ChronoCrash Modders Tools 0.7.9.10.1

[Piccolo]

Oh wow, the bugs i fixed few days ago suddenly came back out of nowhere and i see that the text files reverted to some old ass versions from few days ago, because i remember fixing specifc lines of code which magically reappeared despite me fixing them....
If this is because of sessions code or something of that nature, i want to disable this premanently ,this is not ready to be used .
And i have no clue now how muych other stuff was brought back without my permisison, how something like this can happen?
Is it autosave code? What is it ? I dont want it to ever happen again, so id rather do my own autosave and not let the software to store entire text files and bring them back, this should not happen.
I know its vague as hell but, yeah i have no idea wha happun... and i dont want it to happen again pretty much.
Im just glad it was a visual change and this is why i noticed it and checked if it changed some scripts, but i wonder how much it changed where it wont show up visually....

Eh I'm more inclined to assume this one is on you. There is absolutely no way the program can do that. The session system doesn't contain content of files at all. It just keeps track of all files that were opened (the paths of the files not their content). So when you reopen the app, it's basically just like shortcuts to the actual files, to let you reopen them.

What's more plausible is the file (the changes) never being saved in the first place for some reason. But reverting to a previous file, no this can never happen. Even the backup manager of CMT doesn't let you do that (you have to copy/paste the content manually).

It's very easy to suspect specific causes to problems, but the track record shows that the suspected causes are rarely the corrects ones. For example, Ilu is no longer experiencing issues since he disabled an option I added recently, and he attributes this change to that option. But knowing how the program works and knowing what changed recently, I'm pretty sure it has nothing to do with that option and more to do with me fixing two core issues around the same time this option was added.

 

[ABK]

Well it did it again yesterday - removed my changes and brought back older version of the text file despite me working in text editor all the time, but if i cant reproduce it, theres no way to address it , i posted to just let people know to be careful.
Maybe i do something unusual who knows, i pretty much edit scripts in text editor, sometimes checking frames in animation viewer etc, maybe i saved in animation viewer and it meessed things up in text editor, no clue.
Imo animation editor and text editor should reflush everything when switching back and forth so theyre both "on the same page" when it comes to current state of txt file, cause i can see sometimes animation names in text editor and animation editor dont match after i removed some anims/stuff in text editor, shouldnt text editor have a priority and animation editor just works with whats in text editor ?
Need more checks for existing anims in animation editor.

 

[O Ilusionista]

There is absolutely no way the program can do that.

I am sorry to sound a dick, but yes, this can happens and even happened to me - scroll back some pages and you will see me talking about it. I had to rely on dropbox to recover my files.

For example, Ilu is no longer experiencing issues since he disabled an option I added recently, and he attributes this change to that option.

No the option per se, but the session feature. Once I stopped using it (I never manually used it, to be honest), I stopped having those issues.

For sure the only person who knows why this happens is you. But coincidentally I no longer had this type of problem when the option of not using an old session was implemented.

 

[ABK]

I pretty much stopped using session and so far it is stable but can be false assumption , the session thing stresses me out

 

[Piccolo]

I am sorry to sound a dick, but yes, this can happens and even happened to me - scroll back some pages and you will see me talking about it. I had to rely on dropbox to recover my files.

Nope I already said that after investigation it was something else entirely. Every report you made here the issue was either related to :
- unsaved data for entity or level lost when loading file first from session panel and then from library. Now this is fixed.
- when closing a file in entity or level editor, the next file in the stack would get the content of the closed file, so if you saved that file it would be overwritten with wrong content. This is fixed as well.

Again I insist, both these issues has nothing to do with reverting a SAVED file to what it was before. There is absolutely no way the program can revert a file to what it was days ago because it doesn't store the previous content. Now it's me that might sound like an entitled d*ck, but bear in mind that for these issues, I first investigated them thoroughly based on your hypotheses. But in the process I eventually found out that your hypotheses were incorrect (even though it was understandable why you made them). There were issues, but not caused at all by what you suspected (for example nothing to do with switch project, nothing to do with renaming, even though they were the main targets of the hypotheses. And now it's the session system that is the target and I can say for sure you're wrong here too). So yeah after spending a lot of time understanding these issues I won't let you say things about the program that I now know for certain you're wrong

And again, that is not to say there is no issue in the program, just that whatever happened it's not what you described here. I already said it in my previous post, but what is most probable is the changes you made were not saved at all. I can certainly believe that in some unknown conditions the program can fail to save some edited changes. And from that if you think it did save the changes, and tomorrow you see that the changes are not here, you might think the app has reverted the file to what it was before. But it didn't, it just didn't save the changes in the first place. This is believable. Reverting a saved file to what it was before ? It's just not believable.

