Index: src/main/java/org/apache/tika/parser/microsoft/ExcelParser.java =================================================================== --- src/main/java/org/apache/tika/parser/microsoft/ExcelParser.java (revision 596536) +++ src/main/java/org/apache/tika/parser/microsoft/ExcelParser.java (working copy) @@ -33,31 +33,29 @@ return "application/vnd.ms-excel"; } - protected String extractText(POIFSFileSystem filesystem) throws IOException{ - StringBuilder builder = new StringBuilder(); + protected void extractText(POIFSFileSystem filesystem, Appendable builder) throws IOException{ extractText(new HSSFWorkbook(filesystem), builder); - return builder.toString(); } - private void extractText(HSSFWorkbook book, StringBuilder builder) { + private void extractText(HSSFWorkbook book, Appendable builder) throws IOException { for (int i = 0; book != null && i < book.getNumberOfSheets(); i++) { extractText(book.getSheetAt(i), builder); } } - private void extractText(HSSFSheet sheet, StringBuilder builder) { + private void extractText(HSSFSheet sheet, Appendable builder) throws IOException { for (int i = 0; sheet != null && i <= sheet.getLastRowNum(); i++) { extractText(sheet.getRow(i), builder); } } - private void extractText(HSSFRow row, StringBuilder builder) { + private void extractText(HSSFRow row, Appendable builder) throws IOException { for (short i = 0; row != null && i < row.getLastCellNum(); i++) { extractText(row.getCell(i), builder); } } - private void extractText(HSSFCell cell, StringBuilder builder) { + private void extractText(HSSFCell cell, Appendable builder) throws IOException { if (cell != null) { switch (cell.getCellType()) { case HSSFCell.CELL_TYPE_STRING: @@ -73,14 +71,11 @@ } } - private void addText(String text, StringBuilder builder) { + private void addText(String text, Appendable builder) throws IOException { if (text != null) { text = text.trim(); if (text.length() > 0) { - if (builder.length() > 0) { - builder.append(' '); - } - builder.append(text); + builder.append(text).append(' '); } } } Index: src/main/java/org/apache/tika/parser/microsoft/OfficeParser.java =================================================================== --- src/main/java/org/apache/tika/parser/microsoft/OfficeParser.java (revision 596536) +++ src/main/java/org/apache/tika/parser/microsoft/OfficeParser.java (working copy) @@ -56,7 +56,9 @@ XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata); xhtml.startDocument(); - xhtml.element("p", extractText(filesystem)); + xhtml.startElement("p"); + extractText(filesystem, xhtml); + xhtml.endElement("p"); xhtml.endDocument(); } @@ -70,7 +72,7 @@ /** * Extracts the text content from a Microsoft document input stream. */ - protected abstract String extractText(POIFSFileSystem filesystem) + protected abstract void extractText(POIFSFileSystem filesystem, Appendable appendable) throws IOException, TikaException; private void getMetadata( @@ -177,4 +179,4 @@ } } -} \ No newline at end of file +} Index: src/main/java/org/apache/tika/parser/microsoft/PowerPointExtractor.java =================================================================== --- src/main/java/org/apache/tika/parser/microsoft/PowerPointExtractor.java (revision 596536) +++ src/main/java/org/apache/tika/parser/microsoft/PowerPointExtractor.java (working copy) @@ -32,12 +32,12 @@ static Logger LOG = Logger.getRootLogger(); /** Buffer holding the content of the file */ - private final StringBuilder builder; + private final Appendable builder; /** * Constructs Listener to get content of PowerPoint file. */ - public PowerPointExtractor(StringBuilder builder) { + public PowerPointExtractor(Appendable builder) { this.builder = builder; } Index: src/main/java/org/apache/tika/parser/microsoft/PowerPointParser.java =================================================================== --- src/main/java/org/apache/tika/parser/microsoft/PowerPointParser.java (revision 596536) +++ src/main/java/org/apache/tika/parser/microsoft/PowerPointParser.java (working copy) @@ -35,8 +35,7 @@ return "application/vnd.ms-powerpoint"; } - protected String extractText(POIFSFileSystem filesystem) throws IOException { - StringBuilder builder = new StringBuilder(); + protected void extractText(POIFSFileSystem filesystem, Appendable builder) throws IOException { InputStream stream = filesystem.createDocumentInputStream(POWERPOINT); try { @@ -44,8 +43,6 @@ } finally { stream.close(); } - - return builder.toString(); } } Index: src/main/java/org/apache/tika/parser/microsoft/Word6Extractor.java =================================================================== --- src/main/java/org/apache/tika/parser/microsoft/Word6Extractor.java (revision 596536) +++ src/main/java/org/apache/tika/parser/microsoft/Word6Extractor.java (working copy) @@ -33,8 +33,11 @@ class Word6Extractor { - public Word6Extractor() + private final Appendable appendable; + + public Word6Extractor(Appendable appendable) { + this.appendable = appendable; } /** @@ -45,7 +48,7 @@ * @return The text from the document * @throws Exception If there are any unexpected exceptions. */ - public String extractText(byte[] mainStream) throws IOException { + public void extractText(byte[] mainStream) throws IOException { int fcMin = LittleEndian.getInt(mainStream, 0x18); int fcMax = LittleEndian.getInt(mainStream, 0x1C); @@ -58,7 +61,7 @@ List textRuns = chpTable.getTextRuns(); // iterate through the - WordTextBuffer finalTextBuf = new WordTextBuffer(); + WordTextBuffer finalTextBuf = new WordTextBuffer(appendable); Iterator runsIt = textRuns.iterator(); while(runsIt.hasNext()) { @@ -76,8 +79,6 @@ } } } - - return finalTextBuf.toString(); } /** Index: src/main/java/org/apache/tika/parser/microsoft/WordParser.java =================================================================== --- src/main/java/org/apache/tika/parser/microsoft/WordParser.java (revision 596536) +++ src/main/java/org/apache/tika/parser/microsoft/WordParser.java (working copy) @@ -47,7 +47,7 @@ * * @param in The InputStream representing the Word file. */ - public String extractText(POIFSFileSystem fsys) + public void extractText(POIFSFileSystem fsys, Appendable appendable) throws IOException, TikaException { // load our POIFS document streams. DocumentEntry headerProps = @@ -74,8 +74,8 @@ case 103: case 104: // this is a Word 6.0 doc send it to the extractor for that version. - Word6Extractor oldExtractor = new Word6Extractor(); - return oldExtractor.extractText(header); + Word6Extractor oldExtractor = new Word6Extractor(appendable); + oldExtractor.extractText(header); } //get the location of the piece table @@ -123,7 +123,7 @@ int currentTextStart = currentPiece.getStart(); int currentTextEnd = currentPiece.getEnd(); - WordTextBuffer finalTextBuf = new WordTextBuffer(); + WordTextBuffer finalTextBuf = new WordTextBuffer(appendable); // iterate through all text runs extract the text only if they haven't been // deleted @@ -157,7 +157,7 @@ runStart = currentTextStart; currentTextEnd = currentPiece.getEnd (); } else { - return finalTextBuf.toString(); + return; } } String str = currentPiece.substring(0, runEnd - currentTextStart); @@ -172,7 +172,6 @@ finalTextBuf.append(str); } } - return finalTextBuf.toString(); } /** Index: src/main/java/org/apache/tika/parser/microsoft/WordTextBuffer.java =================================================================== --- src/main/java/org/apache/tika/parser/microsoft/WordTextBuffer.java (revision 596536) +++ src/main/java/org/apache/tika/parser/microsoft/WordTextBuffer.java (working copy) @@ -22,16 +22,16 @@ */ public class WordTextBuffer { - StringBuffer _buf; + Appendable _buf; boolean _hold; - public WordTextBuffer() + public WordTextBuffer(Appendable appendable) { - _buf = new StringBuffer(); + _buf = appendable; _hold = false; } - public void append(String text) + public void append(String text) throws java.io.IOException { char[] letters = text.toCharArray(); for (int x = 0; x < letters.length; x++) @@ -57,9 +57,4 @@ } } - public String toString() - { - return _buf.toString(); - } - } Index: src/main/java/org/apache/tika/parser/opendocument/OpenOfficeParser.java =================================================================== --- src/main/java/org/apache/tika/parser/opendocument/OpenOfficeParser.java (revision 596536) +++ src/main/java/org/apache/tika/parser/opendocument/OpenOfficeParser.java (working copy) @@ -101,7 +101,9 @@ XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata); xhtml.startDocument(); - xhtml.element("p", xp.concatOccurrence(xmlDoc, "//*", " ")); + xhtml.startElement("p"); + xp.concatOccurrence(xmlDoc, "//*", " ", xhtml); + xhtml.endElement("p"); xhtml.endDocument(); } Index: src/main/java/org/apache/tika/parser/xml/XMLParser.java =================================================================== --- src/main/java/org/apache/tika/parser/xml/XMLParser.java (revision 596536) +++ src/main/java/org/apache/tika/parser/xml/XMLParser.java (working copy) @@ -70,13 +70,14 @@ XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata); xhtml.startDocument(); - xhtml.element("p", concatOccurrence(xmlDoc, "//*", " ")); + xhtml.startElement("p"); + concatOccurrence(xmlDoc, "//*", " ", xhtml); + xhtml.endElement("p"); xhtml.endDocument(); } - public String concatOccurrence(Object xmlDoc, String xpath, String concatSep) { + public void concatOccurrence(Object xmlDoc, String xpath, String concatSep, Appendable chaineConcat) throws IOException { - StringBuilder chaineConcat = new StringBuilder(); try { JDOMXPath xp = new JDOMXPath(xpath); List ls = xp.selectNodes(xmlDoc); @@ -108,7 +109,7 @@ if (StringUtils.isNotEmpty(text)) { chaineConcat.append(text); if (ls.size() == 1) { - return chaineConcat.toString().trim(); + return; } else { if (ls.size() != j) { chaineConcat.append(' ') @@ -121,7 +122,6 @@ } catch (JaxenException j) { logger.error(j.getMessage()); } - return chaineConcat.toString().trim(); } public List getAllDocumentNs(org.jdom.Document doc) { Index: src/main/java/org/apache/tika/sax/ContentHandlerDecorator.java =================================================================== --- src/main/java/org/apache/tika/sax/ContentHandlerDecorator.java (revision 596536) +++ src/main/java/org/apache/tika/sax/ContentHandlerDecorator.java (working copy) @@ -16,6 +16,8 @@ */ package org.apache.tika.sax; +import java.io.IOException; + import org.xml.sax.Attributes; import org.xml.sax.ContentHandler; import org.xml.sax.Locator; @@ -27,7 +29,7 @@ * instance. Subclasses can provide extra decoration by overriding one or more * of the SAX event methods. */ -public class ContentHandlerDecorator implements ContentHandler { +public class ContentHandlerDecorator implements ContentHandler, Appendable { /** * Decorated SAX event handler. @@ -35,12 +37,18 @@ private final ContentHandler handler; /** + * Decorated SAX event handler. + */ + private final Appendable appendable; + + /** * Creates a decorator for the given SAX event handler. * * @param handler SAX event handler to be decorated */ public ContentHandlerDecorator(ContentHandler handler) { this.handler = handler; + this.appendable = (handler instanceof Appendable ? (Appendable)handler : null); } public void startPrefixMapping(String prefix, String uri) @@ -93,4 +101,36 @@ handler.skippedEntity(name); } + public Appendable append(char c) throws IOException { + if (appendable != null) { + appendable.append(c); + } else { + append(Character.toString(c)); + } + return this; + } + + public Appendable append(CharSequence charSequence, int start, int end) throws IOException { + if (charSequence != null) { + if (appendable != null) { + appendable.append(charSequence, start, end); + } else { + char[] chars = charSequence.toString().toCharArray(); + try { + handler.characters(chars, start, (end - start)); + } catch (SAXException e) { + throw new IOException(e.toString()); + } + } + } + return this; + } + + public Appendable append(CharSequence charSequence) throws IOException { + if (charSequence != null) { + append(charSequence, 0, charSequence.length()); + } + return this; + } + } Index: src/main/java/org/apache/tika/sax/WriteOutContentHandler.java =================================================================== --- src/main/java/org/apache/tika/sax/WriteOutContentHandler.java (revision 596536) +++ src/main/java/org/apache/tika/sax/WriteOutContentHandler.java (working copy) @@ -26,7 +26,7 @@ * SAX event handler that writes all character content out to * a {@link Writer} character stream. */ -public class WriteOutContentHandler extends DefaultHandler { +public class WriteOutContentHandler extends DefaultHandler implements Appendable { /** * The character stream. @@ -45,5 +45,19 @@ throw new SAXException("Error writing out character content", e); } } + public Appendable append(char c) throws IOException { + writer.append(c); + return this; + } + public Appendable append(CharSequence charSequence, int start, int end) throws IOException { + writer.append(charSequence, start, end); + return this; + } + + public Appendable append(CharSequence charSequence) throws IOException { + writer.append(charSequence); + return this; + } + }