matz / ruby-lang.org (Yukihiro Matsumoto) wrote in message news:<1059381472.890681.25332.nullmailer / picachu.netlab.jp>...
> Hello,
> 
> I just put preview5 archive on the ftp server.
> 
>    ftp://ftp.ruby-lang.org/pub/ruby/1.8/ruby-1.8.0-preview5.tar.gz
> 
> Unless something wrong happens, this would be the last preview before
> the final 1.8.0 release.
> 
> 							matz.


Looks pretty good on Tru64 5.1A (osf1 5.1a).  Fewer compiler warnings
than preview4 and the stuff you said you fixed looks fixed.  See below
for the rest:

29 warnings like this (signed/unsigned ptr), down from 95 in preview4:
cc: Warning: io.c, line 868: In the initializer for p, the referenced
type of th
e pointer value "((f)->_ptr)" is "unsigned char", which is not
compatible with "
const char" because they differ by signed/unsigned attribute.
(ptrmismatch1)
            const char *p = READ_DATA_PENDING_PTR(f);
----------------------------^
I'm not very worried about those.

A warning you said could be ignored:
cc: Warning: stubs.c, line 94: Source file does not contain any
declarations. (e
mptyfile)
#endif
------^

A warning due to _XOPEN_SOURCE_EXTENDED not set in curses.h, probably
not very important and dangerous to fix:
cc: Warning: curses.c, line 434: In this statement, "keyname(...)" of
type "int"
, is being converted to "pointer to const char". (cvtdiftypes)
  name = keyname(NUM2INT(c));
--^

And that's it, no more warnings.  Thanks matz!

Doug