summaryrefslogtreecommitdiff
path: root/writerfilter
diff options
context:
space:
mode:
authorVasily Melenchuk <vasily.melenchuk@cib.de>2022-04-05 19:13:05 +0300
committerMiklos Vajna <vmiklos@collabora.com>2022-04-08 11:20:10 +0200
commitd72dece2bc61e3bab8db5968d53dc0e98a3bea4d (patch)
tree88f691fee71e8e0fc0711451ade6b6b7cbe2ab5d /writerfilter
parent267555f2458e6670a03013a5def67e87d6dc56cd (diff)
tdf#95706: RTF import: Use fontname suffixes to detect encoding
Font names like "Arial CE", "Times New Roman Cyr" are not special fonts. They are classical Arial, Times New Roman... And these suffixes can be used to detect encoding used for RTF text. Most interesting: for MS Word these suffixes have priority: {\f34\cpg1253\fcharset161 Arial Baltic;} will have cp1257 and not cp1253. Looks like compatibility issue came from dark ages. Change-Id: Ife8e781d5d04c3f6a8c11fcf604357c74bf33055 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132584 Tested-by: Jenkins Reviewed-by: Miklos Vajna <vmiklos@collabora.com> Signed-off-by: Xisco Fauli <xiscofauli@libreoffice.org> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132681 Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
Diffstat (limited to 'writerfilter')
-rw-r--r--writerfilter/source/rtftok/rtfcharsets.cxx9
-rw-r--r--writerfilter/source/rtftok/rtfcharsets.hxx13
-rw-r--r--writerfilter/source/rtftok/rtfdocumentimpl.cxx39
3 files changed, 59 insertions, 2 deletions
diff --git a/writerfilter/source/rtftok/rtfcharsets.cxx b/writerfilter/source/rtftok/rtfcharsets.cxx
index 69e416ce8bc5..886e161e3c89 100644
--- a/writerfilter/source/rtftok/rtfcharsets.cxx
+++ b/writerfilter/source/rtftok/rtfcharsets.cxx
@@ -9,6 +9,7 @@
#include "rtfcharsets.hxx"
#include <sal/macros.h>
+#include <rtl/textenc.h>
namespace writerfilter::rtftok
{
@@ -50,6 +51,14 @@ RTFEncoding const aRTFEncodings[] = {
int nRTFEncodings = SAL_N_ELEMENTS(aRTFEncodings);
+RTFFontNameSuffix const aRTFFontNameSuffixes[] = {
+ { "Baltic", RTL_TEXTENCODING_MS_1257 }, { "CE", RTL_TEXTENCODING_MS_1250 },
+ { "Cyr", RTL_TEXTENCODING_MS_1251 }, { "Greek", RTL_TEXTENCODING_MS_1253 },
+ { "Tur", RTL_TEXTENCODING_MS_1254 }, { "(Hebrew)", RTL_TEXTENCODING_MS_1255 },
+ { "(Arabic)", RTL_TEXTENCODING_MS_1256 }, { "(Vietnamese)", RTL_TEXTENCODING_MS_1258 },
+ { "", RTL_TEXTENCODING_DONTKNOW } // End of array
+};
+
} // namespace writerfilter::rtftok
/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/writerfilter/source/rtftok/rtfcharsets.hxx b/writerfilter/source/rtftok/rtfcharsets.hxx
index 865a9310289e..826dea271f6b 100644
--- a/writerfilter/source/rtftok/rtfcharsets.hxx
+++ b/writerfilter/source/rtftok/rtfcharsets.hxx
@@ -19,6 +19,19 @@ struct RTFEncoding
};
extern RTFEncoding const aRTFEncodings[];
extern int nRTFEncodings;
+
+/// Font name can contain special suffixes used
+/// to determine encoding for given font table entry
+/// For example "Arial CE" is "Arial" with CP1250 encoding
+/// List of these suffixes is not official and detected in a empirical
+/// way thus may be inexact and incomplete.
+struct RTFFontNameSuffix
+{
+ const char* suffix;
+ int codepage;
+};
+extern RTFFontNameSuffix const aRTFFontNameSuffixes[];
+
} // namespace writerfilter::rtftok
/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/writerfilter/source/rtftok/rtfdocumentimpl.cxx b/writerfilter/source/rtftok/rtfdocumentimpl.cxx
index af4ba47158c3..8ea858abada5 100644
--- a/writerfilter/source/rtftok/rtfdocumentimpl.cxx
+++ b/writerfilter/source/rtftok/rtfdocumentimpl.cxx
@@ -50,6 +50,7 @@
#include "rtfskipdestination.hxx"
#include "rtftokenizer.hxx"
#include "rtflookahead.hxx"
+#include "rtfcharsets.hxx"
using namespace com::sun::star;
@@ -1370,14 +1371,48 @@ void RTFDocumentImpl::text(OUString& rString)
case Destination::FONTTABLE:
case Destination::FONTENTRY:
{
- m_aFontNames[m_nCurrentFontIndex] = aName;
+ // Old documents can contain no encoding information in fontinfo,
+ // but there can be font name suffixes: Arial CE is not a special
+ // font, it is ordinal Arial, but with used cp 1250 encoding.
+ // Moreover these suffixes have priority over \cpgN and \fcharsetN
+ // in MS Word.
+ OUString aFontSuffix;
+ OUString aNameNoSuffix(aName);
+ sal_Int32 nLastSpace = aName.lastIndexOf(' ');
+ if (nLastSpace >= 0)
+ {
+ aFontSuffix = aName.copy(nLastSpace + 1);
+ aNameNoSuffix = aName.copy(0, nLastSpace);
+ sal_Int32 nEncoding = RTL_TEXTENCODING_DONTKNOW;
+ for (int i = 0;
+ aRTFFontNameSuffixes[i].codepage != RTL_TEXTENCODING_DONTKNOW; i++)
+ {
+ if (aFontSuffix.equalsAscii(aRTFFontNameSuffixes[i].suffix))
+ {
+ nEncoding = aRTFFontNameSuffixes[i].codepage;
+ break;
+ }
+ }
+ if (nEncoding > RTL_TEXTENCODING_DONTKNOW)
+ {
+ m_nCurrentEncoding = nEncoding;
+ m_aStates.top().setCurrentEncoding(m_nCurrentEncoding);
+ }
+ else
+ {
+ // Unknown suffix: looks like it is just a part of font name, restore it
+ aNameNoSuffix = aName;
+ }
+ }
+
+ m_aFontNames[m_nCurrentFontIndex] = aNameNoSuffix;
if (m_nCurrentEncoding >= 0)
{
m_aFontEncodings[m_nCurrentFontIndex] = m_nCurrentEncoding;
m_nCurrentEncoding = -1;
}
m_aStates.top().getTableAttributes().set(NS_ooxml::LN_CT_Font_name,
- new RTFValue(aName));
+ new RTFValue(aNameNoSuffix));
writerfilter::Reference<Properties>::Pointer_t const pProp(
new RTFReferenceProperties(m_aStates.top().getTableAttributes(),