[sword-devel] CVS Tree lost icu-sword branches
Sun, 4 Apr 2004 06:59:51 -0600
Hi Troy / Jonathan,
Just wondering ... have the references to CBuilder CodeGuard (from the
Enterprise version?) been removed? Seems like at one point that was a
residual stumbling block for people using the personal version.
----- Original Message -----
From: "Troy A. Griffitts" <firstname.lastname@example.org>
Sent: Sunday, April 04, 2004 2:38 AM
Subject: Re: [sword-devel] CVS Tree lost icu-sword branches
> My apologies for not mentioning this early. I recall that you said you
> found an icu dll in your OpenOffice directory? This cannot be a usable
> dll for sword IMO. The correct DLL can be found on our alpha under the
> link 'misc alpha files'. Chris Little maintains ICU and I think this
> can only be built currently with the MSVC compiler. I believe it is
> merely the data for ICU. If you cannot find the correct file, I would
> ask Chris which one you should use. This may very well be the problem.
> [more below]
> Jonathan Mickelson wrote:
> > Troy,
> > I had commented out the offending lines in BibleCS and though it
> > executed, all it did was show the splash screen, hide it, show the
> > outline of the main form and immediately terminated. I stepped thru the
> > program, but it did not make sense as to why it quit
> > application->run(). There were no exceptions and no errors. In less
> > than 20 steps, the program was finished executing. I'm still too
> > unfamiliar with the 30+ million lines of code the BCB6 said it compiled
> > to produce a 4+MB executable (I know a lot of that was header files).
> > I've read the many forum postings concerning BCB6 issues and the one
> > about the two good fellows who got BCB6 to work properly. I had hoped
> > that the code would have been tagged in the CVS tree as a milestone, so
> > I went looking for that build knowing that the current CVS tree was in
> > flux. That's how I found "the rest of the story" concerning icu-sword.
> > From the forum postings, I got the impression that BCB6 is the unwanted
> > stepchild of the Sword project. I don't mind hanging in there with BCB6
> > if there's an actual desire to see it come to fruition. However, It
> > will take some coordination.
> BCB6 should work just fine. I just never upgraded. I was waiting for
> BCB7 but it just seems like that may never exist. I'm still using BCB5
> mostly but have a free version BCB6 on a VMWare image that can use. I'm
> trying a build right now...
> > On the Linux side, I presume your are using KDE, KDevelop, Qt3 based on
> > web site information for BibleTime. Has that changed?
> I'm sure that is probably what the BibleTime guys use. They can speak
> better than me. I use Fedora Core 1, Gnome2, Enlightenment, and mostly
> > I'm impressed with the different projects that are going on
> > (multiplatform development, language modules, flash cards/tutors,
> > serious web interface development and various side projects that I'm
> > picking up on). I can tell that development is moving at a break neck
> > pace with a lot of excitement behind the scenes about the progress that
> > is being made on some really cool pet projects that really raise the bar
> > for this genre of software. Is there a list of the various projects and
> > who is working on them? BTW, who has responsibility for BibleCS? I'm
> > guessing that Joachim has responsibility for BibleTime.
> > I'm surprised at Borland for making BCB6 non-backwards-compatible with
> > BCB5. Any insight as the the major differences and pifalls?
> It's pretty backwards-compatible. It's really not much more than just
> keeping the project files in sync, since I don't use it all the time.
> I'd be glad for you to own the BCB6 project files. :)
> > Finally, I've taken a look at BibleStudy's source code using wxWindows
> > last updated April 2003. Is there any thought of having a basic
> > cross-platform frontend for sword among the core developers -OR- is it
> > preferred that everyone have the privilege of creating their own
> > front-end for the platforms they want to use. I haven't figured out the
> > end-goal yet. My observations are that all the slick, frontline
> > development is put into BibleTime with all the other applications
> > playing catch up. This implies there is intimate development
> > cooperation between BibleTime developers and SWORD API developers.
> > Wonderful! It makes for a great team. If this is indeed the case, what
> > do we do with BibleCS?
> Well, believe it or not (the BibleTime guys may not agree :) ), BibleCS
> usually uses more of the latest features than any of the frontends.
> It's not very asthetically pleasing, but I think has the same
> functionality as most of the other frontends. I originally wrote
> BibleCS, and it is good for me to have the experience of building a
> complex client of the API. I would very much like to have other major
> contributors to the Windows frontend. You can also find a prototype of
> a new user interface on the alpha page and probably in CVS somewhere.
> It's hard for me to get time to work on the Windows side.
> > If BibleTime is the avant guard of Crosswire's development, it would
> > behoove us keep permanent milestones of working libraries that the rest
> > of the Crosswire family can code to while BibleTime forges ahead with
> > the latest and greatest. This would allows hobbyists and other
> > researchers to develop their own custom application based on a stable
> > API set and upgrade later if and when they desire. How do you envision
> > all this working? Is there already a model in place that I've missed?
> > I'm really intrigued to know the inside scoop cause I'm picking up a lot
> > of excited activity on the radar.
> We usually keep source releases of all public version available on our
> ftp site: ftp://ftp.crosswire.org/pub/sword/source/
> Any on these should be stable to build clients against.
> Coordination is usually done live at: irc.freenode.net #sword
> And on each project's mailing list:
> MacSword and GnomeSword have lists on SF, I think.
> I hope this answers some of your questions, and that more people might
> reply to others.
> sword-devel mailing list