Apache
/
poi
mirror of https://github.com/apache/poi.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

[bug-63575] support capitalized text in XWPFWordExtractor 

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@1903729 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
PJ Fanning 2022-08-28 12:19:08 +00:00
parent ab5cb372e5
commit 80f89a3674
3 changed files with 21 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
poi-ooxml/src/main/java/org/apache/poi/xwpf/usermodel/XWPFRun.java
View File

@ -27,6 +27,7 @@ import java.math.RoundingMode;
import java.util.ArrayList; import java.util.ArrayList;
import java.util.Arrays; import java.util.Arrays;
import java.util.List; import java.util.List;
import java.util.Locale;
import javax.xml.namespace.QName; import javax.xml.namespace.QName;
@ -35,10 +36,7 @@ import org.apache.poi.ooxml.POIXMLException;
import org.apache.poi.ooxml.util.DocumentHelper; import org.apache.poi.ooxml.util.DocumentHelper;
import org.apache.poi.ooxml.util.POIXMLUnits; import org.apache.poi.ooxml.util.POIXMLUnits;
import org.apache.poi.openxml4j.exceptions.InvalidFormatException; import org.apache.poi.openxml4j.exceptions.InvalidFormatException;
import org.apache.poi.util.HexDump; import org.apache.poi.util.*;
import org.apache.poi.util.Internal;
import org.apache.poi.util.Removal;
import org.apache.poi.util.Units;
import org.apache.poi.wp.usermodel.CharacterRun; import org.apache.poi.wp.usermodel.CharacterRun;
import org.apache.xmlbeans.*; import org.apache.xmlbeans.*;
import org.apache.xmlbeans.impl.values.XmlAnyTypeImpl; import org.apache.xmlbeans.impl.values.XmlAnyTypeImpl;
@ -1381,7 +1379,13 @@ public class XWPFRun implements ISDTContents, IRunElement, CharacterRun {
// come up as instances of CTText, but we don't want them // come up as instances of CTText, but we don't want them
// in the normal text output // in the normal text output
if (!("instrText".equals(node.getLocalName()) && XWPFDocument.NS_OOXML_WP_MAIN.equals(node.getNamespaceURI()))) { if (!("instrText".equals(node.getLocalName()) && XWPFDocument.NS_OOXML_WP_MAIN.equals(node.getNamespaceURI()))) {
text.append(((CTText) o).getStringValue()); String textValue = ((CTText) o).getStringValue();
if (textValue != null) {
if (isCapitalized() || isSmallCaps()) {
textValue = textValue.toUpperCase(LocaleUtil.getUserLocale());
}
text.append(textValue);
}
} }
} }
@ -1391,7 +1395,9 @@ public class XWPFRun implements ISDTContents, IRunElement, CharacterRun {
if (ctfldChar.getFldCharType() == STFldCharType.BEGIN) { if (ctfldChar.getFldCharType() == STFldCharType.BEGIN) {
if (ctfldChar.getFfData() != null) { if (ctfldChar.getFfData() != null) {
for (CTFFCheckBox checkBox : ctfldChar.getFfData().getCheckBoxList()) { for (CTFFCheckBox checkBox : ctfldChar.getFfData().getCheckBoxList()) {
text.append((checkBox.getDefault() != null && POIXMLUnits.parseOnOff(checkBox.getDefault().xgetVal())) ? "|X|" : "|_|"); String textValue = checkBox.getDefault() != null && POIXMLUnits.parseOnOff(checkBox.getDefault().xgetVal()) ?
"|X|" : "|_|";
text.append(textValue);
} }
} }
} }

9
poi-ooxml/src/test/java/org/apache/poi/xwpf/extractor/TestXWPFWordExtractor.java
View File

@ -478,4 +478,13 @@ class TestXWPFWordExtractor {
assertEquals(expected, actual); assertEquals(expected, actual);
} }
} }
@Test
void testCapitalizedFlag() throws IOException {
try (XWPFDocument doc = XWPFTestDataSamples.openSampleDocument("capitalized.docx");
XWPFWordExtractor extractor = new XWPFWordExtractor(doc)) {
String txt = extractor.getText();
assertEquals( "The following word is: CAPITALIZED.", txt.trim());
}
}
} }

BIN
test-data/document/capitalized.docx Normal file
View File

Binary file not shown.