public int serialize(int offset, byte[] data) {
throw new RecordFormatException("Cannot serialize a dummy record");
}
- protected final int getDataSize() {
+ public final int getRecordSize() {
throw new RecordFormatException("Cannot serialize a dummy record");
}
}
}
return getRecordSize();
}
- protected int getDataSize() {
+
+ public int getRecordSize() {
if (escherRecords.size() == 0 && rawData != null) {
return rawData.length;
}
package org.apache.poi.hssf.record;
+import org.apache.poi.util.HexDump;
import org.apache.poi.util.LittleEndian;
+import org.apache.poi.util.LittleEndianOutput;
/**
* Title: Continue Record(0x003C) - Helper class used primarily for SST Records <P>
* @author Andrew C. Oliver (acoliver at apache dot org)
* @author Csaba Nagy (ncsaba at yahoo dot com)
*/
-public final class ContinueRecord extends Record {
+public final class ContinueRecord extends StandardRecord {
public final static short sid = 0x003C;
private byte[] _data;
return _data.length;
}
- public int serialize(int offset, byte[] data) {
- return write(data, offset, null, _data);
+ public void serialize(LittleEndianOutput out) {
+ out.write(_data);
}
/**
* get the data for continuation
* @return byte array containing all of the continued data
*/
- public byte [] getData()
- {
+ public byte[] getData() {
return _data;
}
- public String toString()
- {
+ public String toString() {
StringBuffer buffer = new StringBuffer();
buffer.append("[CONTINUE RECORD]\n");
- buffer.append(" .id = ").append(Integer.toHexString(sid))
- .append("\n");
+ buffer.append(" .data = ").append(HexDump.toHex(_data)).append("\n");
buffer.append("[/CONTINUE RECORD]\n");
return buffer.toString();
}
- public short getSid()
- {
+ public short getSid() {
return sid;
}
- /**
- * Fill the fields. Only thing is, this record has no fields --
- *
- * @param in the RecordInputstream to read the record from
- */
- public ContinueRecord(RecordInputStream in)
- {
- _data = in.readRemainder();
+ public ContinueRecord(RecordInputStream in) {
+ _data = in.readRemainder();
}
public Object clone() {
- return new ContinueRecord(_data);
- }
-
- /**
- * Writes the full encoding of a Continue record without making an instance
- */
- public static int write(byte[] destBuf, int destOffset, Byte initialDataByte, byte[] srcData) {
- return write(destBuf, destOffset, initialDataByte, srcData, 0, srcData.length);
- }
- /**
- * @param initialDataByte (optional - often used for unicode flag).
- * If supplied, this will be written before srcData
- * @return the total number of bytes written
- */
- public static int write(byte[] destBuf, int destOffset, Byte initialDataByte, byte[] srcData, int srcOffset, int len) {
- int totalLen = len + (initialDataByte == null ? 0 : 1);
- LittleEndian.putUShort(destBuf, destOffset, sid);
- LittleEndian.putUShort(destBuf, destOffset + 2, totalLen);
- int pos = destOffset + 4;
- if (initialDataByte != null) {
- LittleEndian.putByte(destBuf, pos, initialDataByte.byteValue());
- pos += 1;
- }
- System.arraycopy(srcData, srcOffset, destBuf, pos, len);
- return 4 + totalLen;
+ return new ContinueRecord(_data);
}
}
public void processChildRecords() {
convertRawBytesToEscherRecords();
}
- protected int getDataSize() {
- // TODO - convert this to a RecordAggregate
- return grossSizeFromDataSize( getRawDataSize() ) - 4;
+
+ public int getRecordSize() {
+ // TODO - convert this to a RecordAggregate
+ return grossSizeFromDataSize(getRawDataSize());
}
private int getRawDataSize() {
*
* @author Glen Stampoultzis (glens at apache.org)
*/
-public class EscherAggregate extends AbstractEscherHolderRecord
-{
- public static final short sid = 9876;
+public final class EscherAggregate extends AbstractEscherHolderRecord {
+ public static final short sid = 9876; // not a real sid - dummy value
private static POILogger log = POILogFactory.getLogger(EscherAggregate.class);
public static final short ST_MIN = (short) 0;
return size;
}
- protected int getDataSize() {
- // TODO - convert this to RecordAggregate
+ public int getRecordSize() {
+ // TODO - convert this to RecordAggregate
convertUserModelToRecords();
List records = getEscherRecords();
int rawEscherSize = getEscherRecordSize( records );
Record r = (Record) iterator.next();
tailRecordSize += r.getRecordSize();
}
- return drawingRecordSize + objRecordSize + tailRecordSize - 4;
+ return drawingRecordSize + objRecordSize + tailRecordSize;
}
/**
public int serialize(int offset, byte [] data) {
throw new RecordFormatException("Label Records are supported READ ONLY...convert to LabelSST");
}
- protected int getDataSize() {
+ public int getRecordSize() {
throw new RecordFormatException("Label Records are supported READ ONLY...convert to LabelSST");
}
return sb.toString();
}
- protected int getDataSize() {
+ public int getRecordSize() {
if (_uninterpretedData != null) {
- return _uninterpretedData.length;
+ return _uninterpretedData.length + 4;
}
int size = 0;
for (int i=subrecords.size()-1; i>=0; i--) {
size++;
}
}
- return size;
+ return size + 4;
}
public int serialize(int offset, byte[] data) {
/**
* Title: Record
- * Description: All HSSF Records inherit from this class. It
- * populates the fields common to all records (id, size and data).
