[ruby-core:24559] Re: Expected change in behavior for ARGF.eof?

Hi,

2009/7/25 Mike Kasick <[e-mail]>:
> Hi folks,
>
> I originally asked this question in ruby-talk:342383, but the lack of
> significant response suggests to me that it's a bit esoteric.
>
> Anyways, is the following an expected change in behavior between Ruby 1.8=
.7
> and Ruby 1.9.1p0? =C2=A0It seems to also hold for the latest 1.9.2 svn tr=
unk.
>
> $ echo foo > /tmp/foo
> $ ruby1.8 -e 'p ARGF.eof?' /tmp/foo
> false
> $ ruby1.9 -e 'p ARGF.eof?' /tmp/foo
> true
>
I guess it is a bug of 1.9.x.
Here is a patch for 1.9.1-p243

--- io.c        2009-07-15 20:57:55.000000000 +0900
+++ io.c.new    2009-07-25 15:38:04.000000000 +0900
@@ -8040,7 +8040,7 @@
 static VALUE
 argf_eof(VALUE argf)
 {
-    if (ARGF.current_file) {
+    if (!NIL_P(ARGF.current_file)) {
        if (ARGF.init_p =3D=3D 0) return Qtrue;
        ARGF_FORWARD(0, 0);
        if (rb_io_eof(ARGF.current_file)) {


> Also, is there a way to advance ARGF to the next (in ARGV) file once EOF
> for the current file has been reached?
>
In my test, ARGF.gets advanced ARGF to the next file in ARGV
Try with
$ ruby -e 'loop do;s=3DARGF.gets;break if s.nil?;puts s;end' /tmp/foo /tmp/=
bar

Regards,

Park Heesob