This is an automated email from the ASF dual-hosted git repository.
fanningpj pushed a commit to branch trunk
in repository https://gitbox.apache.org/repos/asf/poi.git
The following commit(s) were added to refs/heads/trunk by this push:
new 9d37011683 workaround stray whitespace in sst index (#874)
9d37011683 is described below
commit 9d370116839056efd542612ca9a939d009e78575
Author: PJ Fanning <[email protected]>
AuthorDate: Tue Aug 5 12:00:54 2025 +0100
workaround stray whitespace in sst index (#874)
* tolerate whitespace in ints
* try to workaround stray whitespace in sst index
* Update TestAllFiles.java
---
.../java/org/apache/poi/stress/TestAllFiles.java | 3 ++
.../xssf/eventusermodel/XSSFSheetXMLHandler.java | 21 ++++++++----
.../org/apache/poi/xssf/usermodel/XSSFCell.java | 23 ++++++++++---
.../eventusermodel/TestXSSFSheetXMLHandler.java | 37 +++++++++++++++++++++
.../apache/poi/xssf/usermodel/TestXSSFBugs.java | 28 ++++++++++++++++
test-data/spreadsheet/bug69769.xlsx | Bin 0 -> 13976 bytes
6 files changed, 100 insertions(+), 12 deletions(-)
diff --git
a/poi-integration/src/test/java/org/apache/poi/stress/TestAllFiles.java
b/poi-integration/src/test/java/org/apache/poi/stress/TestAllFiles.java
index 957b804ff5..8988c8a180 100644
--- a/poi-integration/src/test/java/org/apache/poi/stress/TestAllFiles.java
+++ b/poi-integration/src/test/java/org/apache/poi/stress/TestAllFiles.java
@@ -100,6 +100,9 @@ public class TestAllFiles {
// stress docs
"document/deep-table-cell.docx",
+ // invalid files
+ "spreadsheet/bug69769.xlsx",
+
// NOTE: Expected failures should usually be added in file
"stress.xls" instead
// of being listed here in order to also verify the expected exception
details!
};
diff --git
a/poi-ooxml/src/main/java/org/apache/poi/xssf/eventusermodel/XSSFSheetXMLHandler.java
b/poi-ooxml/src/main/java/org/apache/poi/xssf/eventusermodel/XSSFSheetXMLHandler.java
index dbe3bcdee6..110d343d7b 100644
---
a/poi-ooxml/src/main/java/org/apache/poi/xssf/eventusermodel/XSSFSheetXMLHandler.java
+++
b/poi-ooxml/src/main/java/org/apache/poi/xssf/eventusermodel/XSSFSheetXMLHandler.java
@@ -253,7 +253,7 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
} else if ("row".equals(localName)) {
String rowNumStr = attributes.getValue("r");
if (rowNumStr != null) {
- rowNum = Integer.parseInt(rowNumStr) - 1;
+ rowNum = parseInt(rowNumStr) - 1;
} else {
rowNum = nextRowNum;
}
@@ -284,7 +284,7 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
XSSFCellStyle style = null;
if (stylesTable != null) {
if (cellStyleStr != null) {
- int styleIndex = Integer.parseInt(cellStyleStr);
+ int styleIndex = parseInt(cellStyleStr);
style = stylesTable.getStyleAt(styleIndex);
} else if (stylesTable.getNumCellStyles() > 0) {
style = stylesTable.getStyleAt(0);
@@ -392,7 +392,7 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
if (this.formatString != null) {
try {
// Try to use the value as a formattable number
- double d = Double.parseDouble(fv);
+ double d = parseDouble(fv);
thisStr = formatter.formatRawCellContents(d,
this.formatIndex, this.formatString);
} catch (Exception e) {
// Formula is a String result not a Numeric one
@@ -416,10 +416,10 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
break;
case SST_STRING:
- String sstIndex = value.toString();
+ String sstIndex = value.toString().trim();
if (!sstIndex.isEmpty()) {
try {
- int idx = Integer.parseInt(sstIndex);
+ int idx = parseInt(sstIndex);
RichTextString rtss =
sharedStringsTable.getItemAt(idx);
thisStr = rtss.toString();
} catch (NumberFormatException ex) {
@@ -433,7 +433,7 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
if (this.formatString != null && !n.isEmpty()) {
try {
thisStr = formatter.formatRawCellContents(
- Double.parseDouble(n), this.formatIndex,
this.formatString);
+ parseDouble(n), this.formatIndex,
this.formatString);
} catch (Exception e) {
LOG.atInfo().log(
"Error formatting cell '{}' - will use its
raw value instead (format '{}')",
@@ -515,7 +515,6 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
}
}
-
/**
* Output an empty-cell comment.
