Hi Aleksandr,
I started working on the project of adding Cyrillic numerals, and I
decided that first I would add the digits and then, once I figured
that out, work on the algorithm. However, I am stuck. I've added the
digits to nativenumbersupplier.cxx (see attached)
Btw, the attachment made it only through in my personal copy, it was
stripped on the l10n list. And please use patch attachments that are
produced by git format-patch, not just git show.
, and the software
compiles, but when I attempt to format the digits in the Church Slavic
locale, I get gibberish (some kind of CJK ideographs).
Likely because you forgot to adjust NumberChar_Count in
i18npool/source/nativenumber/data/numberchar.h and instead added
NumberChar_cu after:
--- a/i18npool/source/nativenumber/data/numberchar.h
+++ b/i18npool/source/nativenumber/data/numberchar.h
@@ -54,6 +54,7 @@ static const sal_Int16 NumberChar_he = 28;
static const sal_Int16 NumberChar_ne = 29;
static const sal_Int16 NumberChar_dz = 30;
static const sal_Int16 NumberChar_Count = 31;
+static const sal_Int16 NumberChar_cu = 32;
NumberChar_Count is used to allocate and loop over ranges, see places in
i18npool/source/nativenumber/nativenumbersupplier.cxx where
NumberChar_Count is used. If you access with NumberChar_cu beyond that
you'll get arbitrary memory.
So instead it should be
+static const sal_Int16 NumberChar_cu = 31;
-static const sal_Int16 NumberChar_Count = 31;
+static const sal_Int16 NumberChar_Count = 32;
Eike