxBuilder method writeDocxForDoc.

public static void writeDocxForDoc(TrpDoc doc, boolean wordBased, boolean writeTags, boolean doBlackeningSensibleData, File file, Set<Integer> pageIndices, IProgressMonitor monitor, boolean createTitle, boolean markUnclear, boolean expandAbbreviations, boolean replaceAbbrevs, boolean keepLineBreaks, boolean showSuppliedInBrackets, boolean ignoreSuppliedTags, ExportCache cache) throws JAXBException, IOException, Docx4JException, InterruptedException {
    // ch.qos.logback.classic.Logger root = logger.getClass().get(ch.qos.logback.classic.Logger) org.slf4j.LoggerFactory.getLogger(ch.qos.logback.classic.Logger.ROOT_LOGGER_NAME);
    ((ch.qos.logback.classic.Logger) logger).setLevel(ch.qos.logback.classic.Level.DEBUG);
    exportTags = writeTags;
    doBlackening = doBlackeningSensibleData;
    tagnames = cache.getOnlySelectedTagnames(ExportUtils.getOnlyWantedTagnames(CustomTagFactory.getRegisteredTagNames()));
    markUnclearWords = markUnclear;
    expandAbbrevs = expandAbbreviations;
    preserveLineBreaks = keepLineBreaks;
    substituteAbbrevs = replaceAbbrevs;
    showSuppliedWithBrackets = showSuppliedInBrackets;
    ignoreSupplied = ignoreSuppliedTags;
    /*
		 * get all names of tags
		 */
    // tagnames = CustomTagFactory.getRegisteredTagNames();
    // main document part
    wordMLPackage = WordprocessingMLPackage.createPackage();
    MainDocumentPart mdp = wordMLPackage.getMainDocumentPart();
    org.docx4j.wml.ObjectFactory factory = Context.getWmlObjectFactory();
    List<TrpPage> pages = doc.getPages();
    int totalPages = pageIndices == null ? pages.size() : pageIndices.size();
    if (monitor != null) {
        monitor.beginTask("Exporting to docx", totalPages);
    }
    int c = 0;
    boolean atLeastOnePageWritten = false;
    // can be used as page break every time we need one
    Br objBr = new Br();
    objBr.setType(STBrType.PAGE);
    P pageBreakP = factory.createP();
    pageBreakP.getContent().add(objBr);
    for (int i = 0; i < pages.size(); ++i) {
        if (pageIndices != null && !pageIndices.contains(i))
            continue;
        if (!atLeastOnePageWritten && createTitle) {
            addTitlePage(doc, mdp);
            // add page break
            mdp.addObject(pageBreakP);
        }
        if (monitor != null) {
            if (monitor.isCanceled()) {
                throw new InterruptedException("Export canceled by the user");
            // logger.debug("docx export cancelled!");
            // return;
            }
            monitor.subTask("Processing page " + (c + 1));
        }
        // TrpPage page = pages.get(i);
        // TrpTranscriptMetadata md = page.getCurrentTranscript();
        // JAXBPageTranscript tr = new JAXBPageTranscript(md);
        // tr.build();
        JAXBPageTranscript tr = null;
        if (cache != null) {
            tr = cache.getPageTranscriptAtIndex(i);
        }
        if (tr == null) {
            TrpPage page = pages.get(i);
            TrpTranscriptMetadata md = page.getCurrentTranscript();
            // md.getStatus().equals("Done");
            tr = new JAXBPageTranscript(md);
            tr.build();
        }
        TrpPageType trpPage = tr.getPage();
        logger.debug("writing docx for the page " + (i + 1) + "/" + doc.getNPages());
        writeDocxForTranscriptWithTables(mdp, trpPage, wordBased, preserveLineBreaks);
