commit 68c9de338fffcd564da990b88fe7119880339df7
author Khaled Hosny <khaled@aliftype.com> Sat Apr 01 23:34:59 2023 +0200
committer Adolfo Jayme Barrientos <fitojb@ubuntu.com> Sun Apr 02 22:43:05 2023 +0200
tree 34ff4ec08b8b740d2b26980b68d5b15a97efd55a
parent 0aa559589035d496a5b80d0986b4cddf02437bab

tdf#154549: Make combining marks inherit the script of their base

... regardless of the script of the mark itself.

Change-Id: I15a5d8cb2e1d7b7a26316689ffa11c278d1f6365
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149925
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
(cherry picked from commit b320a33c01e4b87d42c2f3828a1dfd88c386df67)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149884
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

vcl/source/gdi/scrptrun.cxx

diff --git a/vcl/source/gdi/scrptrun.cxx b/vcl/source/gdi/scrptrun.cxx
index fbc68ca..19cb547 100644
--- a/vcl/source/gdi/scrptrun.cxx
+++ b/vcl/source/gdi/scrptrun.cxx
@@ -41,6 +41,7 @@
#include <sal/config.h>

#include <rtl/character.hxx>
#include <unicode/uchar.h>
#include <unicode/utypes.h>
#include <unicode/uscript.h>

@@ -119,14 +120,21 @@ struct PairIndices

};

// There are three Unicode script codes for Japanese text, but only one
// OpenType script tag, so we want to keep them in one run as splitting is
// pointless for the purpose of OpenType shaping.
UScriptCode getScript(UChar32 ch, UErrorCode* status)
{
    // tdf#154549
    // Make combining marks inherit the script of their bases, regardless of
    // their own script.
    if (u_getIntPropertyValue(ch, UCHAR_GENERAL_CATEGORY) == U_NON_SPACING_MARK)
        return USCRIPT_INHERITED;

    UScriptCode script = uscript_getScript(ch, status);
    if (U_FAILURE(*status))
        return script;

    // There are three Unicode script codes for Japanese text, but only one
    // OpenType script tag, so we want to keep them in one run as splitting is
    // pointless for the purpose of OpenType shaping.
    if (script == USCRIPT_KATAKANA || script == USCRIPT_KATAKANA_OR_HIRAGANA)
        return USCRIPT_HIRAGANA;
    return script;