SxcDocumentDeserializer.java

Go to the documentation of this file.

/*
 * This file is part of the LibreOffice project.
 *
 * This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this
 * file, You can obtain one at http://mozilla.org/MPL/2.0/.
 *
 * This file incorporates work covered by the following license notice:
 *
 *   Licensed to the Apache Software Foundation (ASF) under one or more
 *   contributor license agreements. See the NOTICE file distributed
 *   with this work for additional information regarding copyright
 *   ownership. The ASF licenses this file to you under the Apache
 *   License, Version 2.0 (the "License"); you may not use this file
 *   except in compliance with the License. You may obtain a copy of
 *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
 */

package org.openoffice.xmerge.converter.xml.sxc;

import java.io.IOException;
import java.util.Iterator;

import org.openoffice.xmerge.ConvertData;
import org.openoffice.xmerge.ConvertException;
import org.openoffice.xmerge.Document;
import org.openoffice.xmerge.DocumentDeserializer;
import org.openoffice.xmerge.converter.xml.OfficeConstants;
import org.openoffice.xmerge.converter.xml.Style;
import org.openoffice.xmerge.converter.xml.StyleCatalog;
import org.openoffice.xmerge.util.Debug;
import org.w3c.dom.Element;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;

public abstract class SxcDocumentDeserializer implements OfficeConstants,
    DocumentDeserializer {

    private SpreadsheetDecoder decoder = null;

    private org.w3c.dom.Document settings = null;

    private org.w3c.dom.Document doc = null;

    private final ConvertData cd ;

    private StyleCatalog styleCat = null;

    private int textStyles = 1;
    private int colStyles = 1;
    private int rowStyles = 1;

    public SxcDocumentDeserializer(ConvertData cd) {
        this.cd = cd;
    }

    public abstract SpreadsheetDecoder createDecoder(String workbook, String[] worksheetNames, String password)
        throws IOException;

    protected abstract String getWorkbookName(ConvertData cd) throws IOException;

    protected abstract String[] getWorksheetNames(ConvertData cd) throws IOException;

    public Document deserialize() throws ConvertException,
        IOException {

        // Get the name of the WorkBook from the ConvertData.
        String[] worksheetNames = getWorksheetNames(cd);
        String workbookName = getWorkbookName(cd);

        // Create a document
        SxcDocument sxcDoc = new SxcDocument(workbookName);
        sxcDoc.initContentDOM();
        sxcDoc.initSettingsDOM();

        // Default to an initial 5 entries in the catalog.
        styleCat = new StyleCatalog(5);

        doc = sxcDoc.getContentDOM();
        settings = sxcDoc.getSettingsDOM();
        initFontTable();
        // Little fact for the curious reader: workbookName should
        // be the name of the StarCalc file minus the file extension suffix.

        // Create a Decoder to decode the DeviceContent to a spreadsheet document
        // ToDo - we aren't using a password in StarCalc, so we can
        // use any value for password here.  If StarCalc XML supports
        // passwords in the future, we should try to get the correct
        // password value here.

        decoder = createDecoder(workbookName, worksheetNames, "password");

        Debug.log(Debug.TRACE, "<?xml version=\"1.0\" encoding=\"UTF-8\"?>");
        Debug.log(Debug.TRACE, "<DEBUGLOG>");

        decoder.addDeviceContent(cd);
        decode();

        Debug.log(Debug.TRACE, "</DEBUGLOG>");

        return sxcDoc;
    }

    private void initFontTable() {

        String fontTable[]= new String[] { "Tahoma", "Tahoma", "swiss", "variable",
                                           "Courier New", "&apos;Courier New&apos;",
                                           "modern", "fixed" };
        //  Traverse to the office:body element.
        //  There should only be one.
        NodeList list = doc.getElementsByTagName(TAG_OFFICE_FONT_DECLS);
        Node root = list.item(0);

        for(int i=0;i<fontTable.length;) {

            // Create an element node for the table
            Element tableElement = doc.createElement(TAG_STYLE_FONT_DECL);