        atLeastOnePageWritten = true;
        ++c;
        if (monitor != null) {
            monitor.worked(c);
        }
    }
    P p = factory.createP();
    mdp.getContent().add(p);
    addComplexField(p, " INDEX \\e \"", "\" \\c \"1\" \\z \"1031\"");
    FieldUpdater updater = new FieldUpdater(wordMLPackage);
    updater.update(true);
    // write tags at end of last page
    if (false) {
        // RtfText headline = RtfText.text("Person names in this document (amount of found persons: " + persons.size() + ")", "\n");
        logger.debug("export tags ");
        boolean firstExport = true;
        // tagnames = all user choosen tags via export dialog
        for (String currTagname : tagnames) {
            // logger.debug("curr tagname " + currTagname);
            // get all custom tags with currTagname and text
            HashMap<CustomTag, String> allTagsOfThisTagname = cache.getTags(currTagname);
            // one paragraph for each tagname
            org.docx4j.wml.P p4Tag = factory.createP();
            if (allTagsOfThisTagname.size() > 0 && !currTagname.equals("textStyle") && !currTagname.equals("gap") && !currTagname.equals("comment")) {
                // new page if tag export starts
                if (firstExport) {
                    // Br objBr = new Br();
                    // objBr.setType(STBrType.PAGE);
                    p4Tag.getContent().add(objBr);
                    firstExport = false;
                }
                // logger.debug("allTagsOfThisTagname " + allTagsOfThisTagname.size());
                // one run for headline and thanfor each entry
                org.docx4j.wml.Text t = factory.createText();
                t.setValue(currTagname + " tags in this document: " + allTagsOfThisTagname.size());
                t.setSpace("preserve");
                org.docx4j.wml.R run = factory.createR();
                run.getContent().add(t);
                org.docx4j.wml.RPr rpr = factory.createRPr();
                org.docx4j.wml.BooleanDefaultTrue b = new org.docx4j.wml.BooleanDefaultTrue();
                b.setVal(true);
                U u = factory.createU();
                u.setVal(UnderlineEnumeration.SINGLE);
                rpr.setB(b);
                rpr.setU(u);
                run.setRPr(rpr);
                // this Br element is used break the current and go for next line
                Br br = factory.createBr();
                run.getContent().add(br);
                p4Tag.getContent().add(run);
                // ArrayList<RtfText> tagTexts = new ArrayList<RtfText>();
                Collection<String> valueSet = allTagsOfThisTagname.values();
                int l = 0;
                for (String currEntry : valueSet) {
                    org.docx4j.wml.R currRun = factory.createR();
                    org.docx4j.wml.Text currText = factory.createText();
                    currText.setValue(currEntry);
                    currText.setSpace("preserve");
                    currRun.getContent().add(currText);
                    // reuse linebreak
                    currRun.getContent().add(br);
                    p4Tag.getContent().add(currRun);
                }
            }
            mdp.getContent().add(p4Tag);
        }
    }
    // finally save the file
    wordMLPackage.save(file);
    logger.info("Saved " + file.getAbsolutePath());
}

