int
rb_enc_get_index(VALUE obj)
{
int i = -1;
VALUE tmp;
if (SPECIAL_CONST_P(obj)) {
if (!SYMBOL_P(obj)) return -1;
obj = rb_sym2str(obj);
}
switch (BUILTIN_TYPE(obj)) {
case T_STRING:
case T_SYMBOL:
case T_REGEXP:
i = enc_get_index_str(obj);
break;
case T_FILE:
tmp = rb_funcallv(obj, rb_intern("internal_encoding"), 0, 0);
if (NIL_P(tmp)) {
tmp = rb_funcallv(obj, rb_intern("external_encoding"), 0, 0);
}
if (is_obj_encoding(tmp)) {
i = enc_check_encoding(tmp);
}
break;
case T_DATA:
if (is_data_encoding(obj)) {
i = enc_check_encoding(obj);
}
break;
default:
break;
}
return i;
}
static int
enc_check_encoding(mrb_state *mrb, mrb_value obj)
{
if (SPECIAL_CONST_P(obj) || !is_data_encoding(obj)) {
return -1;
}
return check_encoding(mrb, RDATA(obj)->data);
}
/*
* call-seq:
* Encoding.find(string) -> enc
*
* Search the encoding with specified <i>name</i>.
* <i>name</i> should be a string.
*
* Encoding.find("US-ASCII") #=> #<Encoding:US-ASCII>
*
* Names which this method accept are encoding names and aliases
* including following special aliases
*
* "external":: default external encoding
* "internal":: default internal encoding
* "locale":: locale encoding
* "filesystem":: filesystem encoding
*
* An ArgumentError is raised when no encoding with <i>name</i>.
* Only <code>Encoding.find("internal")</code> however returns nil
* when no encoding named "internal", in other words, when Ruby has no
* default internal encoding.
*/
static VALUE
enc_find(VALUE klass, VALUE enc)
{
int idx;
if (RB_TYPE_P(enc, T_DATA) && is_data_encoding(enc))
return enc;
idx = str_to_encindex(enc);
if (idx == UNSPECIFIED_ENCODING) return Qnil;
return rb_enc_from_encoding_index(idx);
}
static inline int
enc_capable(VALUE obj)
{
if (SPECIAL_CONST_P(obj)) return SYMBOL_P(obj);
switch (BUILTIN_TYPE(obj)) {
case T_STRING:
case T_REGEXP:
case T_FILE:
return TRUE;
case T_DATA:
if (is_data_encoding(obj)) return TRUE;
default:
return FALSE;
}
}
/*
* call-seq:
* enc.inspect -> string
*
* Returns a string which represents the encoding for programmers.
*
* Encoding::UTF_8.inspect #=> "#<Encoding:UTF-8>"
* Encoding::ISO_2022_JP.inspect #=> "#<Encoding:ISO-2022-JP (dummy)>"
*/
static VALUE
enc_inspect(VALUE self)
{
rb_encoding *enc;
if (!is_data_encoding(self)) {
not_encoding(self);
}
if (!(enc = DATA_PTR(self)) || rb_enc_from_index(rb_enc_to_index(enc)) != enc) {
rb_raise(rb_eTypeError, "broken Encoding");
}
return rb_enc_sprintf(rb_usascii_encoding(),
"#<%"PRIsVALUE":%s%s%s>", rb_obj_class(self),
rb_enc_name(enc),
(ENC_DUMMY_P(enc) ? " (dummy)" : ""),
enc_autoload_p(enc) ? " (autoload)" : "");
}
int
rb_enc_get_index(VALUE obj)
{
int i = -1;
VALUE tmp;
if (SPECIAL_CONST_P(obj)) {
if (!SYMBOL_P(obj)) return -1;
obj = rb_id2str(SYM2ID(obj));
}
switch (BUILTIN_TYPE(obj)) {
as_default:
default:
case T_STRING:
case T_REGEXP:
i = ENCODING_GET_INLINED(obj);
if (i == ENCODING_INLINE_MAX) {
VALUE iv;
iv = rb_ivar_get(obj, rb_id_encoding());
i = NUM2INT(iv);
}
break;
case T_FILE:
tmp = rb_funcall(obj, rb_intern("internal_encoding"), 0, 0);
if (NIL_P(tmp)) obj = rb_funcall(obj, rb_intern("external_encoding"), 0, 0);
else obj = tmp;
if (NIL_P(obj)) break;
case T_DATA:
if (is_data_encoding(obj)) {
i = enc_check_encoding(obj);
}
else {
goto as_default;
}
break;
}
return i;
}
int
rb_data_is_encoding(VALUE obj)
{
return is_data_encoding(obj);
}
