World: r3wp
[Red] Red language group
older newer | first last |
Kaj 7-Aug-2011 [2825x2] | It's the same bug that crashes cURL. It doesn't even get to opening the window, it prevents the SDL library from properly initialising |
When I add C library initialisation, SDL also initialises | |
Dockimbel 7-Aug-2011 [2827] | So it works out of the box on Linux but not on Syllable OS? |
Kaj 7-Aug-2011 [2828] | Yes, exactly like cURL |
Dockimbel 7-Aug-2011 [2829] | Have you investigated what is causing that difference on Syllable? |
Kaj 7-Aug-2011 [2830] | Not deeper than the LSB spec you found that lists possible functionality that any implementation can put in there |
Dockimbel 7-Aug-2011 [2831] | I mean: it looks like Syllable is not initializing the C lib the same way as other OS. |
Kaj 7-Aug-2011 [2832] | It would be surprising if it would, because this is very OS specific |
Dockimbel 7-Aug-2011 [2833] | Maybe you could find out what is causing that in Syllable and fix it? |
Kaj 7-Aug-2011 [2834x2] | There's nothing to fix. This is how it's implemented in Syllable's case |
Most of the adaptation of Syllable to standard software is in the C library. This is where the handshake between kernel and userland is done in common systems | |
Dockimbel 7-Aug-2011 [2836] | Well, then I am ok to add that libC init call in Syllable's runtime part (runtime/syllable.reds) as long as it doesn't need any other OS-specific additions in Red/System. |
Kaj 7-Aug-2011 [2837x2] | OK, thanks. It will also be needed on other systems, though, in forms specific to their C library. I've already included the deviations for Linux |
I haven't tested it yet, but I'm sure that without it, things such as get-env won't work | |
Dockimbel 7-Aug-2011 [2839] | I won't add it for other OS until it becomes mandatory for Red. |
Kaj 7-Aug-2011 [2840] | That's fair enough. I also like to postpone things until they're proven necessary :-) |
Dockimbel 7-Aug-2011 [2841] | BTW, this call could prevent Red from working on Syllable as that libC init is a blackbox. I especially fear a clash with Red and libC thread init routines. |
Kaj 7-Aug-2011 [2842x2] | It would be much worse if it weren't a black box. :-) You'd have to duplicate all the initialisations done in there for all systems and all C libraries and all versions of them |
If Red threading will be incompatible with C libraries, it will be usable in pure Red programs, but not in any program that uses any library that uses the C library | |
Dockimbel 7-Aug-2011 [2844] | Some incompatibilities with multi-threaded C libs might happen, but mono-threaded C libs should be safe. |
Kaj 7-Aug-2011 [2845] | cURL is mono-threaded, though |
Dockimbel 7-Aug-2011 [2846] | I don't know exactly what libC is doing in its thread init routine nor I have time now to investigate that. We'll see if any issue arises once Red gets multithreading support. |
Kaj 7-Aug-2011 [2847] | I'm fairly sure they would be fixable |
Dockimbel 7-Aug-2011 [2848] | BTW, Red will probably be distributed both as a standalone exe and as a dynamic library. So if people wants or needs to deeply rely on C libs, they could embed Red in a C project easily. |
Kaj 7-Aug-2011 [2849] | For Syllable Desktop I have to disable CDROM support, because we have only recently ported that to SDL. But now I get the windows, both with and without border |
Pekr 8-Aug-2011 [2850] | it should not be difficult to port View engine, no? SDL is surely very welcomed .... |
Kaj 8-Aug-2011 [2851x2] | The View engine is under unknown licence, so it's useless |
Do you see them as competing? If the View engine were BSD, I would still put it on top of SDL | |
Henrik 8-Aug-2011 [2853] | Isn't it enough to just make it behave like View? |
Kaj 8-Aug-2011 [2854x2] | I hope so |
There will always be subtle differences, though, so like Red, it won't be fully compatible | |
Henrik 8-Aug-2011 [2856] | View is events, basic graphics structure, DRAW and window handling, as I see it. |
Kaj 8-Aug-2011 [2857x3] | All that except DRAW is done by SDL. For extra functionality, there are many libraries |
This is what I want to make: | |
http://www.youtube.com/watch?v=8nA4WaJTc7Y&NR=1 | |
Endo 8-Aug-2011 [2860] | GEOS on C64 is much better :) http://www.youtube.com/watch?v=j1Mnvead8Tc |
Dockimbel 8-Aug-2011 [2861x4] | Support for functions with variable number of arguments just released. |
New attribute 'variadic documentation: http://static.red-lang.org/red-system-specs.html#section-6.3.3 | |
New attribute 'typeinfo documentation: http://static.red-lang.org/red-system-specs.html#section-6.3.4 | |
'typeinfo is a variant of 'variadic that provides also argument type at runtime. | |
Kaj 8-Aug-2011 [2865x2] | Shouldn't that be type-info? |
Endo, it's in development | |
Dockimbel 8-Aug-2011 [2867x3] | I want attributes to be single words, so as callback is also wrote without an hyphen, I decided to do the same for typeinfo. |
I might change 'typeinfo to 'type or 'info only, but these words alone are too common, so, not enough accurate. | |
If someone has a better proposition for 'typeinfo, I am ready to replace it. | |
Kaj 8-Aug-2011 [2870x2] | Callback is written like that in English (otherwise it would be a verb instead of a noun), but type info is written as two words |
TYPED would also do | |
Gregg 8-Aug-2011 [2872] | TYPED sounds good, based on a glance at the docs. |
Kaj 8-Aug-2011 [2873x2] | I can now draw a pixel on an SDL screen in a format of 32 bits per pixel |
Humble beginnings, but I've had a series of issues already | |
older newer | first last |