import java.util.zip.GZIPInputStream;\r
import java.util.zip.GZIPOutputStream;\r
\r
+import org.apache.poi.xssf.model.SharedStringsTable;\r
+\r
/**\r
* Sheet writer that supports gzip compression of the temp files.\r
*/\r
public GZIPSheetDataWriter() throws IOException {\r
super();\r
}\r
+ \r
+ /**\r
+ * @param sharedStringsTable the shared strings table, or null if inline text is used\r
+ */\r
+ public GZIPSheetDataWriter(SharedStringsTable sharedStringsTable) throws IOException {\r
+ super(sharedStringsTable);\r
+ }\r
\r
/**\r
* @return temp file to write sheet data\r
import org.apache.poi.ss.formula.udf.UDFFinder;
import org.apache.poi.ss.usermodel.Row.MissingCellPolicy;
import org.apache.poi.ss.util.CellRangeAddress;
+import org.apache.poi.xssf.model.SharedStringsTable;
/**
* Streaming version of XSSFWorkbook implementing the "BigGridDemo" strategy.
*
- * @author Alex Geller, Four J's Development Tools
-*/
+ * SXSSFWorkbook defaults to using inline strings instead of a shared strings
+ * table. This is very efficient, since no document content needs to be kept in
+ * memory, but is also known to produce documents that are incompatible with
+ * some clients. With shared strings enabled all unique strings in the document
+ * has to be kept in memory. Depending on your document content this could use
+ * a lot more resources than with shared strings disabled.
+ *
+ * Carefully review your memory budget and compatibility needs before deciding
+ * whether to enable shared strings or not.
+ */
public class SXSSFWorkbook implements Workbook
{
/**
*/
private boolean _compressTmpFiles = false;
+ /**
+ * shared string table - a cache of strings in this workbook
+ */
+ private SharedStringsTable _sharedStringSource = null;
+
/**
* Construct a new workbook
*/
* @param compressTmpFiles whether to use gzip compression for temporary files
*/
public SXSSFWorkbook(XSSFWorkbook workbook, int rowAccessWindowSize, boolean compressTmpFiles){
+ this(workbook,rowAccessWindowSize, compressTmpFiles, false);
+ }
+
+ /**
+ * Constructs an workbook from an existing workbook.
+ * <p>
+ * When a new node is created via createRow() and the total number
+ * of unflushed records would exceed the specified value, then the
+ * row with the lowest index value is flushed and cannot be accessed
+ * via getRow() anymore.
+ * </p>
+ * <p>
+ * A value of -1 indicates unlimited access. In this case all
+ * records that have not been flushed by a call to flush() are available
+ * for random access.
+ * <p>
+ * <p></p>
+ * A value of 0 is not allowed because it would flush any newly created row
+ * without having a chance to specify any cells.
+ * </p>
+ *
+ * @param workbook the template workbook
+ * @param rowAccessWindowSize
+ * @param compressTmpFiles whether to use gzip compression for temporary files
+ * @param useSharedStringsTable whether to use a shared strings table
+ */
+ public SXSSFWorkbook(XSSFWorkbook workbook, int rowAccessWindowSize, boolean compressTmpFiles, boolean useSharedStringsTable){
setRandomAccessWindowSize(rowAccessWindowSize);
setCompressTempFiles(compressTmpFiles);
if (workbook == null)
{
_wb=new XSSFWorkbook();
+ if(useSharedStringsTable){
+ _sharedStringSource = _wb.getSharedStringSource();
+ }
}
else
{
_wb=workbook;
+ if(useSharedStringsTable){
+ _sharedStringSource = _wb.getSharedStringSource();
+ }
for ( int i = 0; i < _wb.getNumberOfSheets(); i++ )
{
XSSFSheet sheet = _wb.getSheetAt( i );
SheetDataWriter createSheetDataWriter() throws IOException {
if(_compressTmpFiles) {
- return new GZIPSheetDataWriter();
+ return new GZIPSheetDataWriter(_sharedStringSource);
} else {
- return new SheetDataWriter();
+ return new SheetDataWriter(_sharedStringSource);
}
}
import org.apache.poi.ss.usermodel.CellStyle;\r
import org.apache.poi.ss.usermodel.FormulaError;\r
import org.apache.poi.ss.util.CellReference;\r
+import org.apache.poi.xssf.model.SharedStringsTable;\r
+import org.apache.poi.xssf.usermodel.XSSFRichTextString;\r
