Test Arithmetics hangs

[GoogleCodeExporter]

What steps will reproduce the problem?

1. download ypsilon-0.9.5.update1
2. type make
3. type make check

What is the expected output? What do you see instead?

I expect make check to complete.  Instead, I see the command hang at this point:

...Test Arithmetics

Passed 598

Interestingly, I can't ctrl-C out of it, I have to ctrl-Z, then  kill %1

What version of the product are you using? On what operating system?

ypsilon-0.9.5.update1, on Mac OS X 10.5.4

Please provide any additional information below.

I understand that update 1 is supposed to fix this, but it didn't on my system, 
unless I'm doing 
something wrong.

Original issue reported on code.google.com by dylan.mc...@gmail.com on 16 Jul 2008 at 3:51

[GoogleCodeExporter]

Thank you for your bug report!

Original comment by y.fujita...@gmail.com on 16 Jul 2008 at 2:36

• Changed state: Accepted

[GoogleCodeExporter]

The latest version of Ypsilon fixed several bugs. (include bug which cause
unpredictable crash)
Please try latest version and let me know if it solve the problem or not. Thank 
you!

You can checkout latest source code by using Xcode3.
Please refer following URL for information.
http://code.google.com/p/ypsilon/wiki/Xcode3checkout

Original comment by y.fujita...@gmail.com on 18 Jul 2008 at 7:41

• Changed state: Started

[GoogleCodeExporter]

I built "trunk" from the SVN repo, and still have the same behavior.  Is that 
the correct branch to be building?

Original comment by dylan.mc...@gmail.com on 20 Jul 2008 at 5:26

[GoogleCodeExporter]

>I built "trunk" from the SVN repo, and still have the same behavior. Is that 
the
correct branch to be building?
Yes, it is correct one.

Please try following session and let me know if it hangs. 
(I want identify which expression trigger the bug if possible.)

-----------
$ ./ypsilon
Ypsilon 0.9.5-trunk Copyright (c) 2008 Y.Fujita, LittleWing Company Limited.
> (string->number "#e1e-2000")
1/100000.........0

> (string->number "#e1e1998")
10000000........0

> (* (string->number "#e1e-2000") (string->number "#e1e1998"))
1/100
-----------

Also, please let me know an output of 'gcc --version'.
(I want to try same gcc on my machine to reproduce problem.)

Thank you!

Original comment by y.fujita...@gmail.com on 20 Jul 2008 at 2:39

[GoogleCodeExporter]

all of the above expressions hang.

here's the output of gcc --version:

i686-apple-darwin9-gcc-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5484)

Original comment by dylan.mc...@gmail.com on 20 Jul 2008 at 3:27

[GoogleCodeExporter]

a bit more info: not all "string->number" expressions fail:

> (string->number "#e1e199") 
10000...0

works, which made me wonder if #e1e1998 wasn't hanging, but was instead just 
taking a long time. So I let it 
run. 5 minutes later it was still running (consuming 1 of my 2 CPU's), so it 
_seems_ to be in a loop. How long 
should this test take?  (I'm on a MacBook Pro, 2.2GHz processors)

Original comment by dylan.mc...@gmail.com on 20 Jul 2008 at 4:51

[GoogleCodeExporter]

Thank you for your input!

Regarding (string->number "#e1e1998"), it takes approximately 1/10000 second if 
it
works correctly.
Your insight about 'be in loop' give me some idea :)

I have updated trunk directory.
Please try and let me know if it work.

Original comment by y.fujita...@gmail.com on 21 Jul 2008 at 8:12

[GoogleCodeExporter]

I updated trunk, and only got a new Makefile. I did a make clean and rebuild, 
and was surprised to see that it 
worked! "Passed all tests." Congrats. 

Original comment by dylan.mc...@gmail.com on 21 Jul 2008 at 4:00

[GoogleCodeExporter]

I am very happy to hear that. Thanks a lot!!!

Original comment by y.fujita...@gmail.com on 22 Jul 2008 at 12:12

• Changed state: Fixed