From: Joel VanderWerf <vjoel@...>
Date: 2009-02-09T05:43:28+09:00
Subject: Re: A Tk window

Hidetoshi NAGAI wrote:
> From: Albert Schlef <albertschlef@gmail.com>
> Subject: Re: A Tk window
> Date: Thu, 5 Feb 2009 22:55:04 +0900
> Message-ID: <584be7f2ea0b447f512b957e71a96d83@ruby-forum.com>
>>>> What does this entail? Will I have to compile Tcl/Tk/Ruby myself?
>>> Nevermind, I've found instructions, here:
>>>
>>> http://www.tkdocs.com/tutorial/install.html
>> I gave up, it's too complicated for me. I'll just wait for the next 
>> release of Ubuntu.
> 
> Did you see '<ruby-src-tree>/ext/tk/README.tcltklib' ?
> Can't it help you ?

I haven't gotten ruby/tk to work with 8.5. It's been ok with 8.4 but 
segfaults with 8.5.

This is on ubuntu. I build tcl8.5.6 and tk8.5.6 with

./configure --disable-threads --prefix=/usr/local
make && make install

The tk demos run correctly.

Then I build ruby (1.8.6-p287) with:

./configure --disable-pthread --with-tcl-dir=/usr/local 
--with-tk-dir=/usr/local --disable-tcl-thread
make && make install

(I tried just rebuilding ext/tk as described in README.tcltklib, but the 
effect was the same.)

Then *some* of the ruby/tk demos run but some segfault. For example:

$ ruby 24hr_clock.rb
*** glibc detected *** ruby: free(): invalid next size (fast): 
0x08448d78 ***
======= Backtrace: =========
/lib/tls/i686/cmov/libc.so.6[0xb7dbca85]
/lib/tls/i686/cmov/libc.so.6(cfree+0x90)[0xb7dc04f0]
/usr/local/lib/libtcl8.5.so(TclpFree+0x1d)[0xb7a0ee8d]
/usr/local/lib/libtcl8.5.so(Tcl_Free+0x1d)[0xb7a1872d]
/usr/local/lib/libtcl8.5.so[0xb7aa4550]
/usr/local/lib/libtcl8.5.so[0xb7a161ca]
/usr/local/lib/libtcl8.5.so(Tcl_GetByteArrayFromObj+0x3b)[0xb7a1623b]
/usr/local/lib/ruby/site_ruby/1.8/i686-linux/tcltklib.so[0xb7bfe1ae]
/usr/local/lib/ruby/site_ruby/1.8/i686-linux/tcltklib.so[0xb7bfedf5]
/usr/local/lib/ruby/site_ruby/1.8/i686-linux/tcltklib.so[0xb7bff2dd]
/usr/local/lib/ruby/site_ruby/1.8/i686-linux/tcltklib.so[0xb7bffb65]
/usr/local/lib/ruby/site_ruby/1.8/i686-linux/tcltklib.so[0xb7bffdce]

(and more backtrace and dump data)

Any ideas?

-- 
       vjoel : Joel VanderWerf : path berkeley edu : 510 665 3407