Example 10 with JAXBPageTranscript

use of eu.transkribus.core.model.beans.JAXBPageTranscript in project TranskribusCore by Transkribus.

the class TrpXlsxTableBuilder method writeXlsxForTables.

public static void writeXlsxForTables(TrpDoc doc, File exportFile, Set<Integer> pageIndices, IProgressMonitor monitor, ExportCache cache) throws NoTablesException, IOException, InterruptedException {
    // TrpTableRegionType is contained in the regions too
    List<TrpPage> pages = doc.getPages();
    String exportPath = exportFile.getPath();
    int totalPages = pageIndices == null ? pages.size() : pageIndices.size();
    if (monitor != null) {
        monitor.beginTask("Exporting tables to Excel", totalPages);
    }
    wb = new XSSFWorkbook();
    int c = 0;
    int tableId = 0;
    for (int i = 0; i < pages.size(); ++i) {
        if (pageIndices != null && !pageIndices.contains(i))
            continue;
        if (monitor != null) {
            if (monitor.isCanceled()) {
                throw new InterruptedException("Export was canceled by user");
            // logger.debug("Xlsx export cancelled!");
            // return;
            }
            monitor.subTask("Processing page " + (c + 1));
        }
        TrpPage page = pages.get(i);
        // try to get previously loaded JAXB transcript
        JAXBPageTranscript tr = null;
        if (cache != null) {
            tr = cache.getPageTranscriptAtIndex(i);
        }
        if (tr == null) {
            TrpTranscriptMetadata md = page.getCurrentTranscript();
            tr = new JAXBPageTranscript(md);
            tr.build();
        }
        TrpPageType trpPage = tr.getPage();
        List<TrpRegionType> regions = trpPage.getRegions();
        for (int j = 0; j < regions.size(); ++j) {
            TrpRegionType r = regions.get(j);
            if (r instanceof TrpTableRegionType) {
                tableId++;
                logger.debug("is table");
                TrpTableRegionType table = (TrpTableRegionType) r;
                int cols = table.getNCols();
                int rows = table.getNRows();
                // double maxX = PageXmlUtils.buildPolygon(table.getCoords().getPoints()).getBounds().getMaxX();
                // double minX = PageXmlUtils.buildPolygon(table.getCoords().getPoints()).getBounds().getMinX();
                // int tablesize = (int) (maxX - minX);
                List<List<TrpTableCellType>> allRowCells = new ArrayList<List<TrpTableCellType>>();
                for (int k = 0; k < rows; k++) {
                    allRowCells.add(table.getRowCells(k));
                }
                List<HashMap<Integer, TrpTableCellType>> allRows = new ArrayList<HashMap<Integer, TrpTableCellType>>();
                HashMap<Integer, TrpTableCellType> nextRowMap = new HashMap<Integer, TrpTableCellType>();
                for (List<TrpTableCellType> rowCells : allRowCells) {
                    HashMap<Integer, TrpTableCellType> currRowMap = new HashMap<Integer, TrpTableCellType>();
                    /*
		            	 * fill up all cells which are not set in TRP (needed for vertical cell merge)
		            	 * the nextRowMap contains already all cells which span vertically with the cells above - means they got merged 
		            	 * in the table but have to be considered here 
		            	 */
                    currRowMap.putAll(nextRowMap);
                    nextRowMap.clear();
                    for (TrpTableCellType cell : rowCells) {
                        // logger.debug("table cell text " + cell.getUnicodeTextFromLines());
                        currRowMap.put(cell.getCol(), cell);
                        // only one row or col span is considered -> FIXME: do it for all spans, but may happens never?
                        if (cell.getRowSpan() > 1) {
                            nextRowMap.put(cell.getCol(), null);
                        }
                        if (cell.getColSpan() > 1) {
                            currRowMap.put(cell.getCol() + 1, null);
                        }
                    }
                    allRows.add(currRowMap);
                }
                createTable(rows, cols, allRows, tableId);
            }
            logger.debug("writing xlsx for page " + (i + 1) + "/" + doc.getNPages());
            ++c;
            if (monitor != null) {
                monitor.worked(c);
            }
        }
    }
    /*
		 * auto size the columns
		 */
    for (int i = 0; i < wb.getNumberOfSheets(); i++) {
        int numberOfCells = 0;
        Iterator rowIterator = wb.getSheetAt(i).rowIterator();
        /**
         * Escape the header row *
         */
        if (rowIterator.hasNext()) {
            Row headerRow = (Row) rowIterator.next();
            // get the number of cells in the header row
            numberOfCells = headerRow.getPhysicalNumberOfCells();
            for (int j = 0; j < numberOfCells; j++) {
                wb.getSheetAt(i).autoSizeColumn(j, true);
            }
        }
    }
    FileOutputStream fOut;
    try {
        // means no tables at all
        if (wb.getNumberOfSheets() == 0) {
            throw new NoTablesException("Sorry - No tables available for export");
        }
        fOut = new FileOutputStream(exportPath);
        wb.write(fOut);
        fOut.close();
    } catch (IOException e) {
        if (!(e instanceof NoTablesException)) {
            logger.error(e.getMessage(), e);
        }
        throw e;
    }
    logger.info("wrote xlsx to: " + exportPath);
}