 

[Piccolo]

But coincidentally I no longer had this type of problem when the option of not using an old session was implemented.

Yes because as I said, this option was added around the same time the two core issues (that could mess files) were fixed. It's just a coincidence.

If I reverted the core issues and kept the new option to not load previous session, I can assure you you'll re-experience file corruption again. Because it didn't matter at all whether the session was reloaded or not. The issue was the opened list and library list were not properly linked. So while working on your game, if you used both to open the same entity or level, the unsaved changes would be lost (because the entity/level was considered different). But this could totally happen if you did not reopen a session. It would just be delayed because the "opened list" would start empty, so at first there would be less chances to open the same file from library list and open list. But as you work you would open files and the opened list would no longer be empty, and eventually you would trigger the bug.

 

[Piccolo]

Imo animation editor and text editor should reflush everything when switching back and forth so theyre both "on the same page" when it comes to current state of txt file

It does exactly that. So if there is a discrepancy, it's not because there was no reflush, it's because animation editor did not reconstruct text properly.

And text view has precedence. Basically each time you go from "Text" to "Animation", everything that is in animation is wiped out and replaced with what was in "Text".

And when going from "Animation" to "Text", the text is rebuilt based on what it was before for the unchanged animations/lines, and what it should now be for the changed animations/lines.

So again, if there is an issue, it's in the reconstruction from "Animation" to "Text". And it's a big process, so the issues need to be more precisely identified to be fixed.

 

[Piccolo]

cause i can see sometimes animation names in text editor and animation editor dont match after i removed some anims/stuff in text editor,

Pretty sure there's no way it can happen.

If it happens then it means you found a way to completely bypass the wipe out from "Text" to "Anim" (this wipe out delete all data from "anim" editor and replaces it with what is in "Text"). And this wipe out is triggered :
- each time you open a file, whether from library or opened list
- each time you switch view tabs ("text", "animation", "level")
- each time you close a file

If you access "anim" using another way, let me know.

Also I assume you're using the last version of CMT (so the core issues fixed weeks ago are out of the equation).

 

[O Ilusionista]

I won't let you say things about the program that I now know for certain you're wrong

Again, I don't want to sound a dick, but this sentence didn't sounded... good.

Maybe it's a language barrier, but it sounds like a demotivation to feedback to me. I can't speak for other people, but it sounded like someone was accusing you of something, or belittling your work. Which is absolutely not my intention.

As I said a few times, I even felt bad always bringing a problem to you to solve, even though I know that this is part of my "job". And this phrase you said kind of reinforced this feeling.

I may have actually gotten this wrong, and I sincerely hope I got it wrong. Because I can't hide my dissatisfaction with reading this sentence.

Well, move on.

 

[Piccolo]

Again, I don't want to sound a dick, but this sentence didn't sounded... good.

Maybe it's a language barrier, but it sounds like a demotivation to feedback to me. I can't speak for other people, but it sounded like someone was accusing you of something, or belittling your work. Which is absolutely not my intention.

As I said a few times, I even felt bad always bringing a problem to you to solve, even though I know that this is part of my "job". And this phrase you said kind of reinforced this feeling.

I may have actually gotten this wrong, and I sincerely hope I got it wrong. Because I can't hide my dissatisfaction with reading this sentence.

Well, move on.

It's precisely because of feedback that I can say that. I spent hours digging every aspect of the code, trying to reproduce the issue based on the clues you gave me, trying to understand where it might come from and so on. So when I eventually found and fix these issues, and people think the issue is still there and still say that it's caused by something that is not related to it, of course I will intervene to say it's wrong.

Imagine working hard on a new car. The car works great but there is an issue that makes it's engine stops working. Users of the car try to understand the issue so as to not cause it. Some found out that when not using any electrical features, the issue never happens. So they start to tell other users that there is an issue with the electrical circuit and if you avoid using it you'll have no issue. The constructor works hard to identify and fix the issue, and eventually discovers that the issue is actually related to the way you press the buttons to trigger some electrical features of the car. They fix the buttons, inform the users that it's fixed. And then when another issue happens with the car, they're still people telling others users that they shouldn't use the electrical features of the car because the electrical circuit of this car has issues. What you expect the constructor to say ? "Oh whatever, let people think what they want, we don't care". No they will intervene and say that it's actually wrong, there is no issue with the electrical engine and people shouldn't avoid it, the new issue is not related to it. That's all there was to my remark.

