[SciPy-user] Weave on W2K problems

[eric jones]

Hey Chris,

> I am trying to set up weave/scipy on a very fast machine,
unfortunately
> running W2K, rather than *NIX, and am having a couple of problems with
> weave.  First, although the scipy binary puts its weave components in
> $PYTHONHOME\Lib\site-packages as I would expect, the full weave binary
> puts its stuff in $PYTHONHOME.  

This may be because the standalone version was built with 2.1.3 and I
believe it defaulted to putting things in $PYTHONHOME on windows
distributions.  This was corrected in Python2.2 (which is used to build
SciPy).

> Second, though I have the cygwin gcc
> compiler (with g++ also) installed and listed first on my user path, I
get
> the following error claiming g++ cannot be found.
> 
> ======================================================================
> ERROR: result[1:-1,1:-1] = (b[1:-1,1:-1] + b[2:,1:-1] + b[:-2,1:-1]
> ----------------------------------------------------------------------
> Traceback (most recent call last):
>   File
"T:\Python22\Lib\site-packages\weave\tests\test_blitz_tools.py",
> line 150, in check_5point_avg_2d
>     self.generic_2d(expr)
>   File
"T:\Python22\Lib\site-packages\weave\tests\test_blitz_tools.py",
> line 124, in generic_2d
>     mod_location)
>   File
"T:\Python22\Lib\site-packages\weave\tests\test_blitz_tools.py",
> line 80, in generic_test
>     blitz_tools.blitz(expr,arg_dict,{},verbose=0) #,
>   File "T:\Python22\Lib\site-packages\weave\blitz_tools.py", line 72,
in
> blitz
>     type_converters = converters.blitz,
>   File "T:\Python22\Lib\site-packages\weave\inline_tools.py", line
439, in
> compile_function
>     verbose=verbose, **kw)
>   File "T:\Python22\Lib\site-packages\weave\ext_tools.py", line 332,
in
> compile
>     verbose = verbose, **kw)
>   File "T:\Python22\Lib\site-packages\weave\build_tools.py", line 221,
in
> build_extension
>     setup(name = module_name, ext_modules = [ext],verbose=verb)
>   File "T:\Python22\lib\distutils\core.py", line 157, in setup
>     raise SystemExit, "error: " + str(msg)
> CompileError: error: command 'g++' failed: No such file or directory
> 
> ----------------------------------------------------------------------
> Ran 184 tests in 250.797s
> 
> FAILED (errors=1)
> <unittest.TextTestRunner instance at 0x00D90AD0>
> 
> I need this stuff to work in order to make my dynamic optimizations
run
> faster, so any assistance is most welcome.

It looks like all the weave tests passed accept for the one blitz test.
If your not planning on using blitz, then I think you may be OK.

I just tried running weave with my cygwin installation with:

----------------------------------------------------------------------
Ran 184 tests in 74.055s

FAILED (failures=1, errors=22)

The blitz() test failed but not like yours.  I actually got incorrect
results.  The 22 errors were all of the MSC++ tests failing because they
couldn't find the C++ file they were asked to compile:

sc_abccabed66f253b594f63a36ae4b0cdb0.cpp
fatal error C1083: Cannot open source file:
'/cygdrive/c/DOCUME~1/eric/LOCALS~1/
Temp/@1336.2/sc_abccabed66f253b594f63a36ae4b0cdb0.cpp': No such file or
directory
E

I'll have to explore these.

As for the error of not finding g++, I don't know why it is happening.
If it is in the path and you can execute the following:

eric at ERICDESKTOP ~/wrk
$ python
Python 2.2.1 (#1, Jun 25 2002, 10:55:46)
[GCC 2.95.3-5 (cygwin special)] on cygwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import os
>>> os.system('g++')
g++: No input files
256

weave shouldn't complain -- especially since it appears to be finding it
in many of the other 184 tests.

Hope something here helps,
eric




> 
> Thanks,
> Chris Fonnesbeck
> --
> ______________________________________________
> http://www.linuxmail.org/
> Now with POP3/IMAP access for only US$19.95/yr
> 
> Powered by Outblaze
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user