When in STLXML a character with diacritical chars is represented with two Unicode codepoints instead of a single one (e.g. J́ , which cannot be represented with a single codepoint), it can occur that this character is located at such a bad position that STLXML-SplitBlocks will move the two codepoints into different TTI blocks.
However this will contradict the correct later conversion from STLXML to STL, as in Unicode the combining diacritical mark is a suffix. In STL in contrast, the corresponding diacritical mark is a prefix. While STLXML2STL swaps the order of such two codepoints, it cannot in the described case, as this only affects a single TTI block, but does not go beyond TTI block borders.
So this module must not store such codepoint pair in different TTI blocks.
When in STLXML a character with diacritical chars is represented with two Unicode codepoints instead of a single one (e.g.
J́, which cannot be represented with a single codepoint), it can occur that this character is located at such a bad position that STLXML-SplitBlocks will move the two codepoints into different TTI blocks.However this will contradict the correct later conversion from STLXML to STL, as in Unicode the combining diacritical mark is a suffix. In STL in contrast, the corresponding diacritical mark is a prefix. While
STLXML2STLswaps the order of such two codepoints, it cannot in the described case, as this only affects a single TTI block, but does not go beyond TTI block borders.So this module must not store such codepoint pair in different TTI blocks.