Owh kk thanks for the reply, i think i’m gonna held my project for some time until i have the will again since this just dropped my mood to start over again.
Hey all, don’t know if it’s just me, but it seems like the web version of choicescript IDE is down at the moment. I tried it on both chrome and firefox, and I signed into my dropbox just to make sure it wasn’t a problem with that.
This is what it looks like when I log into it:
Hi @ViIsBae , please can you try the suggestions above? [CSIDE] The ChoiceScript IDE (v1.3.0 Now Available — 18/06/2019) - #1126 by CJW
I have to head to a lab now, but I tried clearing my cache and opening it in an incognito tab, and neither of those worked. For some reason, F12 didn’t open the console on chrome (don’t know if me being on a tablet with a connected keyboard has something to do with that or not), but just thought I’d give you a rundown of what I did so far. I’ll try more when I get home in a few hours and update this then.
No worries, I seem to be able to reproduce it now. It’s related to a fix I made yesterday, which I’ve reverted, but whilst the page loads Dropbox still isn’t recognising my credentials correctly. This might be a tricky one to fix, please bear with me.
Edit: looks like things are in a better place now, although I’m personally having trouble logging into Dropbox, so can’t confirm 100%.
1 Like
It seems to working working now on my end!
1 Like
Hi, I hope it’s ok to ask a CSIDE question in here?
I have been having trouble getting my randomtests to finish, when I get to about the thousand seed the entire thing just slogs down to a stop and the test-window turns white (I usually have display choices w/o text + scroll to bottom). I have tried waiting for over an hour, but I am a very impatient woman. Should I wait longer? Or am I missing something, or doing something wrong? I have tried to look for ‘infinite loops’ but I barely understand the term, let alone know what to look for. 
Grateful for any help as manually going through and testing every choice is just unfeasible, and I’m missing a lot of errors. wails in despair
(can I also squeeze in a huge thank you to the dev of CSIDE. It’s so intuitive, it’s just me that is probably committing heinous crimes against it)
1 Like
I have had this issue in the past you can get around it by shortening the number of tests you do. I think it’s caused when randomtest hits a loop. But I’ve noticed that if I don’t show choices the test goes through completely fine.
@CJW will be able to answer the question better though.
2 Likes
Hello!
The HTML version of Randomtest (which CSIDE uses), can be rather flaky with large numbers of iterations. You may also be hitting an infinite loop, as @Nocturnal_Stillness says.
I’d advise sticking to fewer runs, each with an adjusted seed (so you still get good coverage).
If you suspect you’ve got an infinite loop you can try to start from a seed near 1,000 and see if you hit it more quickly. If you don’t see it at the same seed but rather see it again after a similar number of iterations, then I’d put it down as a performance issue.
I hope that the next major update to CSIDE will move to using the command line tests in separate processes (i.e. your test will be run in the background and output will be dumped to a text file).
E.g.
GIF
2 Likes
Is there a chance of having the legacy versions as well?
The text files will be great for in-depth analysis, but the html version still looks useful for quick on-spot testing.
Maybe I am mistaken, idk.
1 Like
Thank you so much for the thorough response! You are a gem of a human for not only making CSIDE (and it’s so accessible! I could gush about it for days.) but also stick around to answer Q’s for code-illiterate peeps like me 
.
Ok! I will try it with less iterations! Please forgive me for my ignorance, but what is an adjusted seed? I googled around a bit and it was incredibly hard to even figure out what a seed is, lol. What I gathered is that you can choose different seeds, which will increase the probability of different random results; but it also changes the distribution of results when you ask for sequences of random results. Is that correct?
So if I write “1000” in the seed-box in the randomtest window, I will get a thousand algorithmically different iterations of the “playthrough” the test goes through? That would explain a lot, that if the seed is set 0 it runs through the same sequence of options and slogs down at around the thousand iteration, it is a performance issue. But then I must ask again, do you mean that it’s a hardware issue or is that where the infinite looping comes in?
Can I put a *looplimit somewhere in my code to try to find it, or should I just replicate the completed randomtest (seed 1000 or wherever it starts looping) manually somehow? Is there anywhere I can learn a bit more about infinite loops and how to identify them?
(And wow, that update sounds fantastic. Having saved logs would be amazing! Thank you for all the hard work you do!)
Again, thank you so much. Have a nice day!
1 Like
By adjusted seed he means you run 1000 iterations from seed 0 (covers tests 0-999) then you run another 1000 iterations from seed 1000 (covers tests 1000-1999) and so on. Hope this helps 
3 Likes
That makes total sense! Thank you so much for your reply (earlier too!). 
1 Like
You are welcome. Hope it helps!
2 Likes
I don’t currently plan to support both, that would be quite a bit of work. But I’ll be receptive to feedback, if you try it and don’t feel like it’s as convenient in some way.
Currently, I run the test… pass or fail pops up on my screen.
In the future version, I need to go to a text file and retrieve it. I run the tests often enough, such a start-stop workflow would be very detrimental to my development and writing… unless I am not understanding the process from your gif.
I see the benefits of the command line testing, but I also see the benefit of the current html legacy version as well.
Alright, it works now. Thanks!
1 Like
You’ll get a notification telling you whether or not it’s passed or failed, which will have a quick access button to open the log file. I’m very confident it won’t be a regression in service, in any way, shape or form. But, if you try do it and don’t like it, let me know and we’ll improve it until you’re happy
1 Like
Other than *comment ing the code out or deleting the code is there any way to run quick/randomtest while using the save system?
For randomtest you could try ‘if not(choice_randomtest)’, but I think for quicktest you’re out of luck.