[ruby-core:28499] Re: [Feature #2759] Regexp /g and /G options

2010/3/5 Caleb Clausen <[e-mail]>:
> Can't you rewrite that as something like this:
>
> =A0 /*untested, intended more as an example*/
> =A0 case T_REGEXP:
> =A0 =A0 w_uclass(obj, rb_cRegexp, arg);
> =A0 =A0 {
> =A0 =A0 =A0 int opts =3D rb_reg_options(obj);
> =A0 =A0 =A0 if (opts<=3D0xFF)
> =A0 =A0 =A0 =A0 w_byte(TYPE_REGEXP, arg);
> =A0 =A0 =A0 else
> =A0 =A0 =A0 =A0 w_byte(TYPE_WIDEREGEXP, arg);
> =A0 =A0 =A0 w_bytes(RREGEXP_SRC_PTR(obj), RREGEXP_SRC_LEN(obj), arg);
> =A0 =A0 =A0 if (opts<=3D0xFF)
> =A0 =A0 =A0 =A0 w_byte((char)opts, arg);
> =A0 =A0 =A0 else
> =A0 =A0 =A0 =A0 w_int(opts, arg);
> =A0 =A0 }
>
> That is, use the old format if the regexp options will fit in a byte,
> else use a new format with a wider field for options. There'd be
> similar logic on the unmarshal side. This new regexp format would be
> unreadable on older ruby versions, but then they won't know what to do
> with 256 in the options field anyway, so there's no loss there.

This code allows 1.9.2 to dump and load older regexps and new extended rege=
xps.
But when older than Ruby 1.9.2 can't load all of dumped data
which include such dumped wideregexp,

I'm thinking the way to dump the regexp as normal object and invisible inst=
ance
encoding.

--=20
NARUSE, Yui
[e-mail]