Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Josh @ Dreamland

2161
Announcements / Re: New Interests
« on: April 02, 2010, 04:31:03 PM »
Then it will be have an integrated circuit for garbage collection.

2162
Announcements / Re: Metaphase
« on: April 02, 2010, 12:34:25 PM »
*chuckles a little*

2163
Announcements / Re: Metaphase
« on: April 02, 2010, 12:03:43 PM »
Serious or no, they're still a consideration. I can't think of a good reason to support them, even if ISO still finds them useful.

2164
Announcements / Re: New Interests
« on: April 02, 2010, 11:03:19 AM »
Nothing like a little team work. ^_^

2165
Announcements / Re: Metaphase
« on: April 02, 2010, 10:52:44 AM »
That textbox will just trick users into thinking something wasn't commented, when it is. Hopefully a checkbox would filter that out from happening, but...

You know, I myself would happily leave trigraphs to fall to antiquity. It's true; basically no one does use them. If they did, maybe GCC wouldn't make you pass a flag to enable them, rather than having you pass one to disable them.

We're pretty close to a release. Too much so to be thinking about this sort of thing now, anyway. Bring it back up when you have someone that actually wants them. Or when Ism and I are long caught up with everything else and are looking bored.

2166
Announcements / Re: Metaphase
« on: April 02, 2010, 10:42:05 AM »
Ism and I discussed having an option for them as G++ does. I can devote fifteen minutes of my day to adding them in conditionally as long as she can support them in the code editor. I don't want any surprises, particularly involving escaped newlines outside of macros.

We've already removed preprocessors from them for simplicity (though you can use macros as defined elsewhere, of course).
If she adds the option and the highlighting (for escaped comments at the very least), they'll be in. As it stands, we're already behind on C++-strings, which is why they too are supported optionally. The code my end for either is simple; hers, not so much.

2167
Announcements / Re: Metaphase
« on: April 02, 2010, 10:18:30 AM »
Oh, you got me. Guess the project's a bust. Everyone, go home; I'll postpone the release until mid-January of next year and make sure I support things we don't need.

Like I said, GCC warns on them. It ignores them in doing so. You have to flip a special switch to enable them.

2168
Announcements / Re: Metaphase
« on: April 02, 2010, 10:13:39 AM »
Bhahahahhaha, C does those.
It warns about them, actually. You can tell it to ignore the warnings and process them.
And no, begin and end are lousy enough. Find me a really nice trigraph and I'll support it.
As in, one with a useful application that C++ wouldn't have something else for.

2169
Announcements / Re: Metaphase
« on: April 02, 2010, 10:12:28 AM »
No.
The implication is that they are like "#" in GML--stored like anything else, and dealt with at run.

2170
Announcements / Re: Metaphase
« on: April 02, 2010, 10:11:10 AM »
Well, I'm reading that "digraphs and trigraphs are sequences of two and three characters respectively which are interpreted as one character by the programming language." That sounds just like those two to me.

2171
Announcements / Re: Metaphase
« on: April 02, 2010, 10:05:40 AM »
You're right. For unicode, short and wchar_t should be enough. I have no idea why anyone would need a third character.

2172
Announcements / Re: Metaphase
« on: April 02, 2010, 09:00:07 AM »
All you have to do is post something if a piece of code won't compile. I will do my best to print helpful info about what may have went wrong, but as long as you are testing simple codes, we should get to the bottom of it relatively painlessly.

I'm just afraid a lot of things will go wrong, and there'll be reports everywhere.

@retep-

2173
Announcements / Re: New Interests
« on: April 02, 2010, 08:57:37 AM »
One of the nice things about ENIGMA is that people will leave knowing one of the most popular languages in the world.
That's why I added all these C++ layers without infringing on the GML ones. Users will slowly get accustomed to C++ features, and before you know it, they'll understand the system.

2174
Announcements / Metaphase
« on: April 01, 2010, 09:27:54 PM »
April Fool's has finally passed, with only moderate ripple effects, I might add.

A lot of insanely awesome stuff is going on right now, a lot of which I'll keep a surprise, but I'm happy to report that Ism can now successfully pass me resource data via the DLL. We are still missing a few things that R3 had support for, which I am trying to scramble up support for. Expect a testing phase commencing tomorrow. This should not overflow into the following day.

What that test will actually look like depends on a number of factors. Right now, I'm too excited to sleep. So the project has me for a while.

LGM has a bug, seemingly only on Windows, that we'll need to isolate and resolve. Isn't a very obtrusive bug and doesn't really fall on me, so.

Other things that need done with LGM include re-implementing the syntax check button now that ENIGMA's a DLL, and configuring it to inform me of changes in the C++ resource to avoid wasting an additional second during compile.

I think I may also implement the new instance system before I begin the test phase.

Also, keep this phase relatively quiet: I'm considering making it community-members only. Two reasons for this:
1) A huge part of the point of this is to alleviate stress caused by the release having a fatal flaw that has been overlooked.
2) I want the testing phase to be a small sample of people who can actually tell me what isn't working, or what's missing.

This testing phase will hopefully be both an opportunity to debug and to request small functions, maybe systems. While you test, I will be implementing sounds. Do not expect them from day 1.

In fact, since at this point I'm getting things that were done already in R3 working, expect the entire project to be a miserable piece of failure, and then be enamored by all the wonderful things it turns out that it can do.

Of course, if it doesn't do a particular wonderful thing, such as R3.0's inability to use non-numeral-prefixed constants (such as .5), and R2's lack of comments, DO REPORT THAT.

Again, though, keep it low for now. Release should be ready by the end of tomorrow. We'll see about getting something from Luda in that. I can't stress enough not to count on all the systems that will be in R4 to be that way immediately; many of them still need integration that is better done during the testing phase. Don't expect the bitmask system yet; Luda said he'd try to have the rectangle part of his design ready, though. Also, I'm looking at backgrounds and tiles right now, frowning. So, we'll see what this gets us.

2175
Announcements / Re: New Interests
« on: April 01, 2010, 07:12:48 PM »
*shrug*
Can't compete with age.

*fixes post count*