Pages: 1
  Print  
Author Topic: Regression Testers Wanted  (Read 5434 times)
Offline (Male) Josh @ Dreamland
Posted on: February 22, 2020, 02:43:20 PM

Prince of all Goldfish
Developer
Location: Pittsburgh, PA, USA
Joined: Feb 2008
Posts: 2960

View Profile Email
Hi folks,

We're looking to merge #1875 (New Events) into master. We've tested it pretty thoroughly, but there's still a chance it will break something; this recodes a huge chunk of the event system. We will be merging it into the EGM feature branch today, and expect it to hit master on or before the completion of EGM.

If you have time, and would not mind testing this change out, please issue git checkout NewEvents and give the system a go.

Please report any issues either in this thread or on the pull request page on GitHub. If no one says anything before EGM is complete, these changes will be merged and we'll deal with the fallout, then.

Cheers
« Last Edit: February 22, 2020, 05:56:50 PM by Josh @ Dreamland » Logged
"That is the single most cryptic piece of code I have ever seen." -Master PobbleWobble
"I disapprove of what you say, but I will defend to the death your right to say it." -Evelyn Beatrice Hall, Friends of Voltaire
Offline (Male) time-killer-games
Reply #1 Posted on: February 23, 2020, 09:03:32 AM

Contributor
Location: Virginia Beach
Joined: Jan 2013
Posts: 1167

View Profile Email
I noticed a regression with my key to success game.  pressing space to unpause the game will also make the player jump, as though I pressed the spacebar twice in the same step, when i only pressed it once. So well done, you fucked up keyboard input. Also noticed some minor problems with the collision event, specifically once you've climbed up to the top of a ladder in some levels it behaves different than GM now. I'm too lazy to include a small test case reproducible for these two regressions, so feel free to gut away my key to success game yourself to narrow it down. https://samuel-venable.itch.io/key-to-success

If you need further details on the collision bug I can provide a video if you aren't able to find out what I'm talking about on your own. The compiled executables are built with an older version of enigma which behave the same as GM, so you may use that as a reference. Ctrl+Enter to skip levels until you get to level 3 and that's where you'll notice the collision regression at the top of the ladder if you pay close attention, at the very beginning of the level..

Also what magical reason did you need to rewrite events for, if they were working just perfect previously, but they are not in this branch of yours?

You get mad at me for causing a so-called "regression" with Robert's game, but in his case it was a bad programming practice of his, not mine. What I did was take a bad implementation and made it a good one. He was using a string where it would make more sense logically by the function's name and purpose to use a number instead. Then you, as the big almighty project lead developer came out here with this big pr that potentially breaks every game made with enigma in existence, with both broken input and collisions, you out of arrogance are too afraid to merge my icon pull request, because of a "regression" I caused, that again, in reality, was just Robert being stupid and something really weird possessed him to put quotes around a number argument, turning it into a string for absolutely no reason. You blame his bad practices on my good practices. get_integer() should take a number, as the name implies, and as the function's intended purposes serve, it would only make natural sense it would take a number argument, yet you call Robert expecting to use a string where a number makes more sense to be used, "a regression".

I question the state your mental health if this is honestly how you feel, even now. Just don't use that crap as a weird-ass excuse to be afraid to merge my icon pr. As this pr of yours clearly demonstrates, you are more prone to killing the entire project than I am, where as I was very lucky to have very loosely-speaking "break" a single game. Do you have any idea how many games even use get_integer()? Robert's might have been the only one in our community that used it. You, on the other hand, broke our entire input and collision systems, in ways that are very specific, quirky, and extremely hard to narrow down, and debug, the root causes for it.

Good day to you, ween.
« Last Edit: February 23, 2020, 09:34:58 AM by time-killer-games » Logged
Pages: 1
  Print