> I have managed to put a small group of developers together to > work on what amounts to a combined wish list. We have all used > various toolkits in the past, but have always wanted to use FLTK. > However, the thing we all agree on is that there are a few things > we would like FLTK to do differently. [...] Our discussion also > resulted in some initial tasks, which I've listed below.
> 1. CMake - We all use CMake so the first order of business is to > create a CMake build. There is already [at least?] one STR requesting a correct CMake configuration. It would be really great if you could provide input that would help to resolve that. > 2. Doxygen - We plan to add Doxygen headers throughout the source. > We believe that it will create a valuable reference resource. It > will also be a good way to get more familiar with the current source. One of the major goals of the current fltk-1.3.x development is the move from hard-code separate documentation to generated documentation using Doxygen. The first implementation is almost complete, with just some tidying up required. Obviously there is room for improvement. > 3. Platform layer - Analyze and rework (where necessary) any > platform specific code (including #ifdef) and push it down to the > platform layer rewriting in c, where necessary. There's always room for improvement, especially cleaning out old code to do with old platform/compiler combinations. > 4. SDL - Add SDL support as a test of the platform layer design. > This should be possible without making any changes above the > platform layer. This might be a bit more problematic. FLTK has traditionally shied away from the mandatory inclusion of external libraries (including the standard C++ library) because there are users who require the Fast and Light aspects of FLTK for their embedded systems. However you really need to have some feedback from some of those developers whether this is still the same killer requirement that it used to be. > These 4 items would represent the first milestone. We intend to > make snapshots available for each of the above items as they are > completed. > > Note: I should probably mention that we don't intend to include > FLUID in this work. [...] FLUID can be a bit quirky, but it is a very powerful tool. It will never be a full environment like Visual Studio, Kdevelop or Eclipse but it is very good at what it does. If you use FLUID to develop GUI elements that are effectively abstract classes, you can then flesh these out in a main-stream coding environment tool. It does take a little discipline, and unfortunately isn't really described anywhere in the documentation. It comes up frequently in the forums. > Disclaimer: > > We/I have the utmost respect for the developers/creators of FLTK. > We are not doing this work because we believe our skills or ideas > to be any better than anyone else's. We are just a group of people > with a slighly different vision of how we would like to use FLTK. > We are stepping up to do the work ourselves, instead of expecting > someone else to do it for us. We do, and always will, give full > credit and our thanks to the FLTK development team for the excellent > work they have done. It would be very nice if our work could benefit > the FLTK project in some way. This is the world of open source, where you are free to take the code and modify it to scratch your own itch. But from what you have written above, three of your four goals above are already being implemented in the fltk1.3.x development, or are part of the future development. Therefore would it not make more sense to contribute to the FLTK-1.3.x development directly to improve the code base in line with your goals? The FLTK community has already seen that the FLTK2 fork ran out of steam, as did the unofficial 1.2 fork to add UTF-8 to FLTK1. Does efltk also count as a fork? At least two FLTK1 users had their own forks ring-fenced for their commercial products, where they had to backport the 1.2 changes. It is only now that FLTK1 has been able to merge these UTF-8 versions for the FLTK-1.3.x development. Cheers Duncan _______________________________________________ fltk mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk
