tdf#104597 related: restore the mirroring of Bidi_Mirrored characters

The method PDFIProcessor::mirrorString() was removed, and
::comphelper::string::reverseString was used, by Kevin Suo in commit
69e9925ded584113e52f84ef0ed7c224079fa061.

The old PDFIProcessor::mirrorString() did two things:
1) Reverse the code points in the string. However, due to its use of iterateCodePoints(&i)
rather than iterateCodePoints(&i, -1) (i.e. iterating in forward rather than reversed order),
the reverse was not functioning at all.
2) Use GetMirroredChar (i.e. the icu unicode u_charMirror() which was dropped in
commit 720a093461aadff044ac0b1b7b45cf3d416b3785) to get the mirrored-image of
characters with the Bidi_Mirrored properties.

Stephan Bergmann restored the #1 in commit 50d73574b6c3d71f9a539c895a15d6fcda22390b
(by introducing comphelper::string::reverseCodePoints). However, the #2 was not restored.

I do not see a clear reason to drop #2 (at least this is an area I don't quite familiar
with), so it's better to restore it as well, in *this* commit, named as SubstituteBidiMirrored.

Change-Id: Ic263097938310e6e04f3b95bb12e4f4e834198f2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/143041
Tested-by: Jenkins
Reviewed-by: Kevin Suo <suokunlong@126.com>

8 files changed, 33 insertions, 1 deletions

diff --git a/sdext/CppunitTest_sdext_pdfimport.mk b/sdext/CppunitTest_sdext_pdfimport.mk
index 864fb7084c88..a16e1726bae9 100644
--- a/sdext/CppunitTest_sdext_pdfimport.mk
+++ b/sdext/CppunitTest_sdext_pdfimport.mk
@@ -34,6 +34,7 @@ $(eval $(call gb_CppunitTest_use_libraries,sdext_pdfimport,\
     unotest \
     tl \
     xo \
+    i18nutil \
 ))
 
 $(eval $(call gb_CppunitTest_use_library_objects,sdext_pdfimport,pdfimport))
diff --git a/sdext/Executable_pdf2xml.mk b/sdext/Executable_pdf2xml.mk
index b2439164a1ed..81e00eca11c0 100644
--- a/sdext/Executable_pdf2xml.mk
+++ b/sdext/Executable_pdf2xml.mk
@@ -32,6 +32,7 @@ $(eval $(call gb_Executable_use_libraries,pdf2xml,\
     sal \
     tl \
     xo \
+    i18nutil \
 ))
 
 $(eval $(call gb_Executable_use_library_objects,pdf2xml,pdfimport))
diff --git a/sdext/Executable_pdfunzip.mk b/sdext/Executable_pdfunzip.mk
index 1fcca8a863d7..10c179336048 100644
--- a/sdext/Executable_pdfunzip.mk
+++ b/sdext/Executable_pdfunzip.mk
@@ -30,6 +30,7 @@ $(eval $(call gb_Executable_use_libraries,pdfunzip,\
     vcl \
     tl \
     xo \
+    i18nutil \
 ))
 
 $(eval $(call gb_Executable_use_library_objects,pdfunzip,pdfimport))
diff --git a/sdext/Library_pdfimport.mk b/sdext/Library_pdfimport.mk
index b68c106268ae..dc1290cd0eb1 100644
--- a/sdext/Library_pdfimport.mk
+++ b/sdext/Library_pdfimport.mk
@@ -29,10 +29,12 @@ $(eval $(call gb_Library_use_libraries,pdfimport,\
     sal \
     tl \
     xo \
+    i18nutil \
 ))
 
