9 réponses à ce sujet

[ColorsFade]

I went to add ga_force_exit as an action on a conversation node and when I hit the "Refresh" button to pop the parameter textfields, it threw the error, "The script failed to compile. Do you want to remove all parameters?"

I haven't touched that script - it's a default script. So I closed my toolkit and restarted and still got the error. 

I decided to copy and past the code from that script to a new script, and it's reporting unrecognized symbols in ginc_misc.

What the hell is going on here? How do I fix this? This really sucks.  

[kevL]

no problems here w/ ga_force_exit.
references ginc_misc -> ginc_actions ...

[ColorsFade]

This only just started happening to me. I've used ga_force_exit in other areas and it's fine.

I traced it up the include chain to a method that doesn't appear to work (why, I don't quite understand yet). So, I wrote a custom script. The culprit seemed to be a method that gets an object, so I changed my custom script to use GetObjectByTag, and it worked.

Not quite sure all the issues, here, but glad I have something working for now.

[Morbane]

did you try to remove all parameters? after all that may just be a dialog error not a script error.

Its likely im wrong but i threw it out there...

[ColorsFade]

Well, the parameters don't even show up - that's the problem.

On the dialog node, in the Actions tab, I add a new Action. I type ga_force_exit into the dropdown (or select it, either way produces the same issue). I click the "Refresh" button to show the textboxes for the parameters and the error pops. I never even get the textboxes t show in the first place.

I don't mind too much, because my own script actually works fine and has an additional parameter for delay (because telling a group of NPC's to all exit at the same time creates as cluster...). So having a delay has been nice; I can make the NPC's look a little more random in their exit.

I just don't understand why this script decided to fail now.

Now I have to go back and check my other usages to make sure they aren't failing as well.

[Lugaid of the Red Stripes]

What you need to do is replace the ncs file, the compiled script file. Deleting both the nss and ncs file should revert you back to the stock default.

[ColorsFade]

I think what may have happened is I may have named one of my own scripts to the same name as a default script.I just ran into this same exact issue with a gc_journal_entry type function, and then came to discover I had named a library of my own ginc_journal. Once I renamed that and reloaded the toolkit, the problem went away.

[Surango]

A lot of problems are simple like that, consider putting a tag in front of your scripts, like cf_ginc_journal so you can easily tell what's yours and what's stock.

[ColorsFade]

Surango wrote...

A lot of problems are simple like that, consider putting a tag in front of your scripts, like cf_ginc_journal so you can easily tell what's yours and what's stock.

Yep. 

Originally, I had all of my scripts named as d_ or x_ (the x_ being for testing only). But then as I started creating some really elaborate libraries for things, I thought I'd try and stick to the naming conventions for gincs.... Ooops! 

So yeah, the gincs are getting the d_ in front now too. 

[Tchos]

As far as naming conventions go, you'll see "kinc" for campaign-specific libraries (see the expansions) and prefixing module scripts and other resources with "m0*_*". I didn't use those conventions, myself. I went with tchos_inc, tchos_quest_inc, and tchos_fx_inc for my custom libraries.