*/
@@ -524,6 +523,14 @@ public class XSSFSheetXMLHandler extends DefaultHandler {
output.cell(cellRef.formatAsString(), null, comment);
}
+ private static int parseInt(String value) throws NumberFormatException {
+ return Integer.parseInt(value.trim());
+ }
+
+ private static double parseDouble(String value) throws
NumberFormatException {
+ return Double.parseDouble(value.trim());
+ }
+
private enum EmptyCellCommentsCheckType {
CELL,
END_OF_ROW,
diff --git
a/poi-ooxml/src/main/java/org/apache/poi/xssf/usermodel/XSSFCell.java
b/poi-ooxml/src/main/java/org/apache/poi/xssf/usermodel/XSSFCell.java
index 29088a5b48..d23375fda5 100644
--- a/poi-ooxml/src/main/java/org/apache/poi/xssf/usermodel/XSSFCell.java
+++ b/poi-ooxml/src/main/java/org/apache/poi/xssf/usermodel/XSSFCell.java
@@ -21,6 +21,8 @@ import java.time.LocalDateTime;
import java.util.Calendar;
import java.util.Date;
+import org.apache.logging.log4j.Logger;
+import org.apache.poi.logging.PoiLogManager;
import org.apache.poi.ss.SpreadsheetVersion;
import org.apache.poi.ss.formula.FormulaParser;
import org.apache.poi.ss.formula.FormulaRenderer;
@@ -73,6 +75,7 @@ import
org.openxmlformats.schemas.spreadsheetml.x2006.main.STCellType;
*/
public final class XSSFCell extends CellBase {
+ private static final Logger LOG = PoiLogManager.getLogger(XSSFCell.class);
private static final String FALSE_AS_STRING = "0";
private static final String TRUE_AS_STRING = "1";
private static final String FALSE = "FALSE";
@@ -244,7 +247,7 @@ public final class XSSFCell extends CellBase {
return 0.0;
}
try {
- return Double.parseDouble(v);
+ return parseDouble(v);
} catch(NumberFormatException e) {
throw typeMismatch(CellType.NUMERIC, CellType.STRING,
false);
}
@@ -330,12 +333,13 @@ public final class XSSFCell extends CellBase {
} else {
if (_cell.isSetV()) {
try {
- int idx = Integer.parseInt(_cell.getV());
+ int idx = parseInt(_cell.getV());
rt =
(XSSFRichTextString)_sharedStringSource.getItemAt(idx);
} catch (Throwable t) {
if (ExceptionUtil.isFatal(t)) {
ExceptionUtil.rethrow(t);
}
+ LOG.atError().withThrowable(t).log("Failed to parse SST
index '{}'", _cell.getV());
rt = new XSSFRichTextString("");
}
} else {
@@ -1122,12 +1126,12 @@ public final class XSSFCell extends CellBase {
case BOOLEAN:
return TRUE_AS_STRING.equals(_cell.getV());
case STRING:
- int sstIndex = Integer.parseInt(_cell.getV());
+ int sstIndex = parseInt(_cell.getV());
RichTextString rt = _sharedStringSource.getItemAt(sstIndex);
String text = rt.getString();
return Boolean.parseBoolean(text);
case NUMERIC:
- return Double.parseDouble(_cell.getV()) != 0;
+ return parseDouble(_cell.getV()) != 0;
case ERROR:
// fall-through
@@ -1149,13 +1153,14 @@ public final class XSSFCell extends CellBase {
return TRUE_AS_STRING.equals(_cell.getV()) ? TRUE : FALSE;
case STRING:
try {
- int sstIndex = Integer.parseInt(_cell.getV());
+ int sstIndex = parseInt(_cell.getV());
RichTextString rt =
_sharedStringSource.getItemAt(sstIndex);
return rt.getString();
} catch (Throwable t) {
if (ExceptionUtil.isFatal(t)) {
ExceptionUtil.rethrow(t);
}
+ LOG.atError().withThrowable(t).log("Failed to parse SST
index '{}'", _cell.getV());
return "";
}
case NUMERIC:
@@ -1227,4 +1232,12 @@ public final class XSSFCell extends CellBase {
ctCell.setR(r);
}
+ private static int parseInt(String value) throws NumberFormatException {
+ return Integer.parseInt(value.trim());
+ }
+
+ private static double parseDouble(String value) throws
NumberFormatException {
+ return Double.parseDouble(value.trim());
+ }
+
}
diff --git