- * Subclasses should be sure to validate the id,
- * Company:
+ * Description: All HSSF Records inherit from this class.
* @author Andrew C. Oliver
* @author Marc Johnson (mjohnson at apache dot org)
* @author Jason Height (jheight at chariot dot net dot au)
*/
public abstract class Record extends RecordBase {
- /**
- * instantiates a blank record strictly for ID matching
- */
+ protected Record() {
+ // no fields to initialise
+ }
- protected Record()
- {
- }
+ /**
+ * called by the class that is responsible for writing this sucker.
+ * Subclasses should implement this so that their data is passed back in a
+ * byte array.
+ *
+ * @return byte array containing instance data
+ */
+ public final byte[] serialize() {
+ byte[] retval = new byte[ getRecordSize() ];
- /**
- * called by the class that is responsible for writing this sucker.
- * Subclasses should implement this so that their data is passed back in a
- * byte array.
- *
- * @return byte array containing instance data
- */
+ serialize(0, retval);
+ return retval;
+ }
- public final byte[] serialize() {
- byte[] retval = new byte[ getRecordSize() ];
+ /**
+ * get a string representation of the record (for biffview/debugging)
+ */
+ public String toString() {
+ return super.toString();
+ }
- serialize(0, retval);
- return retval;
- }
+ /**
+ * return the non static version of the id for this record.
+ */
- public final int getRecordSize() {
- return 4 + getDataSize();
- }
- /**
- * @return the size of the data portion of this record
- * (does not include initial 4 bytes for sid and size)
- */
- protected abstract int getDataSize();
-
- /**
- * get a string representation of the record (for biffview/debugging)
- */
- public String toString()
- {
- return super.toString();
- }
+ public abstract short getSid();
- /**
- * return the non static version of the id for this record.
- */
+ public Object clone() {
+ if (false) {
+ // TODO - implement clone in a more standardised way
+ try {
+ return super.clone();
+ } catch (CloneNotSupportedException e) {
+ throw new RuntimeException(e);
+ }
+ }
+ throw new RuntimeException("The class "+getClass().getName()+" needs to define a clone method");
+ }
- public abstract short getSid();
+ /**
+ * Clone the current record, via a call to serialize
+ * it, and another to create a new record from the
+ * bytes.
+ * May only be used for classes which don't have
+ * internal counts / ids in them. For those which
+ * do, a full model-aware cloning is needed, which
+ * allocates new ids / counts as needed.
+ */
+ public Record cloneViaReserialise() {
+ // Do it via a re-serialization
+ // It's a cheat, but it works...
+ byte[] b = serialize();
+ RecordInputStream rinp = new RecordInputStream(new ByteArrayInputStream(b));
+ rinp.nextRecord();
- public Object clone() {
- throw new RuntimeException("The class "+getClass().getName()+" needs to define a clone method");
- }
-
- /**
- * Clone the current record, via a call to serialise
- * it, and another to create a new record from the
- * bytes.
- * May only be used for classes which don't have
- * internal counts / ids in them. For those which
- * do, a full record-aware serialise is needed, which
- * allocates new ids / counts as needed.
- */
- public Record cloneViaReserialise()
- {
- // Do it via a re-serialise
- // It's a cheat, but it works...