+import org.openxmlformats.schemas.spreadsheetml.x2006.main.STCellType;\r
\r
/**\r
* Initially copied from BigGridDemo "SpreadsheetWriter".\r
private int _numberOfCellsOfLastFlushedRow; // meaningful only of _numberOfFlushedRows>0\r
private int _numberLastFlushedRow = -1; // meaningful only of _numberOfFlushedRows>0\r
\r
+ /**\r
+ * Table of strings shared across this workbook.\r
+ * If two cells contain the same string, then the cell value is the same index into SharedStringsTable\r
+ */\r
+ private SharedStringsTable _sharedStringSource;\r
+\r
public SheetDataWriter() throws IOException {\r
_fd = createTempFile();\r
_out = createWriter(_fd);\r
}\r
\r
+ public SheetDataWriter(SharedStringsTable sharedStringsTable) throws IOException{\r
+ this();\r
+ this._sharedStringSource = sharedStringsTable;\r
+ }\r
/**\r
* Create a temp file to write sheet data. \r
* By default, temp files are created in the default temporary-file directory\r
break;\r
}\r
case Cell.CELL_TYPE_STRING: {\r
- _out.write(" t=\"inlineStr\">");\r
- _out.write("<is><t");\r
- if(hasLeadingTrailingSpaces(cell.getStringCellValue())) {\r
- _out.write(" xml:space=\"preserve\"");\r
+ if (_sharedStringSource != null) {\r
+ XSSFRichTextString rt = new XSSFRichTextString(cell.getStringCellValue());\r
+ int sRef = _sharedStringSource.addEntry(rt.getCTRst());\r
+\r
+ _out.write(" t=\"" + STCellType.S.toString() + "\">");\r
+ _out.write("<v>");\r
+ _out.write(String.valueOf(sRef));\r
+ _out.write("</v>");\r
+ } else {\r
+ _out.write(" t=\"inlineStr\">");\r
+ _out.write("<is><t");\r
+ if (hasLeadingTrailingSpaces(cell.getStringCellValue())) {\r
+ _out.write(" xml:space=\"preserve\"");\r
+ }\r
+ _out.write(">");\r
+ outputQuotedString(cell.getStringCellValue());\r
+ _out.write("</t></is>");\r
}\r
- _out.write(">");\r
- outputQuotedString(cell.getStringCellValue());\r
- _out.write("</t></is>");\r
break;\r
}\r
case Cell.CELL_TYPE_NUMERIC: {\r
}\r
\r
//Taken from jdk1.3/src/javax/swing/text/html/HTMLWriter.java\r
- protected void outputQuotedString(String s) throws IOException {\r
+ protected void outputQuotedString(String s) throws IOException {\r
if (s == null || s.length() == 0) {\r
return;\r
}\r
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.OutputStream;
+import java.lang.reflect.Field;
import org.apache.poi.ss.usermodel.BaseTestWorkbook;
import org.apache.poi.ss.usermodel.Cell;
import org.apache.poi.ss.usermodel.WorkbookFactory;
import org.apache.poi.ss.util.CellReference;
import org.apache.poi.xssf.SXSSFITestDataProvider;
+import org.apache.poi.xssf.model.SharedStringsTable;
import org.apache.poi.xssf.usermodel.XSSFWorkbook;
public final class TestSXSSFWorkbook extends BaseTestWorkbook {
}
+ public void testUseSharedStringsTable() throws Exception {
+ SXSSFWorkbook wb = new SXSSFWorkbook(null, 10, false, true);
+
+ Field f = SXSSFWorkbook.class.getDeclaredField("_sharedStringSource");
+ f.setAccessible(true);
+ SharedStringsTable sss = (SharedStringsTable)f.get(wb);
+
+ assertNotNull(sss);
+
+ Row row = wb.createSheet("S1").createRow(0);
+
+ row.createCell(0).setCellValue("A");
+ row.createCell(1).setCellValue("B");
+ row.createCell(2).setCellValue("A");
+
+ XSSFWorkbook xssfWorkbook = (XSSFWorkbook) SXSSFITestDataProvider.instance.writeOutAndReadBack(wb);
+ sss = (SharedStringsTable)f.get(wb);
+ assertEquals(2, sss.getUniqueCount());
+ wb.dispose();
+
+ Sheet sheet1 = xssfWorkbook.getSheetAt(0);
+ assertEquals("S1", sheet1.getSheetName());
+ assertEquals(1, sheet1.getPhysicalNumberOfRows());
+ row = sheet1.getRow(0);
+ assertNotNull(row);
+ Cell cell = row.getCell(0);
+ assertNotNull(cell);
+ assertEquals("A", cell.getStringCellValue());
+ cell = row.getCell(1);
+ assertNotNull(cell);
+ assertEquals("B", cell.getStringCellValue());
+ cell = row.getCell(2);
+ assertNotNull(cell);
+ assertEquals("A", cell.getStringCellValue());
+ }
+
public void testAddToExistingWorkbook() {
XSSFWorkbook xssfWorkbook = new XSSFWorkbook();
xssfWorkbook.createSheet("S1");
projects / poi.git / commitdiff