void UNICHARSET::unichar_insert(const char* const unichar_repr) {
if (!ids.contains(unichar_repr)) {
if (strlen(unichar_repr) > UNICHAR_LEN) {
fprintf(stderr, "Utf8 buffer too big, size=%d for %s\n",
int(strlen(unichar_repr)), unichar_repr);
return;
}
if (size_used == size_reserved) {
if (size_used == 0)
reserve(8);
else
reserve(2 * size_used);
}
strcpy(unichars[size_used].representation, unichar_repr);
this->set_script(size_used, null_script);
// If the given unichar_repr represents a fragmented character, set
// fragment property to a pointer to CHAR_FRAGMENT class instance with
// information parsed from the unichar representation. Use the script
// of the base unichar for the fragmented character if possible.
CHAR_FRAGMENT *frag = CHAR_FRAGMENT::parse_from_string(unichar_repr);
this->unichars[size_used].properties.fragment = frag;
if (frag != NULL && this->contains_unichar(frag->get_unichar())) {
this->unichars[size_used].properties.script_id =
this->get_script(frag->get_unichar());
}
this->unichars[size_used].properties.enabled = true;
ids.insert(unichar_repr, size_used);
++size_used;
}
}
void UNICHARSET::unichar_insert(const char* const unichar_repr,
OldUncleanUnichars old_style) {
if (old_style == OldUncleanUnichars::kTrue) old_style_included_ = true;
std::string cleaned =
old_style_included_ ? unichar_repr : CleanupString(unichar_repr);
if (!cleaned.empty() && !ids.contains(cleaned.data(), cleaned.size())) {
const char* str = cleaned.c_str();
GenericVector<int> encoding;
if (!old_style_included_ &&
encode_string(str, true, &encoding, nullptr, nullptr))
return;
if (size_used == size_reserved) {
if (size_used == 0)
reserve(8);
else
reserve(2 * size_used);
}
int index = 0;
do {
if (index >= UNICHAR_LEN) {
fprintf(stderr, "Utf8 buffer too big, size>%d for %s\n", UNICHAR_LEN,
unichar_repr);
return;
}
unichars[size_used].representation[index++] = *str++;
} while (*str != '\0');
unichars[size_used].representation[index] = '\0';
this->set_script(size_used, null_script);
// If the given unichar_repr represents a fragmented character, set
// fragment property to a pointer to CHAR_FRAGMENT class instance with
// information parsed from the unichar representation. Use the script
// of the base unichar for the fragmented character if possible.
CHAR_FRAGMENT* frag =
CHAR_FRAGMENT::parse_from_string(unichars[size_used].representation);
this->unichars[size_used].properties.fragment = frag;
if (frag != nullptr && this->contains_unichar(frag->get_unichar())) {
this->unichars[size_used].properties.script_id =
this->get_script(frag->get_unichar());
}
this->unichars[size_used].properties.enabled = true;
ids.insert(unichars[size_used].representation, size_used);
++size_used;
}
}