From 55ddbfc610d2a00e565ca7bcb0277da33bb90947 Mon Sep 17 00:00:00 2001 From: Khaled Hosny Date: Tue, 28 Mar 2017 19:58:25 +0200 Subject: tdf#106755: Fix script type for combining marks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 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/35811 Tested-by: Jenkins Reviewed-by: Caolán McNamara Tested-by: Caolán McNamara --- i18npool/qa/cppunit/test_breakiterator.cxx | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'i18npool/qa/cppunit') diff --git a/i18npool/qa/cppunit/test_breakiterator.cxx b/i18npool/qa/cppunit/test_breakiterator.cxx index c00c716efac2..fb1a155ae9db 100644 --- a/i18npool/qa/cppunit/test_breakiterator.cxx +++ b/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 -- cgit