// How many char objects can I process to get <= max_limit
// wchar_t objects?
int utf8_codecvt_facet::do_length(
std::mbstate_t &,
const char * from,
const char * from_end,
std::size_t max_limit
) const
{
// RG - this code is confusing! I need a better way to express it.
// and test cases.
// Invariants:
// 1) last_octet_count has the size of the last measured character
// 2) char_count holds the number of characters shown to fit
// within the bounds so far (no greater than max_limit)
// 3) from_next points to the octet 'last_octet_count' before the
// last measured character.
int last_octet_count=0;
std::size_t char_count = 0;
const char* from_next = from;
// Use "<" because the buffer may represent incomplete characters
while (from_next+last_octet_count <= from_end && char_count <= max_limit) {
from_next += last_octet_count;
last_octet_count = (get_octet_count(*from_next));
++char_count;
}
return from_next-from_end;
}
// continuing octets = octets except for the leading octet
static unsigned int get_cont_octet_count(unsigned char lead_octet) {
return get_octet_count(lead_octet) - 1;
}