And to be clear, this has nothing to do with feedback. I appreciate your feedback and welcome it. It never bothered me at all. But when you guys spread false assumptions about some issues of the app, it's not feedback, and for things that I know it's false (like this), I must clarify that it's wrong. Think about it, one of the reasons things started to get hot in the last few posts is because some of these false assumptions started to spread and take root weeks ago. I gently clarified weeks ago that some of these assumptions were false, but here we are. So today when it came back again I said it more clearly. It's annoying to be told you're wrong, I know, but at some point it's needed. Be assured that I don't act that way for every issue, in fact this is the first time I do. I give the benefit of the doubt until I'm basically 100% sure something is false.

 

[DCurrent]

Gentlemen, let's perhaps take a step back and cool off a little. I don't want to have to temp lock the thread or see strife between some of the best and brightest people in this community.

AFAIK, I am the only native English speaker between us, and though it may not be intentional on any of your parts, some of the language is indeed getting a little abrasive on all sides. I also understand the frustration on both sides of an issue (user / developer), so it's very important we all try to meet each other half way.

DC

 

[ABK]

Well i do ctrl+s to save , and it did save cause ive seen the bug fixed within animation, but in next 2 days the animation came back to a state from day ago that contained the bug so i was surprised, even tho i did seen it fixed in the game.

Pretty much no way to explain why or what caused it, session saver is just an assumption that can be total false lead , i dont know the tool that well and didnt read the source, so now i dont do ctrl S , i go to the text editor, copy all the code, and then save, and then i run the game to see if all is good.

Also i reported for the people so they know, im aware that in order to catch this one i need more context and a way to recreate that scenario.

Also also i was interested if others had it happen, all for the cause of nailing the cause of it so it can be solved faster.

The tool has too much good stuff to prevent me from using it cause of situations like this and i do not want to hide a fact that strange thing happened, on one hand i want it to happen again, on other hand the stuff i work on is on strict deadline so i dont have much time to experiment...

But yeah, i do realize that my post was just typical whiner that lost some data, i provided no context and how to recreate it, hopefully ill figure out how to trigger it again.

 

[Piccolo]

Well i do ctrl+s to save , and it did save cause ive seen the bug fixed within animation, but in next 2 days the animation came back to a state from day ago that contained the bug so i was surprised, even tho i did seen it fixed in the game.

Ok so what must have happened is this :
- You opened CMT
- You made changes in a file, save the file
- You test the file in engine, the changes appear, meaning the file is actually saved outside CMT
- Then before closing CMT you did something that for some reason reverted the file to what it was (probably using the transient data in anim editor)
- You use either CTRL + s or click yes when closing and it asks to save unsaved changes
- You close CMT, outside CMT the file is no longer saved with the new changes (even though you don't know it yet)
- 2 days later you open the file again in CMT, you find out the changes are no longer here, and assume they were just now reverted (whereas in fact they were already reverted two days ago, while CMT was still open)


What can't have happened :
- You opened CMT
- You made changes in a file, save the file
- You test the file in engine, the changes appear, meaning the file is actually saved
- You close CMT, outside CMT the file is still saved with the new changes.
- You reopen CMT 2 days later, CMT revert the file to a previous version


There's a world in which scenario 1 can happen. But scenario 2 is totally impossible, because when you close CMT there is no memory of what the files were before. So when launching CMT the second time, it can't revert the file to what it was before.

In other words what we are looking for is an issue where in the same session (hence why it has nothing to do with reopening sessions) :
1 / a file was changed
2 / then saved
3 / and then is reverted to what it was at the start of the current session

The only way I can see this happen is :
- First you edit file in Animation (or Level) view. Doesn't really matter if you edit things or not, just that you open the file it in that view
- Then you go in Text view and make new changes. You save the file (or you don't, actually I suspect that it doesn't matter)
- Animation view is now both out of focus and contains data that is not up to date
- Text view is focused and contains data changes that are up to date
- Then following an unknown sequence you can either :
- option 1 / access animation tab without the animation data being replaced by "text" data (that is, keeping the old anim data it already had). Normally this should always happen.
- option 2 / trigger the process that makes animation data (which is out of date) go to text view without leaving animation view (normally this process happens only when you leave animation view, because when you leave animation view it means animation view had the most up to date data)


I'll investigate the issue based on that, it's still not very clear but at least I know what to trigger

 

[Kratus]

Also also i was interested if others had it happen

Honestly I never had this "rollback changes" issue before, the only problem that can happen due to my fault is when I'm editing the same file in both CMT/External editor like Notepad++ or VS Code.