            tableElement.setAttribute(ATTRIBUTE_STYLE_NAME, fontTable[i++]);
               tableElement.setAttribute(ATTRIBUTE_FO_FONT_FAMILY, fontTable[i++]);
               tableElement.setAttribute(ATTRIBUTE_FO_FONT_FAMILY_GENERIC, fontTable[i++]);
            tableElement.setAttribute(ATTRIBUTE_STYLE_FONT_PITCH, fontTable[i++]);

               root.appendChild(tableElement);
        }

    }

    protected void decode() throws IOException {

        // Get number of worksheets
        int numSheets = decoder.getNumberOfSheets();
        // #i33702# - check for an Empty InputStream.
        if(numSheets == 0)
        {
            System.err.println("Error decoding invalid Input stream");
            return;
        }

        //  Traverse to the office:body element.
        //  There should only be one.
        NodeList list = doc.getElementsByTagName(TAG_OFFICE_BODY);
        Node node = list.item(0);

        for (int i = 0; i < numSheets; i++) {

            // Set the decoder to the correct worksheet
            decoder.setWorksheet(i);

            int len = list.getLength();

            if (len > 0) {

                // Process the spreadsheet
                processTable(node);
            }
        }

        // Add the Defined Name table if there is one
        Iterator<NameDefinition> nameDefinitionTable = decoder.getNameDefinitions();
        if(nameDefinitionTable.hasNext()) {
            processNameDefinition(node, nameDefinitionTable);
        }

        // add settings
        NodeList settingsList = settings.getElementsByTagName(TAG_OFFICE_SETTINGS);
        Node settingsNode = settingsList.item(0);
        processSettings(settingsNode);

    }

    protected void processSettings(Node root) {

        Element configItemSetEntry      = settings.createElement(TAG_CONFIG_ITEM_SET);
        configItemSetEntry.setAttribute(ATTRIBUTE_CONFIG_NAME, "view-settings");
        Element configItemMapIndexed    = settings.createElement(TAG_CONFIG_ITEM_MAP_INDEXED);
        configItemMapIndexed.setAttribute(ATTRIBUTE_CONFIG_NAME, "Views");
        Element configItemMapEntry      = settings.createElement(TAG_CONFIG_ITEM_MAP_ENTRY);
        BookSettings bs = decoder.getSettings();
        bs.writeNode(settings, configItemMapEntry);

        configItemMapIndexed.appendChild(configItemMapEntry);
        configItemSetEntry.appendChild(configItemMapIndexed);
        root.appendChild(configItemSetEntry);
    }

    protected void processNameDefinition(Node root, Iterator<NameDefinition> eNameDefinitions) throws IOException {

        Debug.log(Debug.TRACE, "<NAMED-EXPRESSIONS>");

        Element namedExpressionsElement = doc.createElement(TAG_NAMED_EXPRESSIONS);

        while(eNameDefinitions.hasNext()) {

            NameDefinition tableEntry = eNameDefinitions.next();
            tableEntry.writeNode(doc, namedExpressionsElement);
        }

        root.appendChild(namedExpressionsElement);

        Debug.log(Debug.TRACE, "</NAMED-EXPRESSIONS>");
    }

    protected void processTable(Node root) throws IOException {

        Debug.log(Debug.TRACE, "<TABLE>");

        // Create an element node for the table
        Element tableElement = doc.createElement(TAG_TABLE);

        // Get the sheet name
        String sheetName = decoder.getSheetName();

        // Set the table name attribute
        tableElement.setAttribute(ATTRIBUTE_TABLE_NAME, sheetName);

        // ToDo - style currently hardcoded - get real value
        // Set table style-name attribute
        tableElement.setAttribute(ATTRIBUTE_TABLE_STYLE_NAME, "Default");

        // Append the table element to the root node
        root.appendChild(tableElement);

        Debug.log(Debug.TRACE, "<SheetName>" + sheetName + "</SheetName>");

        // Add the various different table-columns
        processColumns(tableElement);

