(1:30 AM EST) I ran into a very difficult snag when I was on the home stretch regarding getting vector of void * to work happily if one of the objects being pointed to is a vector of char. I think I've figured out a way to work around the problem but it will require a bit of pointer magic (and heavy thinking) on my part. If that doesn't work, then I have a somewhat less elegant solution which I know will. Once this is over with, I will post 0.0.3.


9/28/2000

(12:30 AM EST) Major progress today. I'm now on the home stretch for the 0.0.3 release, which will provide read ability for formatted data (Styles, Info, Events). Look for a module release on sourceforge some time this afternoon.

0.0.4 will feature loading and saving for embedded data (Graphics, Fonts). 0.0.5 will add saving for formatted data. 0.0.6 will be code clean up only. Then we go on to the beta sets before the 1.0.0 release.


9/26/2000

(8:25 PM EST) Last night I sat down and began recoding the section parser from scratch. At this moment, I'm pleased to announce that its almost done. The development snapshot that I released about a week ago is completely invalid. The new code is actually good looking, functional, and reasonably efficient.

Hopefully, I'll have a working version of the formated data parser up for download within a few days. Once that is done, all that is left is the embedded data parser (for encoded binary files embedded into the ascii). With a little luck, Nevu will be ready for beta by the end of the month.

(11:00 PM EST) I written up the new layout for the sssm module. Basically, the base package will include two classes: ssa400support, which holds static support functions, and ssa400script, which holds the data for the SSAv4 file that you are working with. The 1.0.0 release will contain the following files:

• ssa400support.cpp
  Contains the SSAv4 support class code

• ssa400support.h
  Contains the SSAv4 support class prototypes and defines

• ssa400script.cpp
  Contins the SSAv4 script class code

• ssa400script.h
  Contains the SSAv4 script class prototypes and defines

• ssa400globals.h
  Contains defines used throughout the SSAv4 classes and various debug/porting options

• demo.cpp
  Demonstration code showing the use of the sssm module

• Makefile
  Makes/cleans/removes the demo.cpp program

• README
  The documentation for the SSSM module

For the 1.0.0 release, there will also be a seperate test suite available which will allow you to test if all the code is working the way it should. While sssm is in the beta stage, this test suite will be included in the tarballs.


9/18/2000

(12:00 PM EST) Released a development snapshot of my work on 0.0.3. You can get it from the downloads section on the main Nevu page.


9/17/2000

(10:15 PM EST) Didn't get as much done as I would have liked last time I worked on the code. I will be doing some coding tonight and a large part of the day tommorow. Work towards the 0.0.3 release is slow but steady.

Once I finish my work on SSSM, I think I will try my hand at creating a video codec for linux. I'll hunt down some worthy, but unsupported format and implement the code for it.


9/10/2000

(10:00 AM EST) I haven't had time to work on the the sssm module recently. I have been busy moving from California to New York--where I am now attending college. I plan to spend the rest of today in the local Computer Science Lab working on the 0.0.3 release, and I will hopefully be mostly done by tonight. Check the downloads section around midnight for a snapshot of my work.

I've decided to hold off the undo/redo system till the 2.0.0 release. Basically, I have enough on my plate already. My goal for the 0.0.4 release is to have the new read system ready for everything except for the ascii->binary conversion. 0.0.4 should also have full support for the Format: lines, which will mean that we will actually be complying to the specs. :)

Oh, one more thing I forgot to mention . . . I now have access to Ultrasparcs running Solaris 8. That means my code will be testing on this platform as well. Additionally, once I am in the final betas before the 1.0.0 release, I will apply for access to Sourceforges IA64 machines, to make sure that my code works on those as well. Finally, I will also be soliciting people with various CPUs/platforms to test my code as well, so that I can be completely confident that it will work/compile on all platforms. Basically, I want my code to be 100% bugfree (hehe). That means that I'll have to make a series of scripts that will automate the testing process.


8/23/2000

(10:34 PM PST) Didn't get the CVS working a couple of days ago. I've decided not to use CVS for now, since I'm the only developer and nobody else has expressed any interested in this project. Tonight I'm starting the planning for the pre-alpha sssm 0.0.3 release. I'm basically working my way through the docs to figure out what data types (structs) I need to create, and which existing ones I can use for various SSSM variables. One thing that I'm a little worried about is portability since, IIRC, C and C++ do not actually guarantee the sizes of the long and short datatypes. For now, however, I'm not going to worry too much about it--I'll leave that for the 0.0.4 release.

I'm also going to make an online reference for the sssm module.


8/15/2000

(8:15 PM PST) The project was accepting into Sourceforge as I had hoped. I tossed together a spartan site for the Nevu project, which SSSM will now be a part of. I also released version 0.0.2 of my SSSM code to the public, even though I was planning on dumping all of that code and restarting from scratch. I will probably spend the rest of tonight getting acquainted with CVS, and then I will start on a new framework for the ssa400Script class.

Before I forget to mention it, I plan on making good use of the some of the spiffy resources that Sourceforge makes available. I will be able to test the SSSM module on several flavors of Linux, BSD, as well as Linux running on Intel's IA64 chip. The last is particularly important since my new code will start doing bit shifting, and things might get funky on a 64-bit architecture (as opposed to the 32-bit one I'm developing on). I will have to be especially careful to make sure that my code is portable across all platform, since C and C++ do not make any real guarantees about the bit-sizes of variables.

(3:00 AM PST) Today I registered SSSM at sourceforge as part of the Nevu project. Hopefully, the project should be approved within a couple of hours. Once thats all said and done, I will begin work on SSSM 0.0.3, which is basically a rewrite from scratch for all the existing code. The reasons for the rewrite? First, the existing code is ugly and inefficient. Secondly, I am not using the Format: line, which puts my existing code out of compliance with the SSAv4 spec. Once I get my code to the point where I can successfuly load SSAv4 files, I will contact the author of Substation Alpha and attempt to get a link to my code placed on his page.