moparisthebest/poi
1
0
Fork 0
Code Issues Pull Requests Releases 1 Wiki Activity

bug 50955 - incorporate info from the DocumentSummaryInformation for

Browse Source 
guessing the encoding.  Back off to the old method if DocSummInfo is
not available. Thanks to Andreas Beeker for recommending this direction.

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@1791002 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Tim Allison 2017-04-11 17:07:04 +00:00
parent 44d536e4ce
commit 9412d467f3
1 changed files with 30 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
src/scratchpad/src/org/apache/poi/hwpf/HWPFOldDocument.java
View File

@ -19,8 +19,12 @@ package org.apache.poi.hwpf;
import java.io.File; import java.io.File;
import java.io.IOException; import java.io.IOException;
import java.io.OutputStream; import java.io.OutputStream;
import java.io.UnsupportedEncodingException;
import java.nio.charset.Charset; import java.nio.charset.Charset;
import org.apache.poi.hpsf.CustomProperties;
import org.apache.poi.hpsf.DocumentSummaryInformation;
import org.apache.poi.hpsf.Section;
import org.apache.poi.hwmf.record.HwmfFont; import org.apache.poi.hwmf.record.HwmfFont;
import org.apache.poi.hwpf.model.ComplexFileTable; import org.apache.poi.hwpf.model.ComplexFileTable;
import org.apache.poi.hwpf.model.FontTable; import org.apache.poi.hwpf.model.FontTable;
@ -188,7 +192,32 @@ public class HWPFOldDocument extends HWPFDocumentCore {
* @return The detected Charset from the old font table * @return The detected Charset from the old font table
*/ */
private Charset guessCodePage(OldFontTable fontTable) { private Charset guessCodePage(OldFontTable fontTable) {
//try to get it out of the overall document summary information
DocumentSummaryInformation summaryInformation = getDocumentSummaryInformation();
if (summaryInformation != null) {
CustomProperties customProperties = summaryInformation.getCustomProperties();
if (customProperties != null) {
int codePage = customProperties.getCodepage();
try {
return Charset.forName(CodePageUtil.codepageToEncoding(codePage));
} catch (UnsupportedEncodingException e) {
//swallow
}
}
//for now, try to get first valid code page in a valid section
for (Section section : summaryInformation.getSections()) {
if (section.getOffset() < 0) {
continue;
}
int codePage = section.getCodepage();
try {
return Charset.forName(CodePageUtil.codepageToEncoding(codePage));
} catch (UnsupportedEncodingException e) {
//swallow
}
}
}
//if that still doesn't work, pick the first non-default non symbol charset
for (OldFfn oldFfn : fontTable.getFontNames()) { for (OldFfn oldFfn : fontTable.getFontNames()) {
HwmfFont.WmfCharset wmfCharset = HwmfFont.WmfCharset.valueOf(oldFfn.getChs()& 0xff); HwmfFont.WmfCharset wmfCharset = HwmfFont.WmfCharset.valueOf(oldFfn.getChs()& 0xff);
if (wmfCharset != null && if (wmfCharset != null &&