a/poi-ooxml/src/test/java/org/apache/poi/xssf/eventusermodel/TestXSSFSheetXMLHandler.java
b/poi-ooxml/src/test/java/org/apache/poi/xssf/eventusermodel/TestXSSFSheetXMLHandler.java
index b3f1215925..7e5a3071f2 100644
---
a/poi-ooxml/src/test/java/org/apache/poi/xssf/eventusermodel/TestXSSFSheetXMLHandler.java
+++
b/poi-ooxml/src/test/java/org/apache/poi/xssf/eventusermodel/TestXSSFSheetXMLHandler.java
@@ -26,7 +26,9 @@ import org.xml.sax.InputSource;
import org.xml.sax.XMLReader;
import java.io.InputStream;
+import java.util.HashMap;
import java.util.Iterator;
+import java.util.Map;
import static org.junit.jupiter.api.Assertions.assertDoesNotThrow;
import static org.junit.jupiter.api.Assertions.assertEquals;
@@ -100,4 +102,39 @@ class TestXSSFSheetXMLHandler {
}
}
}
+
+ @Test
+ void testSstStrayWhitespace() throws Exception {
+ try (OPCPackage xlsxPackage =
OPCPackage.open(_ssTests.openResourceAsStream("bug69769.xlsx"))) {
+ final XSSFReader reader = new XSSFReader(xlsxPackage);
+ final Iterator<InputStream> iter = reader.getSheetsData();
+ final Map<String, String> cellValues = new HashMap<>();
+
+ try (InputStream stream = iter.next()) {
+ final XMLReader sheetParser =
XMLHelper.getSaxParserFactory().newSAXParser().getXMLReader();
+
+ sheetParser.setContentHandler(new
XSSFSheetXMLHandler(reader.getStylesTable(),
+ new ReadOnlySharedStringsTable(xlsxPackage), new
SheetContentsHandler() {
+ @Override
+ public void startRow(final int rowNum) {
+ }
+
+ @Override
+ public void endRow(final int rowNum) {
+ }
+
+ @Override
+ public void cell(final String cellReference, final String
formattedValue,
+ final XSSFComment comment) {
+ cellValues.put(cellReference, formattedValue);
+ }
+ }, false));
+
+ assertDoesNotThrow(() -> sheetParser.parse(new
InputSource(stream)));
+
+ assertEquals(4, cellValues.size());
+ assertEquals("Mustermann", cellValues.get("B2"));
+ }
+ }
+ }
}
diff --git
a/poi-ooxml/src/test/java/org/apache/poi/xssf/usermodel/TestXSSFBugs.java
b/poi-ooxml/src/test/java/org/apache/poi/xssf/usermodel/TestXSSFBugs.java
index ef8c8542a0..22a9ec10e2 100644
--- a/poi-ooxml/src/test/java/org/apache/poi/xssf/usermodel/TestXSSFBugs.java
+++ b/poi-ooxml/src/test/java/org/apache/poi/xssf/usermodel/TestXSSFBugs.java
@@ -3910,6 +3910,34 @@ public final class TestXSSFBugs extends
BaseTestBugzillaIssues {
}
}
+ @Test
+ void testBug69769() throws Exception {
+ final int expectedCount = 3;
+ try (XSSFWorkbook wb = openSampleWorkbook("bug69769.xlsx")) {
+ SharedStringsTable sst = wb.getSharedStringSource();
+ assertNotNull(sst);
+ assertEquals(expectedCount, sst.getCount());
+ for (int i = 0; i < expectedCount; i++) {
+ assertNotNull(sst.getItemAt(i));
+ }
+ XSSFSheet ws = wb.getSheetAt(0);
+ int nRowCount = ws.getLastRowNum();
+ DataFormatter df = new DataFormatter();
+ for (int r = 0; r <= nRowCount; r++) {
+ XSSFRow row = ws.getRow(r);
+ if (row != null) {
+ for (Cell cell : row) {
+ String cellValue = df.formatCellValue(cell);
+ assertNotNull(cellValue, "Cell value should not be
null");
+ if (cell.getRowIndex() == 1 && cell.getColumnIndex()
== 1) {
+ assertEquals("Mustermann", cellValue);
+ }
+ }
+ }
+ }
+ }
+ }
+
private static void readByCommonsCompress(File temp_excel_poi) throws
IOException {
/* read by commons-compress*/
try (ZipFile zipFile =
ZipFile.builder().setFile(temp_excel_poi).get()) {
diff --git a/test-data/spreadsheet/bug69769.xlsx
b/test-data/spreadsheet/bug69769.xlsx
new file mode 100644
index 0000000000..39262dfff3
Binary files /dev/null and b/test-data/spreadsheet/bug69769.xlsx differ
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
