tdf#106755: Fix script type for combining marks

We are classifying characters in the “Combining Diacritical Marks”
Unicode block with ScriptType::LATIN, but these are combining marks and
can combine with any script and should have been ScriptType::WEAK. Just
removing them from the range in scriptList does the trick as we will
fallback to getting the script classification based on the Unicode
script property.

Change-Id: I3577f4b03360a1c8e094a207f01b6bbb6abbaf30
Reviewed-on: https://gerrit.libreoffice.org/35811Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

i18npool/qa/cppunit/test_breakiterator.cxx

@@ -759,6 +759,10 @@ void TestBreakIterator::testWeak()
         {
             0x0001, 0x0002,
             0x0020, 0x00A0,
+            0x0300, 0x036F, //Combining Diacritical Marks
+            0x1AB0, 0x1AFF, //Combining Diacritical Marks Extended
+            0x1DC0, 0x1DFF, //Combining Diacritical Marks Supplement
+            0x20D0, 0x20FF, //Combining Diacritical Marks for Symbols
             0x2150, 0x215F, //Number Forms, fractions
             0x2160, 0x2180, //Number Forms, roman numerals
             0x2200, 0x22FF, //Mathematical Operators

i18npool/source/breakiterator/breakiteratorImpl.cxx

@@ -442,7 +442,8 @@ struct UBlock2Script
 static const UBlock2Script scriptList[] =
 {
     {UBLOCK_NO_BLOCK, UBLOCK_NO_BLOCK, ScriptType::WEAK},
-    {UBLOCK_BASIC_LATIN, UBLOCK_ARMENIAN, ScriptType::LATIN},
+    {UBLOCK_BASIC_LATIN, UBLOCK_SPACING_MODIFIER_LETTERS, ScriptType::LATIN},
+    {UBLOCK_GREEK, UBLOCK_ARMENIAN, ScriptType::LATIN},
     {UBLOCK_HEBREW, UBLOCK_MYANMAR, ScriptType::COMPLEX},
     {UBLOCK_GEORGIAN, UBLOCK_GEORGIAN, ScriptType::LATIN},
     {UBLOCK_HANGUL_JAMO, UBLOCK_HANGUL_JAMO, ScriptType::ASIAN},