        // Get each cell and add to doc
        processCells(tableElement);

        Debug.log(Debug.TRACE, "</TABLE>");
    }

    protected void processColumns(Node root) throws IOException {

        for(Iterator<ColumnRowInfo> e = decoder.getColumnRowInfos();e.hasNext();) {

            ColumnRowInfo ci = e.next();
            if(ci.isColumn()) {
                ColumnStyle cStyle = new ColumnStyle("Default",SxcConstants.COLUMN_STYLE_FAMILY,
                            SxcConstants.DEFAULT_STYLE, ci.getSize(), null);

                Style result[] = styleCat.getMatching(cStyle);
                String styleName;
                if(result.length==0) {

                        cStyle.setName("co" + colStyles++);
                        styleName = cStyle.getName();
                        Debug.log(Debug.TRACE,"No existing style found, adding " + styleName);
                        styleCat.add(cStyle);
                } else {
                        ColumnStyle existingStyle = (ColumnStyle) result[0];
                        styleName = existingStyle.getName();
                        Debug.log(Debug.TRACE,"Existing style found : " + styleName);
                }

                // Create an element node for the new row
                Element colElement = doc.createElement(TAG_TABLE_COLUMN);
                colElement.setAttribute(ATTRIBUTE_TABLE_STYLE_NAME, styleName);
                if(ci.getRepeated()!=1) {
                    String repeatStr = String.valueOf(ci.getRepeated());
                    colElement.setAttribute(ATTRIBUTE_TABLE_NUM_COLUMNS_REPEATED, repeatStr);
                }
                root.appendChild(colElement);
            }
        }
    }

    protected void processCells(Node root) throws IOException {

        // The current row element
        Element rowElement = null;

        // The current cell element
        Element cellElement = null;

        // The row number - we may not have any rows (empty sheet)
        // so set to zero.
        int row = 0;

        // The column number - This is the expected column number of
        // the next cell we are reading.
        int col = 1;

        // The number of columns in the spreadsheet
        int lastColumn = decoder.getNumberOfColumns();

        Node autoStylesNode = null;

        // Loop over all cells in the spreadsheet
        while (decoder.goToNextCell()) {

            // Get the row number
            int newRow = decoder.getRowNumber();

            // Is the cell in a new row, or part of the current row?
            if (newRow != row) {

                // Make sure that all the cells in the previous row
                // have been entered.
                if (col <= lastColumn && rowElement != null) {
                    int numSkippedCells = lastColumn - col + 1;
                    addEmptyCells(numSkippedCells, rowElement);
                }

                // log an end row - if we already have a row
                if (row != 0) {
                    Debug.log(Debug.TRACE, "</tr>");
                }

                // How far is the new row from the last row?
                int deltaRows = newRow - row;

                // Check if we have skipped any rows
                if (deltaRows > 1) {
                    // Add in empty rows
                    addEmptyRows(deltaRows-1, root, lastColumn);
                }

                // Re-initialize column (since we are in a new row)
                col = 1;

                // Create an element node for the new row
                rowElement = doc.createElement(TAG_TABLE_ROW);

                for(Iterator<ColumnRowInfo> e = decoder.getColumnRowInfos();e.hasNext();) {
                    ColumnRowInfo cri = e.next();
                    if(cri.isRow() && cri.getRepeated()==newRow-1) {
                        // We have the correct Row BIFFRecord for this row
                        RowStyle rStyle = new RowStyle("Default",SxcConstants.ROW_STYLE_FAMILY,
                                    SxcConstants.DEFAULT_STYLE, cri.getSize(), null);

                        Style result[] = styleCat.getMatching(rStyle);
                        String styleName;
                        if(result.length==0) {

                                rStyle.setName("ro" + rowStyles++);
                                styleName = rStyle.getName();
                                Debug.log(Debug.TRACE,"No existing style found, adding " + styleName);
                                styleCat.add(rStyle);
                        } else {
                                RowStyle existingStyle = (RowStyle) result[0];
                                styleName = existingStyle.getName();
                                Debug.log(Debug.TRACE,"Existing style found : " + styleName);
                        }
                        rowElement.setAttribute(ATTRIBUTE_TABLE_STYLE_NAME, styleName);
                        // For now we will not use the repeat column attribute
                    }
                }

