Re: [Tm4j-developers] [PATCH] Merge utility rewrite

[Kal A. <ka...@te...>]

But I am glad I did - this is a good start on rewriting the merge utility=
=2E=20
There is one other thing which I would like to do (perhaps you can do it=20
Florian ?), which is to make the application load topic maps using the=20
TopicMapProvider.addTopicMap() method, rather than the old=20
XTMBuilder/XTMParser way. This is for two reasons:

1) I tell folks that the source for these apps are a good way to learn ho=
w to=20
do basic things with TM4J (like reading and writing topic maps)
2) When we get support for alternate syntaxes (such as LTM) in TM4J, I wo=
uld=20
like that support to be hidden behind addTopicMap() - so using that metho=
d=20
would make the application "forward-compatible" with support for LTM.

However, this is definitely a vast improvement. I have no problems with a=
dding=20
jargs to the build tree as I have used it for other apps and I'm quite ha=
ppy=20
with its quality and usefulness.

Cheers,

Kal

On Wednesday 29 May 2002 18:16, Florian G. Haas wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hello!
>
> While in the process of putting together some functionality for
> XHTML-from-XTM generation -- which will be checked in Real Soon Now --,=
 I
> fiddled a bit with the Merge utility, which, as it happened, resulted i=
n a
> fairly complete re-write. :-)
>
> I used Steve Purcell's JArgs package
> (http://sourceforge.net/projects/jargs) and added some new options. JAr=
gs
> is a largely GNU-compatible command line parser allowing for specifying
> both short (-x) and long (--option) command line args, which makes it e=
asy
> to add new features without writing outrageously complicated parsing
> routines. The rewritten Merge utility still writes either to stdout or =
an
> output file, but now additionally supports reading from stdin so you ca=
n
> pipe output from a TM generator or the like to it. This still lacks a l=
ot
> of documentation, but perhaps one or two of you can make the time for
> taking a look at it. As usual, just a proposal -- no hard feelings if i=
t's
> rejected. Might be worthwhile to take a look at JArgs itself too: I'm a
> bit ambivalent about it, it works fine but uses all sorts of odd
> constructs such as nesting tons of inner classes and the like.
>
> When testing, apply the patch from the root of your TM4J CVS tree.
> After recompiling, invoke Merge with the -? or --help switch to see wha=
t
> options it now supports.
>
> I'd be happy about comments.
> Thanks in advance,
> - -- Florian
>
> - --
> Florian G. Haas <fg...@bk...>
>
> GnuPG public key: http://www.fgh.bkf.at/gpgkey.asc
> Key fingerprint:  3C24 E021 33B1 9E98 B94B  2F21 860A B693 1B33 BB3C
>
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.0.6 (GNU/Linux)
> Comment: For info see http://www.gnupg.org
>
> iD8DBQE89RsJhgq2kxszuzwRAsTvAJ9Feh44cAvW8RhJ3C525l7PjK+5ewCgmBQz
> PWxOu9o3Hj0FljkAGMcsr8Y=3D
> =3DHShR
> -----END PGP SIGNATURE-----

--=20
Kal Ahmed, techquila.com
XML and Topic Map Consultancy

e: ka...@te...
p: +44 7968 529531
w: www.techquila.com