 $(eval $(call gb_Library_use_externals,pdfimport,\
 	boost_headers \
+    icu_headers \
     zlib \
     $(if $(filter-out WNT MACOSX,$(OS)),fontconfig) \
 ))
diff --git a/sdext/source/pdfimport/inc/pdfiprocessor.hxx b/sdext/source/pdfimport/inc/pdfiprocessor.hxx
index fa86a5155905..7cbe7d7a5104 100644
--- a/sdext/source/pdfimport/inc/pdfiprocessor.hxx
+++ b/sdext/source/pdfimport/inc/pdfiprocessor.hxx
@@ -79,6 +79,8 @@ namespace pdfi
 
         static void sortElements( Element* pElement );
 
+        static OUString SubstituteBidiMirrored(const OUString& rString);
+
     private:
         void processGlyphLine();
 
diff --git a/sdext/source/pdfimport/tree/drawtreevisiting.cxx b/sdext/source/pdfimport/tree/drawtreevisiting.cxx
index 2e56bb449662..8338ba6dbfde 100644
--- a/sdext/source/pdfimport/tree/drawtreevisiting.cxx
+++ b/sdext/source/pdfimport/tree/drawtreevisiting.cxx
@@ -123,7 +123,12 @@ void DrawXmlEmitter::visit( TextElement& elem, const std::list< std::unique_ptr<
     }
 
     if (isRTL)  // If so, reverse string
+    {
+        // First, produce mirrored-image for each code point which has the Bidi_Mirrored property.
+        str = PDFIProcessor::SubstituteBidiMirrored(str);
+        // Then, reverse the code points in the string, in backward order.
         str = ::comphelper::string::reverseCodePoints(str);
+    }
 
     m_rEmitContext.rEmitter.beginTag( "text:span", aProps );
 
diff --git a/sdext/source/pdfimport/tree/pdfiprocessor.cxx b/sdext/source/pdfimport/tree/pdfiprocessor.cxx
index 164e06d85bde..46a85a753854 100644
--- a/sdext/source/pdfimport/tree/pdfiprocessor.cxx
+++ b/sdext/source/pdfimport/tree/pdfiprocessor.cxx
@@ -33,7 +33,7 @@
 #include <basegfx/polygon/b2dpolygontools.hxx>
 #include <basegfx/utils/canvastools.hxx>
 #include <basegfx/matrix/b2dhommatrix.hxx>
-#include <vcl/svapp.hxx>
+#include <i18nutil/unicode.hxx>
 
 using namespace com::sun::star;
 
@@ -695,6 +695,21 @@ void PDFIProcessor::sortElements(Element* pEle)
     pEle->Children.sort(lr_tb_sort);
 }
 
+/* Produce mirrored-image for each code point which has the Bidi_Mirrored property, within a string.
+   This need to be done in forward order.
+*/
+OUString PDFIProcessor::SubstituteBidiMirrored(const OUString& rString)
+{
+    const sal_Int32 nLen = rString.getLength();
+    OUStringBuffer aMirror(nLen);
+
+    for (sal_Int32 i = 0; i < nLen;) {
+        const sal_uInt32 nCodePoint = rString.iterateCodePoints(&i);
+        aMirror.appendUtf32(unicode::GetMirroredChar(nCodePoint));
+    }
+    return aMirror.makeStringAndClear();
+}
+
 }
 
 /* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/sdext/source/pdfimport/tree/writertreevisiting.cxx b/sdext/source/pdfimport/tree/writertreevisiting.cxx
index 78c1cf00c865..cc2cee18eb6a 100644
--- a/sdext/source/pdfimport/tree/writertreevisiting.cxx
+++ b/sdext/source/pdfimport/tree/writertreevisiting.cxx
@@ -112,7 +112,12 @@ void WriterXmlEmitter::visit( TextElement& elem, const std::list< std::unique_pt
     }
 
     if (isRTL)  // If so, reverse string
+    {
+        // First, produce mirrored-image for each code point which has the Bidi_Mirrored property.
+        str = PDFIProcessor::SubstituteBidiMirrored(str);
+        // Then, reverse the code points in the string, in backward order.
         str = ::comphelper::string::reverseCodePoints(str);
+    }
 
     m_rEmitContext.rEmitter.beginTag( "text:span", aProps );