- byte[] b = serialize();
- RecordInputStream rinp = new RecordInputStream(
- new ByteArrayInputStream(b)
- );
- rinp.nextRecord();
-
- Record[] r = RecordFactory.createRecord(rinp);
- if(r.length != 1) {
- throw new IllegalStateException("Re-serialised a record to clone it, but got " + r.length + " records back!");
- }
- return r[0];
- }
+ Record[] r = RecordFactory.createRecord(rinp);
+ if(r.length != 1) {
+ throw new IllegalStateException("Re-serialised a record to clone it, but got " + r.length + " records back!");
+ }
+ return r[0];
+ }
}
* @author Josh Micich\r
*/\r
public abstract class StandardRecord extends Record {\r
+ protected abstract int getDataSize();\r
+ public final int getRecordSize() {\r
+ return 4 + getDataSize();\r
+ }\r
@Override\r
public final int serialize(int offset, byte[] data) {\r
int dataSize = getDataSize();\r
\r
\r
/**\r
- * @return four less than the total length of the encoded record(s) \r
- * (in the case when no {@link ContinueRecord} is needed, this is the \r
- * same ushort value that gets encoded after the record sid\r
+ * @return the total length of the encoded record(s) \r
+ * (Note - if any {@link ContinueRecord} is required, this result includes the\r
+ * size of those too)\r
*/\r
- protected final int getDataSize() {\r
+ public final int getRecordSize() {\r
ContinuableRecordOutput out = ContinuableRecordOutput.createForCountingOnly();\r
serialize(out);\r
out.terminate();\r
- return out.getTotalSize() - 4;\r
+ return out.getTotalSize();\r
}\r
\r
public final int serialize(int offset, byte[] data) {\r
import org.apache.poi.hssf.model.Sheet;
import org.apache.poi.hssf.record.NameRecord;
import org.apache.poi.hssf.record.Record;
+import org.apache.poi.hssf.record.RecordBase;
import org.apache.poi.hssf.record.RecordFormatException;
import org.apache.poi.hssf.record.formula.Area3DPtg;
import org.apache.poi.util.LittleEndian;
import org.apache.poi.util.TempFile;
/**
- *
+ * Tests for {@link HSSFWorkbook}
*/
public final class TestHSSFWorkbook extends TestCase {
private static HSSFWorkbook openSample(String sampleFileName) {
b.createSheet("Sheet1");
b.createSheet();
b.createSheet("name1");
- try
- {
+ try {
b.createSheet("name1");
fail();
- }
- catch ( IllegalArgumentException pass )
- {
+ } catch (IllegalArgumentException pass) {
+ // expected during successful test
}
b.createSheet();
- try
- {
- b.setSheetName( 3, "name1" );
+ try {
+ b.setSheetName(3, "name1");
fail();
- }
- catch ( IllegalArgumentException pass )
- {
+ } catch (IllegalArgumentException pass) {
+ // expected during successful test
}
- try
- {
- b.setSheetName( 3, "name1" );
+ try {
+ b.setSheetName(3, "name1");
fail();
- }
- catch ( IllegalArgumentException pass )
- {
+ } catch (IllegalArgumentException pass) {
+ // expected during successful test
}
b.setSheetName( 3, "name2" );
assertEquals("active", expectedActive, sheet.isActive());
assertEquals("selected", expectedSelected, sheet.isSelected());
}
-
+
/**
* If Sheet.getSize() returns a different result to Sheet.serialize(), this will cause the BOF
- * records to be written with invalid offset indexes. Excel does not like this, and such
+ * records to be written with invalid offset indexes. Excel does not like this, and such
* errors are particularly hard to track down. This test ensures that HSSFWorkbook throws
* a specific exception as soon as the situation is detected. See bugzilla 45066
*/
public void testSheetSerializeSizeMismatch_bug45066() {
HSSFWorkbook wb = new HSSFWorkbook();
Sheet sheet = wb.createSheet("Sheet1").getSheet();
- List sheetRecords = sheet.getRecords();
+ List<RecordBase> sheetRecords = sheet.getRecords();
// one way (of many) to cause the discrepancy is with a badly behaved record:
sheetRecords.add(new BadlyBehavedRecord());
// There is also much logic inside Sheet that (if buggy) might also cause the discrepancy
assertTrue(e.getMessage().startsWith("Actual serialized sheet size"));
}
}
-
+
/**
* Checks that us and HSSFName play nicely with named ranges
