WinRT still advances
As people who follow @videolan know, we keep working quite a bit on our port to the WinRT platforms. It is probably the port where the most effort is spent on those days, and is probably the most difficult.
The good news is that we are improving quite a bit, and we are closer to having something on the store.
Precision about VLC ports
Some people said that we stopped working on WinRT to ship VLC on iOS… This is totally wrong, because those are not the same people working on it.
VLC development is closer to a bazaar than to a cathedral building, and while the core advances altogether, people, who are volunteers, work on what they want, and on the features they want. This is very often true for VLC modules, and this is even more true for the ports to the mobile platforms.
Notably:
- VLC for Android is mostly developed by Ludovic Fauvet, Sébastien Toque and Edward Wang;
- VLC for iOS is mostly the work of Felix Kühne and Gleb Pinigin;
- VLC for WinRT is mostly developed by Rafaël Carré, Geoffroy Couprie, Kellen and me.
Therefore, doing work on one platform does not slow down the development of other platforms ports.
On the contrary, porting VLC to more platforms improves the portability of VLC, and helps finding weird bugs or misdesigns that benefit all the other ports, when they are fixed!
For example, the work on OpenGL ES for Android helped the port to iOS. Or the work on WinRT helped the normal VLC for Windows.
WinRT calls
So, last time we spoke about our advance, we had to fix 16 forbidden calls to 4 Windows dll, from 5 dlls and all the socket code. And those required WinRT direct calls from C.
We fixed most of the issues, including the WinRT static calls, meaning that we rewrote a lot of idl, idl tools and header files. We are at 5 forbidden calls (3 tonight, I hope), to 1 Windows dll, from only the VLC core. We still have the socket code to fix though.
The remaining calls are on the threading initialization, and so far, we are not able to create those WinRT objects from a C codebase. We are looking at alternatives, including using a C++ library to work-around this issue.
For the socket code, we have an idea for that too, and I hope I can share it soon to you.
As soon as we are down to 0 calls, we can upload something on the store, for the backers to test it.
Goodies
We’ve done another round of sending of goodies.
Therefore, if you had a certificate, a key-holder, a mug or a cone, you should have received them before the end of the week.
If you had a t-shirt or a hoodie, they might not have reached you yet. Note that, if you had one of these, the keyholder and certificate will arrive at the same time.
If you have not received your goodies, please e-mail us about it, to the email where you received your confirmation from. If you are at a total loss, please mail me or contact me :)
Sorry again for the delay, but we’re doing the best we can, so far. Have fun!