int
ml_unicode_renderer(varnam *handle,
vtoken *previous,
vtoken *current,
strbuf *output)
{
int rc;
vtoken *virama;
bool removed;
rc = vst_get_virama (handle, &virama);
if (rc) return rc;
if (previous != NULL && current->type == VARNAM_TOKEN_VOWEL && strcmp(current->pattern, "r") == 0)
{
strbuf_add (output, current->value3);
return VARNAM_SUCCESS;
}
if (strcmp(current->tag, RENDER_VALUE2_TAG) == 0 && previous != NULL)
{
#ifdef _VARNAM_VERBOSE
varnam_debug (handle, "ml-unicode-renderer - Found %s tag", RENDER_VALUE2_TAG);
#endif
strbuf_add(output, current->value2);
return VARNAM_SUCCESS;
}
if (current->type == VARNAM_TOKEN_VOWEL && previous != NULL && strcmp(previous->tag, CHIL_TAG) == 0)
{
removed = strbuf_remove_from_last (output, previous->value1);
if (!removed) {
removed = strbuf_remove_from_last (output, previous->value2);
}
if (removed)
{
strbuf_add (output, previous->value3);
strbuf_add (output, current->value2);
return VARNAM_SUCCESS;
}
}
return VARNAM_PARTIAL_RENDERING;
}
static strbuf*
sanitize_word (varnam *handle, const char *word)
{
size_t i;
bool is_special = false;
strbuf *string, *to_remove;
string = get_pooled_string (handle);
to_remove = get_pooled_string (handle);
strbuf_add (string, word);
for (i = 0; i < string->length; i++)
{
is_special = is_special_character (string->buffer[i]);
if (is_special)
strbuf_addc (to_remove, string->buffer[i]);
else
break;
}
strbuf_remove_from_first (string, strbuf_to_s (to_remove));
strbuf_clear (to_remove);
for (i = string->length - 1; i >= 0; i--)
{
is_special = is_special_character (string->buffer[i]);
if (is_special)
strbuf_addc (to_remove, string->buffer[i]);
else
break;
}
strbuf_remove_from_last (string, strbuf_to_s (to_remove));
language_specific_sanitization (string);
/* Remove trailing ZWNJ and leading ZWJ */
strbuf_remove_from_first (string, ZWNJ());
strbuf_remove_from_last (string, ZWNJ());
strbuf_remove_from_first (string, ZWJ());
return string;
}
static void
resolve_token(varnam *handle,
struct token *match,
struct strbuf *string)
{
const char *virama = NULL;
struct varnam_token_rendering *rule;
int rc;
char zwnj[] = {'\xe2', '\x80', '\x8c', '\0'};
assert(handle);
assert(match);
assert(string);
/* will be fixed later */
/* if(handle->internal->virama[0] == '\0') { */
/* fill_general_values(handle, handle->internal->virama, "virama"); */
/* } */
virama = handle->internal->virama;
rule = get_additional_rendering_rule(handle);
if(rule != NULL) {
rc = rule->render(handle, match, string);
if(rc == VARNAM_SUCCESS) {
return;
}
}
if(strcmp(match->value1, handle->internal->virama) == 0) {
/* we are resolving a virama. If the output ends with a virama already, add a
ZWNJ to it, so that following character will not be combined.
if output not ends with virama, add a virama and ZWNJ */
if(strbuf_endswith (string, virama)) {
strbuf_add (string, zwnj);
}
else {
strbuf_add (string, virama);
strbuf_add (string, zwnj);
}
return;
}
if(match->type == VARNAM_TOKEN_VOWEL)
{
if(strbuf_endswith(string, virama)) {
/* removing the virama and adding dependent vowel value */
strbuf_remove_from_last(string, virama);
if(match->value2[0] != '\0') {
strbuf_add(string, match->value2);
}
}
else if(handle->internal->last_token_available) {
strbuf_add(string, match->value2);
}
else {
strbuf_add(string, match->value1);
}
}
else {
strbuf_add(string, match->value1);
}
}
