This issue was reported on the Flixel forums. At the moment, in the collision function between a bullet and something else, there's no way to tell what kind of bullet it is; i.e. what weapon it was fired from. The only way to get this is to use a different FlxGroup for each weapon's bullets, and then perform each collision check with each FlxGroup, passing in a different callback function. It'd potentially simplify user code if there was an easy way to retrieve the weapon from a bullet so we can tell what kind of weapon it is.

Hi,

I'm using timescale for some cool effects but when I set it to a value between zero and one, everything slows down except my bullets.

Problem is with how lifespan of bullets are calculated, I think. Because no matter how shrinked your time is, bullet's lifespan will be apparently calculated in real-time's time. For example my pistol' bullet's lifespan is 4 seconds but if you shrink time to 0.1, while bullet is still near the player, because the time is so slow in game, bullet will disappear after 4 seconds of real-time time.

since you probably won't update 1.9 anymore and will do a 2.0 version by xmas, what would u suggest to fix it?

-tnx

Despite what it says, FlxDelay abort() just calls stop(), so the callback is called, regardless.

I find it unnecessary that FlxDelay extends Sprite, when it can be made much more simple and efficient by adding a single static dispatcher that provides the event for all FlxDelay instances:
https://gist.github.com/1525275

Or another solution, why does it extend Sprite and listen to ENTER_FRAME instead of extending FlxPlugin and using it's "update" method?

An an additional idea for FlxDelay, the "newDuration" parameter of "reset" should be optional, so by default, if no parameter is passed, the timer will be reset to the same amount of time it was when first initialized.

Is the ZIP package somehow broken?

DropDownFX.as seems to be almost empty in zip file and it gives errors.
Also getting quite many other errors from missing greensock and other used libs as those are not included.

If the image's height is not divisible by the offset, this results in a crash because a script timeout. The reason being that subtracting the offset from a uint(dropY), can cause it to be negative, but since uint's can't be negative, the result is a very large number.

A simple restructuring in the draw() method fixes the problem.

Changing:

dropY -= offset;

dropRect.y -= offset;

if (dropY <= 0)
{
complete = true;
}

Into:

if (dropY <= offset)
{
    dropY = 0;
    complete = true;
}
else
{
    dropY -= offset;
    dropRect.y -= offset;
}

after line 48 in GlitchFX, need to add:

if (updateFromSource)
{
    sourceRef = source;
}

Hi,

It would be nice if we can somehow find the weapon (it's name or it's reference) from a fired bullet.

-tnx

@photonstorm hi ,because of the constructor parameters, FlxBar cant be recycled by the FlxGroup,can the parameters move to a init method ?

can I set FlxButtonPlus.scrollFactor.x = 0 ?

should make merging in changes easier too.

Just a few issues I have regarding the structure of the repository.

I can't seem to find the actual "code" for Flixel Power Tools other than in the pre-compressed ZIP (which defies the whole purpose of having version control, no? ;) This also makes forks or pull requests impossible.

Why are the Flixel libraries included? Do they have any changes made to them in order to work with some parts of the power tools? It may be easier including them as a SWC instead, but that's just a personal preference.

Trying to setup a sprite that's on a different camera from the default. Even when I set onlyScrollOnscreen to false it wouldn't scroll.
It looks like the draw() function is incorrectly checking if the user wants to override onScreen() checks:
if ((obj.onScreenScroller == true && obj.source.onScreen()) && obj.scrolling == true && obj.source.exists)
Should be:
if ((obj.onScreenScroller == false || obj.source.onScreen()) && obj.scrolling == true && obj.source.exists)

Can anyone see what is wrong with the following code?

package;

import flixel.FlxState;
import flixel.addons.plugin.FlxScrollingText;
import flixel.text.FlxBitmapText;
import flash.geom.Rectangle;

class BlahBlah extends FlxState
{
    override public function create():Void
    {
        var scrolltext = new FlxBitmapText();
        var scroller = FlxScrollingText.add(scrolltext, new Rectangle(20, 20, 200, 40), 1, 0, "*TEST* ");
        add(scroller);
        super.create();
    }

    override public function update(elapsed:Float):Void
    {
        super.update(elapsed);
    }
}

These are my current versions:

$ haxelib list
firetongue: [2.0.0]
flixel-addons: [2.5.0]
flixel-demos: [2.4.3]
flixel-templates: [2.4.6]
flixel-tools: [1.4.1]
flixel-ui: [2.2.0]
flixel: 4.3.0 [git]
hxcpp: [3.4.188]
lime-samples: [4.0.1]
lime: [2.9.1] 6.1.0
nape: [2.0.20]
openfl: [3.6.1] 7.1.1
poly2trihx: [0.1.4]
spinehaxe: [3.5.0]
systools: [1.1.0]
task: [1.0.7]

I am compiling on Linux to neko but I get a static garbled output:

Similar results with html5 and flash - can anyone help? (Please bear with me as I am very new to haxe)

ScrollZone doesn't work too well if you're messing with alpha transparencies.
To fix, in the scroll function, right I set data.source.pixels to a new, empty, transparent BitmapData. It seems to work without any speed/performance problems (but I'm using a tiny Png).

This is my 'fix':

zone.buffer.graphics.endFill();
data.source.pixels = new BitmapData(data.source.pixels.width, data.source.pixels.height, true, 0x00000000); /// <- added
data.source.pixels.draw(zone.buffer, zone.drawMatrix);
...```