Re: [Chessdb-users] Tightly bound to Tk?

[Will M. <wil...@us...>]

I don't know how the duplicate message got through, especially after such a=
 long delay.=0A=0AI've actually started working through the surgery of un-T=
cl-ing the code, and it's really not too bad. So much of the code in there =
is dedicated to dealing with parsing the Tcl argument lists that once you e=
xtract just the code that actually does the work it comes out looking a lot=
 cleaner. Then you just wrap the C interface back in the Tcl to recreate th=
e Tcl interface so that the old GUI will still work.=0A=0AIf I can find the=
 time to finish this first stage of creating a new GUI, then I'll share the=
 results. I could very well just encounter the same problems you describe b=
elow, though. We'll see.=0A=0ACheers,=0AWill=0A=0A----- Original Message --=
--=0AFrom: Michal Rudolf <mr...@kd...>=0ATo: chessdb-users@lists.=
sourceforge.net=0ASent: Sunday, July 15, 2007 5:46:58 PM=0ASubject: Re: [Ch=
essdb-users] Tightly bound to Tk?=0A=0AWill Mason,  niedziela, 8 lipca 2007=
:=0A>Hello, list,=0A>I'm new to the list, so please forgive me if I'm rehas=
hing old news.=0A>=0A>ChessDB seemed to be exactly what I was looking for, =
so I downloaded the=0A> source and built it on Mac OS X. The first thing I =
noticed was that it did=0A> exactly what I was looking for a chess database=
 to do, and the second thing=0A> I noticed was that the UI was amazingly ug=
ly. I poked around in the source=0A> code for a while and came up with the =
following idea.=0A=0A>I think the code in src/tkchessdb.cpp should be refac=
tored. Currently, the=0A> only real interface to the DB code is through Tcl=
/Tk. I think the=0A> primordial interface to the code should be in C, from =
which Tcl, Tk, Java,=0A> Ruby, Python and a host of other interfaces could =
be derived. From what=0A> I've seen of the code this would not be an amazin=
gly difficult thing to do:=0A> factor out the C (I know it's C++, but it re=
ally bears a greater=0A> resemblance to C code), expose the interface descr=
ibed in=0A> http://chessdb.sourceforge.net/tutorial/programmers-reference.p=
hp as a=0A> C-only interface, rebuild the Tcl/Tk interface on top of that, =
and finally=0A> build whatever other interfaces one might want. I, for one,=
 would build a=0A> Java interface, first exposing the C interface described=
 above through JNI.=0A>=0A>I write this suggestion to the list rather than =
simply embarking upon its=0A> implementation myself because I just don't ha=
ve time to undertake this=0A> project right now. Or I should say that if I =
did undertake such a project=0A> it would proceed at an extremely slow pace=
 due to my time restrictions.=0A>=0A>I thought I would throw this idea out =
to see if it might appeal to anyone.=0A> Thanks for your time, Will Mason=
=0AWe started similar project called NewScid, trying to use Scid as underly=
ing =0Adatabase and Java for GUI. But we encountered various problems that =
=0Adiscouraged us and convinced us to start another project instead (ChessX=
).=0A=0AOf course, somebody else may have more luck, but I am not sure if i=
t is not =0Aworth an effort  - Scid code seems quite tightly integrated wit=
h Tcl.=0A=0A-- =0AMichal Rudolf=0A=0A--------------------------------------=
-----------------------------------=0AThis SF.net email is sponsored by DB2=
 Express=0ADownload DB2 Express C - the FREE version of DB2 express and tak=
e=0Acontrol of your XML. No limits. Just data. Click to get it now.=0Ahttp:=
//sourceforge.net/powerbar/db2/=0A_________________________________________=
______=0AChessdb-users mailing lis...@li...=
=0Ahttps://lists.sourceforge.net/lists/listinfo/chessdb-users=0A=0A=0A=0A=
=0A