PA Alpha Build 51531

[/quote]
I really haven't been paying that much attention. Working nights has really restricted my time for play right now. Hoping to concentrate on pa a little more in the future, bit right now my butt is kicked. I do wanna work on my "mod" though, especially since nothing in the base files gets overwritten on disc.

This is work in progress. I've been busting away on getting the real initial unit roster in, and pulled out some old data we had made many many months ago. You don't see the new work yet, because I'm keeping it shelved until it's ready for a full checkin. Once I'm done with initial implementation, I'll hand it off to Scathis for a balance pass, and ForrestTheWoods for some additional feature implementation to support needed behaviors in orbit.

i'm just happy our minions will walk in straight lines, and not loolygag around like a bunch of drunks.

....wait, robots can't get drunk..... unless......

...they have shiny metal you know what?^^

yay! so we may be able to get to the moon soon

Somebody said air-combat?

2000+++ planes on patrol
about 10 fps
memory usage 5.5 GB
_______________We need the "Armstrong Rocket" to get to the moon soon hehe

Video of 2000+++ planes in PA Alpha 51531, Patrolling, some combat, viewing.
Cant wait to get to the moon

http://www.twitch.tv/kennyball81/b/440441481

You can use jQuery to modify the html until html fragments are supported. See viewtopic.php?f=72&t=50277.

You're ******* crazy if you think I'm going to be replacing entire html pages through jquery.

The fact of the matter is: this "modding system" sucks, and I'm not using it. Deal with it. If you can't look at some of the mods I've released and understand why what you are suggesting isn't viable, then I suggest you stop attempting to give out "advice".

Furthermore I'm offended they even wasted their time providing such a thing to us.

Edit: then again, I should have seen this coming from the same people who have left the FPS-crippling resource bars in the game for over 4 weeks now.

If you need to replace an entire html page, you can just create a whole new one, add that new page to the list of pages to load using this mod system then redirect or link to it by modifying one line via jquery. More often than not though, mods only modify a small section of pages, and this is where the jquery method works well. As per below though, this is still only a temporary workaround.

Most of your mods are one or two line javascript changes, which are perfect for this modding system (Even longer javascript sections are easy to do with this system, it's only when html is involved that it becomes slightly more complex). Your lobby mod is more involved, but as the readme for the modding system says, html fragments are a feature that is under construction. But there are still ways of doing it now, and they are viable.

I have looked at your mods, and they are viable using the new system. No, it's not perfect yet, but neither is it as bad as you are making out.

This is unnecessary, as is your comment towards me. Please be civil.

I think you don't understand the system. Of the 2 mods I made I could convert both to use it perfectly. I also fixed up the unit count mod, you made. I didnt put in the timer because I didnt need it, it would have been possible. The unit count alone was like 3 or 4 lines of js.

EDIT:
I felt interested in how hard it is to implement i.e. your player list with the new system
Here is a quick attempt, that works:

Code:

$('.div_pre_game_header').append('<div id="onlineplayerslist" style="height: 400px; float:right; position: absolute; right: 10px; overflow: scroll;" data-bind="style: { height: serverListHeightString() }"></div> ');

model.updatePlayerList = function() {
	engine.asyncCall("ubernet.getCurrentGames").done(function(data) {
		var games = JSON.parse(data).Games;
		var onlinePlayersList = [];
		
		for (var i = 0; i < games.length; i++) {
			if (games[i].PlayerUberNames) {
				for (var j = 0; j < games[i].PlayerUberNames.length; j++) {
					if ($.inArray(games[i].PlayerUberNames[j], onlinePlayersList) == -1) {
						onlinePlayersList.push(games[i].PlayerUberNames[j]);
					}
				}
			}
		}
		
		onlinePlayersList.sort();
		var namePool = "";
		$.each(onlinePlayersList, function(i, val) {
			namePool += val + "<br />";
		});
		$('#onlineplayerslist').html(namePool);
	});
};

setInterval(model.updatePlayerList, 1000);

@Cola_Colin we need Linq support to deal with loops like:

Code:

for (var i = 0; i < games.length; i++) {
         if (games[i].PlayerUberNames) {
            for (var j = 0; j < games[i].PlayerUberNames.length; j++) {
               if ($.inArray(games[i].PlayerUberNames[j], onlinePlayersList) == -1) {
                  onlinePlayersList.push(games[i].PlayerUberNames[j]);
               }
            }
         }
      }

We don't need "support" as such unless there are licensing issues; there's nothing stopping the use of jlinq (or any other free javascript library) by simply including the library files in the mod, then importing them.

I just copied that code for the most part, it works like that so I don't care
Also as raevn said, nobody stops you from importing extra libraries.

Wow, what's with the hostility?

Just played a game, and sorry, but the improved path finding is really nothing to speak of right now. I wouldn't send out an update claiming that quite yet...

I think you don't understand javascript. It is as bad as I am making it out to be, if not worse. This is going to create some seriously massive scope chains and tons of unnecessary function calls. Yeah man, let's just duplicate this for loop, everything will be fineeeeeee :roll:

As stated earlier, there are a number of larger mods I am working on that, when implemented in this fashion, will completely cripple coherent's engine, and impact the framerate more than the already terrible resource bars. (still waiting for you guys to implement the fix that has been sitting in the bug tracker you never seem to look at or reference)

At the end of the day, I won't be using the modding system, and neither should you.

I laughed out loud.

I'm totally agree with you, but I think you quite too aggressive in your posts.

Not everyone here are web developers who have lot of experience with JS and it's problems. It's will be much useful if you just show some performance comparisons to people who argue with you, then just attacking them like that.

Uber say right there in the mod readme file that this not complete.

Yes without HTML support it's not as good as it could be, but I still find it a welcome improvement just for the fact we can develop in such a way we don't need to release updates every patch.

What I did in my now un-usable "live stats" mod to get an html chunck in without editing the markup was to write my html in a "partial" html file and pull that in via ajax (not sure if this will work anymore, hopefully it does) then bind knockout functions to it after I inserted it with jQuery.

I thought this was quite elegant considering what we've currently got. What do you think about that method sneakyness? If it's no good, how do you think Uber should be implementing HTML support? I'm just curious to what you think is the best method.

Edit: A post I wrote on the subject a while back as a suggestion: http://forums.uberent.com/forums/viewtopic.php?f=72&t=47732&p=741859#p741859

I'm personally against reinventing of wheel and I think that all HTML/JS/CSS editing should be done via diff patches. It's give lot of possibilities for automation in something like mods catalog.

Strengths:

1. Less custom tools required, everything already here, stable and working.
2. Easy detection of conflicts. You always sure that patch worked or failed, and it's will be easy to detect what's mods conflicts with each other automatically.
3. If new version of game is released mods compatibility can be tested automatically.
4. It's better for security, it's fairly easy to check what patch change.
5. One big patch can be spitted in few and you able to apply only part of changes.
6. It's easy to forbid changing of specific files globally.
7. It's possible to make patches automatically from given ZIP archive even if author of mod don't know how patches work. E.g modder upload his mod as ZIP and we can automatically make patch from it.

Weaknesses:

1. It's a bit less useful system while basic UI design is unstable.
2. Some hooks should be defined in HTML/CSS/JS. It's make everything easier for mods which only add code, but don't replace anything. It's should be denied to remove/edit these hooks in patches.
3. Probably basic HTML need to be formatted to make it easier to change some easy things.
4. Patches not suitable for JSON and art assets, so different system required.
5. It's all probably overkill for UI modding

Feel free to criticize my idea.

PS: Probably i'll done some design document and then create dedicated topic about it.