As an example: I can make many updates in CMT, save using ctrl+s, open the same file in VS Code without closing CMT, make more updates, switch to CMT to edit more things without refreshing using F5 and save again. So, CMT data will be outdated compared with the VS Code and will revert all the changes made in VS Code posteriorly. I can confirm it because I'm constantly monitoring everything through GitHub Desktop tracking every single change, plus VS Code refreshes and brings all the CMT changes at everytime I alternate apps through alt+tab (@Piccolo it could be an idea for CMT too).

So, I strongly suggest adding your projects to GitHub and then you can see the exact moment where the changes are reverted.

 

[Piccolo]

Honestly I never had this "rollback changes" issue before, the only problem that can happen due to my fault is when I'm editing the same file in both CMT/External editor like Notepad++ or VS Code.

As an example: I can make many updates in CMT, save using ctrl+s, open the same file in VS Code without closing CMT, make more updates, switch to CMT to edit more things without refreshing using F5 and save again. So, CMT data will be outdated compared with the VS Code and will revert all the changes made in VS Code posteriorly. I can confirm it because I'm constantly monitoring everything through GitHub Desktop tracking every single change, plus VS Code refreshes and brings all the CMT changes at everytime I alternate apps through alt+tab (@Piccolo it could be an idea for CMT too).

So, I strongly suggest adding your projects to GitHub and then you can see the exact moment where the changes are reverted.

Yeah I always assume people are sure they are not doing something like that when they report data loss, but it's still important to mention.

I will add a check warning whenever the current file has been modified outside CMT, it's a good idea.

 

[ABK]

I actually dont close chronotools, i have them open for days.

Assuming that some magic force saved the file when tool was closed thats too much fantasy, i have the editor open for many days in a row unless i need update then i do the update and reopen

 

[Piccolo]

I actually dont close chronotools, i have them open for days.

Assuming that some magic force saved the file when tool was closed thats too much fantasy, i have the editor open for many days in a row unless i need update then i do the update and reopen

Ok then that reinforces my hypothesis that it all happens during one session (from the perspective of the app).

 

[Bruce]

@Piccolo

Is it possible that you could please add the following options?

1. Animation --> offset
Apply All icon that lets users copy the offset of the idle or other animation and apply to all other animations.

2. Animation --> Body Box / Attack Box
Simplify on/off version of bbox and attack box

examples:

    bbox.position.x 0
    bbox.position.y 0
    bbox.size.x 0
    bbox.size.y 0
    bbox.size.z.background 0

to bbox 0 0 0 0

    attack.block.penetrate 0
    attack.damage.force 5
    attack.damage.type 1
    attack.effect.hit.flash.disable 0
    attack.position.x 195
    attack.position.y 5
    attack.reaction.fall.force 0
    attack.reaction.pause.time 25
    attack.size.x 165
    attack.size.y 148
    attack.size.z.background 20

to attack 195 5 165 148 5 0 0 0 25 20
can you please also add the dropv to the CCM?

I do understand that it is easier to read as it is, but the code is much longer.
Of course, some people like the way you have, but for me I prefer the shortcut version.

Thank you very much for your hard work!

 

[DCurrent]

@Piccolo

Is it possible that you could please add the following options?

1. Animation --> offset
Apply All icon that lets users copy the offset of the idle or other animation and apply to all other animations.

2. Animation --> Body Box / Attack Box
Simplify on/off version of bbox and attack box

examples:

    bbox.position.x 0
    bbox.position.y 0
    bbox.size.x 0
    bbox.size.y 0
    bbox.size.z.background 0

to bbox 0 0 0 0

    attack.block.penetrate 0
    attack.damage.force 5
    attack.damage.type 1
    attack.effect.hit.flash.disable 0
    attack.position.x 195
    attack.position.y 5
    attack.reaction.fall.force 0
    attack.reaction.pause.time 25
    attack.size.x 165
    attack.size.y 148
    attack.size.z.background 20

to attack 195 5 165 148 5 0 0 0 25 20
can you please also add the dropv to the CCM?

I do understand that it is easier to read as it is, but the code is much longer.
Of course, some people like the way you have, but for me I prefer the shortcut version.

Thank you very much for your hard work!

@Bruce,

I highly advise you avoid this idea. Over time, I will be fully depreciating the old style collision commands.

@Piccolo

Current stable build doesn't do this, but next release allows you to use attack/body none to close a body/attack.

DC

 

[O Ilusionista]

Over time, I will be fully depreciating the old style collision commands.

oh

2. Animation --> Body Box / Attack Box
Simplify on/off version of bbox and attack box

Options > Legacy Commands