Re: Refactoring codepage converters - Help for testing needed

This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--===============2198953237454617076==
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature";
	boundary="------------enig2208F83292BA1A6187969409"

This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--------------enig2208F83292BA1A6187969409
Content-Type: text/plain; charset=ISO-8859-15
Content-Transfer-Encoding: quoted-printable

Hi,

due to the rain today I've got a lot of the unit tests fixed ;-)
Only codepage 278 and 424 are making trouble.
I tend to trust 278 regardless of failing tests, cause I've compared
the tables manually.

Next time I will refactor the classes so that the new implementations
take place and the old ones are still accessible for testing purposes.

Regards
 Martin

Martin Kirst schrieb:
> Hi @all,
>=20
> the last couple of days I've started refactoring the codepage converter=
s
> within TN5250j. The simply reason doing so, was that I didn't catched u=
p
> how to add a new converter like requested in the SF.net forum and the
> big mess how monolithic the current implementation is.
>=20
> I've already commited new implementations, JUnit tests
> and micro benchmarks. Looking at the micro benchmarks,
> the new implementations are _magnitudes faster_.
>=20
> At the moment both implementations existing side by side but the new on=
e
> is not used.
>=20
> Please have a look at "migrating_cp_converters.txt" which will explain
> more technical details.
>=20
> Often I have to figure out, what implementation is wrong and why,
> by comparing to the JTOpen reference or by reading some HTMLs.
> Sometimes the old implementations where not correct.
>=20
> I would appreciate any help on fixing the last errors in the unit tests=
=2E
>=20
> Regards
>  Martin
>=20
>=20
>=20
> -----------------------------------------------------------------------=
-
>=20
> -----------------------------------------------------------------------=
-------
> Crystal Reports - New Free Runtime and 30 Day Trial
> Check out the new simplified licensing option that enables=20
> unlimited royalty-free distribution of the report engine=20
> for externally facing server and web deployment.=20
> http://p.sf.net/sfu/businessobjects
>=20
>=20
> -----------------------------------------------------------------------=
-
>=20
> _______________________________________________
> Tn5250j-develop mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/tn5250j-develop



--------------enig2208F83292BA1A6187969409
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkoQgSoACgkQefmD4St1GNEmcQCfWnJ/dTjFeLvJh8H0YvgW9eG/
gMIAoMRnptjwjQFJ2+yDvfRmXUR4rS36
=v0Lz
-----END PGP SIGNATURE-----

--------------enig2208F83292BA1A6187969409--


--===============2198953237454617076==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables 
unlimited royalty-free distribution of the report engine 
for externally facing server and web deployment. 
http://p.sf.net/sfu/businessobjects
--===============2198953237454617076==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Tn5250j-develop mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/tn5250j-develop

--===============2198953237454617076==--