unsigned int
rb_enc_codepoint_len(const char *p, const char *e, int *len_p, rb_encoding *enc)
{
int r;
if (e <= p)
rb_raise(rb_eArgError, "empty string");
r = rb_enc_precise_mbclen(p, e, enc);
if (!MBCLEN_CHARFOUND_P(r)) {
rb_raise(rb_eArgError, "invalid byte sequence in %s", rb_enc_name(enc));
}
if (len_p) *len_p = MBCLEN_CHARFOUND_LEN(r);
return rb_enc_mbc_to_codepoint(p, e, enc);
}
static VALUE
bug_str_cstr_term_char(VALUE str)
{
long len;
char *s;
int c;
rb_encoding *enc = rb_enc_get(str);
RSTRING_GETMEM(str, s, len);
s += len;
len = rb_enc_mbminlen(enc);
c = rb_enc_precise_mbclen(s, s + len, enc);
if (!MBCLEN_CHARFOUND_P(c)) {
c = (unsigned char)*s;
}
else {
c = rb_enc_mbc_to_codepoint(s, s + len, enc);
if (!c) return Qnil;
}
return rb_enc_uint_chr((unsigned int)c, enc);
}
int
rb_enc_ascget(const char *p, const char *e, int *len, rb_encoding *enc)
{
unsigned int c, l;
if (e <= p)
return -1;
if (rb_enc_asciicompat(enc)) {
c = (unsigned char)*p;
if (!ISASCII(c))
return -1;
if (len) *len = 1;
return c;
}
l = rb_enc_precise_mbclen(p, e, enc);
if (!MBCLEN_CHARFOUND_P(l))
return -1;
c = rb_enc_mbc_to_codepoint(p, e, enc);
if (!rb_enc_isascii(c, enc))
return -1;
if (len) *len = l;
return c;
}