From: Mark Hubbart <discord@...>
Date: 2004-04-30T01:31:06+09:00
Subject: Re: minor problem with ri -c


On Apr 29, 2004, at 6:23 AM, Joel VanderWerf wrote:

> Dave Thomas wrote:
>> On Apr 29, 2004, at 0:44, Joel VanderWerf wrote:
>>>
>>> The last few lines output by 'ri -c' are:
>>>
>>>      TimeExtentionTest, TrueClass, TypeError, UnboundMethod, Values,
>>>      Vector, YAML, ZeroDivisionError, body;, dig = {}, fatal, 
>>> include,
>>>      klass
>>>
>> Hmm... this is what I get (using the latest CVS HEAD)
> ...
>> Is it possible that the snapshot is old?
>
> The ruby executable says it is:
>
> ruby 1.9.0 (2004-04-27) [i686-linux]
>
> I'll remove my /usr/local/share/ri/1.9 it and see if it the junk comes 
> back when I reinstall 1.9...and yes it does:
>
> # ls -lt /usr/local/share/ri/1.9/system | tail
> drwxr-xr-x    2 root     root         4096 Apr 29 06:12 
> SystemStackError/
> drwxr-xr-x    2 root     root         4096 Apr 29 06:12 Vector/
> drwxr-xr-x    2 root     root         4096 Apr 29 06:12 Method/
> drwxr-xr-x    2 root     root         4096 Apr 29 06:12 
> SystemCallError/
> drwxr-xr-x    4 root     root         4096 Apr 29 06:12 File/
> -rw-r--r--    1 root     root           29 Apr 29 06:11 created.rid
> drwxr-xr-x    2 root     root         4096 Apr 29 06:09 body;/
> drwxr-xr-x    2 root     root         4096 Apr 29 06:09 dig = {}/
> drwxr-xr-x    2 root     root         4096 Apr 29 06:09 include/
> drwxr-xr-x    2 root     root         4096 Apr 29 06:09 klass/
>
> # ri -c -T | tail -n3
>      TimeExtentionTest, TrueClass, TypeError, UnboundMethod, Values,
>      Vector, YAML, ZeroDivisionError, body;, dig = {}, fatal, include,
>      klass
>
> Where is this stuff coming from? Interesting that it was created a few 
> seconds before the legitimate stuff.

FWIW:

mark@imac% ruby -v
ruby 1.9.0 (2004-04-11) [powerpc-darwin]
mark@imac% ri -c | tail -3
      TimeExtentionTest, TrueClass, TypeError, UnboundMethod, Values,
      Vector, YAML, ZeroDivisionError, body;, dig = {}, fatal, include,
      klass
mark@imac% grep -rn 'class << body;' .
./lib/cgi.rb:1044:        (class << body; self; end).class_eval do
mark@imac% grep -rn 'dig = {}' .
./lib/set.rb:375:      class << dig = {}                # :nodoc:
mark@imac% grep -rn 'class' . | grep fatal | grep Exception
./error.c:973:    rb_eFatal         = rb_define_class("fatal", 
rb_eException);
mark@imac% grep -rn 'class << klass' .                                  
        ./.ext/tk.rb:3456:    class << klass
./ext/tk/lib/tk.rb:3456:    class << klass
./lib/singleton.rb:159:    class << klass

That explains (maybe) where those are coming from. But I can't figure 
out where rdoc got the idea that "include" was a class...

HTH,
--Mark