[EMAIL PROTECTED] wrote:
+{
+ // The latex command is "\textLANG{<spec>}" and we have to retain
+ // "\textLANG{<spec>" for the first char but only "<spec>" for all
+ // subsequent chars (this also works when we are passed untranslated
+ // unicode).
+ docstring const latex1 = rtrim(encoding.latexChar(c), "}");
+ int length = latex1.length();
+ os << latex1;
+ while (i < size() - 1) {
+ char_type next = getChar(i + 1);
+ if (!Encodings::isKnownLangChar(next, preamble) ||
+ runningChange != lookupChange(i + 1))
+ break;
What happens for the case of a Greek character following a Cyrillic one?
Could the same mechanism be extended to include Hebrew? Japanese?
Icelandic Runes?
Angus