* that point to deleted sheets
*/
- public void testNamesToDeleteSheets() throws Exception {
+ public void testNamesToDeleteSheets() {
HSSFWorkbook b = openSample("30978-deleted.xls");
assertEquals(3, b.getNumberOfNames());
-
+
// Sheet 2 is deleted
assertEquals("Sheet1", b.getSheetName(0));
assertEquals("Sheet3", b.getSheetName(1));
-
+
Area3DPtg ptg;
NameRecord nr;
HSSFName n;
-
+
/* ======= Name pointing to deleted sheet ====== */
-
+
// First at low level
nr = b.getWorkbook().getNameRecord(0);
assertEquals("On2", nr.getNameText());
assertEquals(0, nr.getSheetNumber());
assertEquals(1, nr.getExternSheetNumber());
assertEquals(1, nr.getNameDefinition().length);
-
+
ptg = (Area3DPtg)nr.getNameDefinition()[0];
assertEquals(1, ptg.getExternSheetIndex());
assertEquals(0, ptg.getFirstColumn());
assertEquals(0, ptg.getFirstRow());
assertEquals(0, ptg.getLastColumn());
assertEquals(2, ptg.getLastRow());
-
+
// Now at high level
n = b.getNameAt(0);
assertEquals("On2", n.getNameName());
assertEquals("", n.getSheetName());
- assertEquals("#REF!$A$1:$A$3", n.getReference());
-
-
+ assertEquals("#REF!$A$1:$A$3", n.getRefersToFormula());
+
+
/* ======= Name pointing to 1st sheet ====== */
-
+
// First at low level
nr = b.getWorkbook().getNameRecord(1);
assertEquals("OnOne", nr.getNameText());
assertEquals(0, nr.getSheetNumber());
assertEquals(0, nr.getExternSheetNumber());
assertEquals(1, nr.getNameDefinition().length);
-
+
ptg = (Area3DPtg)nr.getNameDefinition()[0];
assertEquals(0, ptg.getExternSheetIndex());
assertEquals(0, ptg.getFirstColumn());
assertEquals(2, ptg.getFirstRow());
assertEquals(0, ptg.getLastColumn());
assertEquals(3, ptg.getLastRow());
-
+
// Now at high level
n = b.getNameAt(1);
assertEquals("OnOne", n.getNameName());
assertEquals("Sheet1", n.getSheetName());
- assertEquals("Sheet1!$A$3:$A$4", n.getReference());
-
-
+ assertEquals("Sheet1!$A$3:$A$4", n.getRefersToFormula());
+
+
/* ======= Name pointing to 3rd sheet ====== */
-
+
// First at low level
nr = b.getWorkbook().getNameRecord(2);
assertEquals("OnSheet3", nr.getNameText());
assertEquals(0, nr.getSheetNumber());
assertEquals(2, nr.getExternSheetNumber());
assertEquals(1, nr.getNameDefinition().length);
-
+
ptg = (Area3DPtg)nr.getNameDefinition()[0];
assertEquals(2, ptg.getExternSheetIndex());
assertEquals(0, ptg.getFirstColumn());
assertEquals(0, ptg.getFirstRow());
assertEquals(0, ptg.getLastColumn());
assertEquals(1, ptg.getLastRow());
-
+
// Now at high level
n = b.getNameAt(2);
assertEquals("OnSheet3", n.getNameName());
assertEquals("Sheet3", n.getSheetName());
- assertEquals("Sheet3!$A$1:$A$2", n.getReference());
+ assertEquals("Sheet3!$A$1:$A$2", n.getRefersToFormula());
}
-
+
/**
* result returned by getRecordSize() differs from result returned by serialize()
*/
private static final class BadlyBehavedRecord extends Record {
public BadlyBehavedRecord() {
- //
+ //
}
public short getSid() {
return 0x777;
public int serialize(int offset, byte[] data) {
return 4;
}
- protected int getDataSize() {
- return 4;
+ public int getRecordSize() {
+ return 8;
}
}
-
+
/**
* The sample file provided with bug 45582 seems to have one extra byte after the EOFRecord
*/
}
}
}
-
+
/**
* Test to make sure that NameRecord.getSheetNumber() is interpreted as a
* 1-based sheet tab index (not a 1-based extern sheet index)
nr = wb.getWorkbook().getNameRecord(2);
// TODO - render full row and full column refs properly
assertEquals("Sheet2!$A$1:$IV$1", HSSFFormulaParser.toFormulaString(wb, nr.getNameDefinition())); // 1:1
-
+
try {
wb.setRepeatingRowsAndColumns(3, 4, 5, 8, 11);
} catch (RuntimeException e) {
if (e.getMessage().equals("Builtin (7) already exists for sheet (4)")) {
- // there was a problem in the code which locates the existing print titles name record
+ // there was a problem in the code which locates the existing print titles name record
throw new RuntimeException("Identified bug 45720b");
}
throw e;
projects / poi.git / commitdiff