                // Append the row element to the root node
                root.appendChild(rowElement);

                // Update row number
                row = newRow;

                Debug.log(Debug.TRACE, "<tr>");
            }

            if (rowElement == null) {
                //utterly busted
                break;
            }

            // Get the column number of the current cell
            int newCol = decoder.getColNumber();

            // Check to see if some columns were skipped
            if (newCol != col) {

                // How many columns have we skipped?
                int numColsSkipped = newCol - col;

                addEmptyCells(numColsSkipped, rowElement);

                // Update the column number to account for the
                // skipped cells
                col = newCol;
            }

            // Lets start dealing with the cell data
            Debug.log(Debug.TRACE, "<td>");

            // Get the cell's contents
            String cellContents = decoder.getCellContents();

            // Get the type of the data in the cell
            String cellType = decoder.getCellDataType();

            // Get the cell format
            Format fmt = decoder.getCellFormat();

            // Create an element node for the cell
            cellElement = doc.createElement(TAG_TABLE_CELL);

            Node bodyNode = doc.getElementsByTagName(TAG_OFFICE_BODY).item(0);

            // Not every document has an automatic style tag
            autoStylesNode = doc.getElementsByTagName(
                                        TAG_OFFICE_AUTOMATIC_STYLES).item(0);

            if (autoStylesNode == null) {
                autoStylesNode = doc.createElement(TAG_OFFICE_AUTOMATIC_STYLES);
                doc.insertBefore(autoStylesNode, bodyNode);
            }

            CellStyle tStyle = new
            CellStyle(  "Default",SxcConstants.TABLE_CELL_STYLE_FAMILY,
                        SxcConstants.DEFAULT_STYLE, fmt, null);
            String styleName;
            Style result[] = styleCat.getMatching(tStyle);
            if(result.length==0) {

                    tStyle.setName("ce" + textStyles++);
                    styleName = tStyle.getName();
                    Debug.log(Debug.TRACE,"No existing style found, adding " + styleName);
                    styleCat.add(tStyle);
            } else {
                    CellStyle existingStyle = (CellStyle) result[0];
                    styleName = existingStyle.getName();
                    Debug.log(Debug.TRACE,"Existing style found : " + styleName);
            }

            cellElement.setAttribute(ATTRIBUTE_TABLE_STYLE_NAME, styleName);

            // Store the cell data into the appropriate attributes
            processCellData(cellElement, cellType, cellContents);

            // Append the cell element to the row node
            rowElement.appendChild(cellElement);

            // Append the cellContents as a text node
            Element textElement = doc.createElement(TAG_PARAGRAPH);
            cellElement.appendChild(textElement);
            textElement.appendChild(doc.createTextNode(cellContents));

            Debug.log(Debug.TRACE, cellContents);
            Debug.log(Debug.TRACE, "</td>");

            // Increment to the column number of the next expected cell
            col++;
        }

        // Make sure that the last row is padded correctly
        if (col <= lastColumn && rowElement != null) {
            int numSkippedCells = lastColumn - col + 1;
            addEmptyCells(numSkippedCells, rowElement);
        }

        // Now write the style catalog to the document
        if(autoStylesNode!=null) {
            Debug.log(Debug.TRACE, "Well the autostyle node was found!!!");
            NodeList nl = styleCat.writeNode(doc, "dummy").getChildNodes();
            int nlLen = nl.getLength();     // nl.item reduces the length
            for (int i = 0; i < nlLen; i++) {
                autoStylesNode.appendChild(nl.item(0));
            }
        }

        if (row != 0) {
            // The sheet does have rows, so write out a /tr
            Debug.log(Debug.TRACE, "</tr>");
        }
    }

    protected void addEmptyRows(int numEmptyRows, Node root, int numEmptyCells) {

        // Create an element node for the row
        Element rowElement = doc.createElement(TAG_TABLE_ROW);

