Julien Nabet
7e03ddd4eb
As indicated in the upstream patch (already pushed): Deal with some LMBCS group control LMBCS is Lotus Multi-Byte Character Set Group control in LMBCS begin with 0x0F and the control character is the second byte See https://github.com/unicode-org/icu/blob/main/icu4c/source/common/ucnv_lmb.cpp Change-Id: Ia11edb37a760ec676caa1fe06290e2031459b1c4 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/209113 Tested-by: Jenkins Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
63 lines
1.8 KiB
Groff
63 lines
1.8 KiB
Groff
From 2e1d8e8b8ee0ff54d5c8448b6ead95aaf12f1e52
|
|
Author: Julien Nabet <serval2412@yahoo.fr>
|
|
Date: Thu Aug 13 20:59:25 2026 +0200
|
|
|
|
Deal with some LMBCS group control
|
|
|
|
LMBCS is Lotus Multi-Byte Character Set
|
|
Group control in LMBCS begin with 0x0F and the control character is the second byte
|
|
See https://github.com/unicode-org/icu/blob/main/icu4c/source/common/ucnv_lmb.cpp
|
|
|
|
It'll allow to fix Libreoffice bug tdf#148749: FILEOPEN Lotus 123 entity encoding
|
|
|
|
diff -ur libwps.org/src/lib/LotusSpreadsheet.cpp libwps/src/lib/LotusSpreadsheet.cpp
|
|
--- libwps.org/src/lib/LotusSpreadsheet.cpp 2026-08-13 20:01:53.946906461 +0200
|
|
+++ libwps/src/lib/LotusSpreadsheet.cpp 2026-08-13 20:02:27.643098528 +0200
|
|
@@ -3348,6 +3348,7 @@
|
|
WPSFont font=style.getFont();
|
|
m_listener->setFont(font);
|
|
bool prevEOL=false;
|
|
+ bool prevLMBCS_EOL=false;
|
|
std::string text;
|
|
while (!input->isEnd())
|
|
{
|
|
@@ -3431,6 +3432,38 @@
|
|
}
|
|
prevEOL=false;
|
|
break;
|
|
+ case 0xf:
|
|
+ if (pos+1>=endPos)
|
|
+ {
|
|
+ WPS_DEBUG_MSG(("LotusSpreadsheet::sendText: can not read the control character value\n"));
|
|
+ break;
|
|
+ }
|
|
+ c=char(libwps::readU8(input));
|
|
+ if (c >= 0x20 && c <= 0x2F)
|
|
+ {
|
|
+ c -= 0x20;
|
|
+ switch(c)
|
|
+ {
|
|
+ case 0xd:
|
|
+ m_listener->insertUnicodeString(libwps_tools_win::Font::unicodeString(text, fontType));
|
|
+ text.clear();
|
|
+ m_listener->insertEOL();
|
|
+ prevLMBCS_EOL=true;
|
|
+ break;
|
|
+ case 0xa:
|
|
+ if (!prevLMBCS_EOL)
|
|
+ {
|
|
+ WPS_DEBUG_MSG(("LotusSpreadsheet::sendText: find 0xa without 0xd\n"));
|
|
+ }
|
|
+ prevLMBCS_EOL=false;
|
|
+ break;
|
|
+ }
|
|
+ }
|
|
+ else
|
|
+ {
|
|
+ WPS_DEBUG_MSG(("LotusSpreadsheet::sendText: find a control character followed by a character not included between 0x20 and 0x2F\n"));
|
|
+ }
|
|
+ break;
|
|
default:
|
|
if (c)
|
|
text.push_back(c);
|