Fix hebrew character highlight issue in a special document

core/src/fpdftext/fpdf_text.cpp

Index: core/src/fpdftext/fpdf_text.cpp
diff --git a/core/src/fpdftext/fpdf_text.cpp b/core/src/fpdftext/fpdf_text.cpp
index defad925b96b3fb8916c8ffb646de648566e589d..f633e1951703be4aad9282f2cf8dbf86f0df6e9e 100644
--- a/core/src/fpdftext/fpdf_text.cpp
+++ b/core/src/fpdftext/fpdf_text.cpp
@@ -184,7 +184,12 @@ CTextBaseLine* CTextPage::InsertTextBox(CTextBaseLine* pBaseLine, FX_FLOAT basey
     while (offset < len) {
         FX_DWORD ch = pFont->GetNextChar(pStr, offset);
         CFX_WideString unicode_str = pFont->UnicodeFromCharCode(ch);
-        text += unicode_str;
+        if (unicode_str.IsEmpty()) {
+            text += (FX_WCHAR)ch;

[jbreiden, 2014/08/18 19:10:49]

The example PDF file for this bug contains a
mapping between character codes and Unicode.
Is this mapping not being used?

====================

5 0 obj
<< /Length 353 >>
stream
/CIDInit /ProcSet findresource begin
12 dict begin
begincmap
/CIDSystemInfo
<<
  /Registry (Adobe)
  /Ordering (UCS)
  /Supplement 0
/CMapName /Adobe-Identify-UCS def
/CMapType 2 def
1 begincodespacerange
<0000> <FFFF>
endcodespacerange
1 beginbfrange
<0000> <FFFF> <0000>
endbfrange
endcmap
CMapName currentdict /CMap defineresource pop
end
end
endstream
endobj

+        }
+        else {
+            text += unicode_str;
+        }
     }
     pBaseLine->InsertTextBox(leftx, rightx, topy, bottomy, spacew, fontsize_v, text);
     return pBaseLine;