        // ToDo - style currently hardcoded - get real value
        // Set row style-name attribute
        rowElement.setAttribute(ATTRIBUTE_TABLE_STYLE_NAME, "Default");

        // Set cell number-rows-repeated attribute
        rowElement.setAttribute(ATTRIBUTE_TABLE_NUM_ROWS_REPEATED,
                                Integer.toString(numEmptyRows));

        // Append the row element to the root node
        root.appendChild(rowElement);

        // Open Office requires the empty row to have an empty cell (or cells)
        addEmptyCells(numEmptyCells, rowElement);

        // Write empty rows to the log
        for (int i = 0;  i < numEmptyRows; i++) {
            Debug.log(Debug.TRACE, "<tr />");
        }

    }

    protected void addEmptyCells(int numColsSkipped, Node row) {

        // Create an empty cellElement
        Element cellElement = doc.createElement(TAG_TABLE_CELL);

        // ToDo - style currently hardcoded - get real value
        // Set cell style-name attribute
        cellElement.setAttribute(ATTRIBUTE_TABLE_STYLE_NAME, "Default");

        // If we skipped more than 1 cell, we must set the
        // appropriate attribute
        if (numColsSkipped > 1) {

            // Set cell number-columns-repeated attribute
            cellElement.setAttribute(ATTRIBUTE_TABLE_NUM_COLUMNS_REPEATED,
                                     Integer.toString(numColsSkipped));
        }

        // Append the empty cell element to the row node
        row.appendChild(cellElement);

        // Write empty cells to the log
        for (int i = 0;  i < numColsSkipped; i++) {
            Debug.log(Debug.TRACE, "<td />");
        }
    }

    protected void processCellData(Element cellElement, String type,
                                 String contents) {

        // Set cell value-type attribute
        cellElement.setAttribute(ATTRIBUTE_TABLE_VALUE_TYPE, type);

        // Does the cell contain a formula?
        if (contents.startsWith("=")) {

            cellElement.setAttribute(ATTRIBUTE_TABLE_FORMULA, contents);

            cellElement.setAttribute(ATTRIBUTE_TABLE_VALUE, decoder.getCellValue());
            // String data does not require any additional attributes
        } else if (!type.equals(CELLTYPE_STRING)) {

            if (type.equals(CELLTYPE_TIME)) {

                // Data returned in StarOffice XML format, so store in
                // attribute
                cellElement.setAttribute(ATTRIBUTE_TABLE_TIME_VALUE,
                                         contents);

            } else if (type.equals(CELLTYPE_DATE)) {

                // Data returned in StarOffice XML format, so store in
                // attribute
                cellElement.setAttribute(ATTRIBUTE_TABLE_DATE_VALUE,
                                         contents);

            } else if (type.equals(CELLTYPE_BOOLEAN)) {

                // StarOffice XML format requires stored boolean value
                // to be in lower case
                cellElement.setAttribute(ATTRIBUTE_TABLE_BOOLEAN_VALUE,
                                         contents.toLowerCase());

            } else if (type.equals(CELLTYPE_CURRENCY)) {
                // TODO - StarOffice XML format requires a correct style to
                // display currencies correctly.  Need to implement styles.
                // TODO - USD is for US currencies.  Need to pick up
                // the correct currency location from the source file.
                cellElement.setAttribute(ATTRIBUTE_TABLE_CURRENCY, "USD");

                // Data comes stripped of currency symbols
                cellElement.setAttribute(ATTRIBUTE_TABLE_VALUE, contents);

            } else if (type.equals(CELLTYPE_PERCENT)) {
                // Data comes stripped of percent signs
                cellElement.setAttribute(ATTRIBUTE_TABLE_VALUE, contents);

            } else {
                // Remaining data types use table-value attribute

                cellElement.setAttribute(ATTRIBUTE_TABLE_VALUE, contents);
            }
        }
    }
}