Use tre wide char

I know some things have been thrown around about various things 
concerning unicode and crm. I have had some requests myself. UTF-32 and 
UTF-8 versions are the same other than UTF-32 converts to UTF-32 from 
UTF-8 at the start and uses all wide char versions of tre. This fixed 
some memory corruption I was seeing with valgrind memcheck on Athlon 
(32bit) and Athlon X2 (64). Notice the huge difference. Please, do not 
take the crm version at face value, several fixes went into the UTF-8 
version that sped it up greatly. However, look closely at UTF-32 vs. 
UTF-8. The more regex the faster the comparison is. This is because the 
non-wide char tre stuff actually converts to UTF-32 (wchar_t) on every call.

New UTF-32 version
time ./dehtml > /tmp/duh2.html

real    0m0.070s
user    0m0.068s
sys    0m0.001s

Old UTF-8 version
time ./dehtml-utf8 > /tmp/duh2.html

real    0m0.118s
user    0m0.107s
sys    0m0.001s

Original CRM version
time crm myhtml.crm </tmp/duh.html  > /tmp/duh3.html

real    0m0.258s
user    0m0.253s
sys    0m0.002s

Trever

------------------------------------------------------------------------------
Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
-OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
-Strategies to boost innovation and cut costs with open source participation
-Receive a $600 discount off the registration fee with the source code: SFAD
http://p.sf.net/sfu/XcvMzF8H

_______________________________________________
Crm114-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/crm114-general

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmv5MEACgkQTmpiivh5HXMhGgCfeTZGKFFC2j4vAOwmaafDQgP2
6qgAnjWAbFQUukQpTfVJyO7+nVuMNveo
=YSXh
-----END PGP SIGNATURE-----