Stumped on error in descr_settlement_plan.txt plus minor text questions

[Crusader Vanguard]

EDIT: "iirc" = if I recall correctly

Thank you. I'm not the type of person who uses short hand of any type (as you can easily tell by the size of my posts throughout this thread - I'm lucky I understand what "omg" and such mean). The only "short-hand" (I'm not even sure what to call it!) I use is "LOL" because it's the only one I can't "truly long-hand".

The thing is, your code seems perfectly fine... It's really perplexing.

I know!! That's what's maddening about this! Because I cannot find any errors in the 3 files where the CTD error is SUPPOSED to be, that means (although I cannot remember doing so) I added/edited code elsewhere and I screwed that up. I tried this "adding culture plus changing faction" before, got all the code in, and it had the same CTD error, but I had the entire game to find it in, which (if the error isn't in the 3 files I'm looking at) I'm facing yet again! Yet, I am confident they are the same error, as the process monitor (which is listed as a tool in the RTW forum) tells me very similar processes between my "first" broken mod and now this one, the minor differences I can easily explain.

Basically, I'm worried the error is elsewhere (because I cannot find it in the 3 files I think they should be in), yet because the process monitor basically showed me the same processes between my two broken mods, I'm confident they have the same error, and since I don't have all the changes in the mod I'm currently working on as in my "first attempt" mod, I'm thinking the name files are a very likely source of the CTD error for both mods.

If you completely remove the 'character_record' and the 'relative' lines, does the game work?

I didn't realize you can do that. Attempting:

I deleted the "character_record" and "relative" lines from the Ostrogoths and deleted the map.rwm. CTD error still occurs almost immediately after clicking the start campaign button, with the map.rwm being regenerated. I put the character_record and relative segments back in... especially for what comes below:

What MoN is saying is that deliberately introduced errors (that are known to produce an error message) can serve as markers that can show you up to which point in the file the game has not encountered an error. That's because the game parses files linearly and it exits as soon as it encounters the first error.

Eg:

line X
line X+1
...
line Y (with deliberate error)
line Y+1
...
etc

If the game crashes without producing an error about line Y, then the error is somewhere in the lines before that. If it crashes with an error about line Y, then the lines up to Y are fine and the errorless CTD is triggered by a line below that. You repeat the process again and again, and you progressively narrow down the problematic area of code.

That's what I thought. Thank you for clarifying that. However, your statement leads me to ask another question: If the game linearly process all game files, do you know the order in which it does? Even if I can get an error about "line Y" or not, the only thing I can be sure of is the file I put the error in. If I get no error about "line Y" in the descr_strat.txt, then the error is "ahead" of the code I broke, and I can search the descr_strat.txt (above where I put the error) for something wrong, but if I cannot find an error in that file, what files does the game access before it? Where do I search from there? Same thing if I actually get an error from the "line Y"... it comes afterwards... what files does the game access after the descr_strat.txt? You understand what I mean?

Okay... to test this, I'm going to introduce an error into the "buildings" listing of the Sarmatians (they are listed right after the Ostrogoths, the area I want to test), so if I receive an error about the broken Sarmatian building line, then I know the descr_strat.txt file is correct.

Okay, trying this: The Sarmatian capital has walls, so I changed "defenses" to "defense" to see what happens:

YES! KTM error produced when attempting to start campaign:

Script Error in: bi/data/world/maps/campaign/barbarian_invasion/descr_strat.txt, at line 6497, column 14
Building specification type defense not found in building database

So... the descr_strat.txt entry I rebuilt must work! Finally! Some good news! Thank you both for that suggestion! Fixed my "intentional error" as well.

Okay, MasterOfNone or Aradan, I have a few questions about how to proceed with this:

--> For the 3 questions I ask below, in what order does the game access these files?

1. Due to the fact I've been at this for so much longer than I wanted, I want to test the entirety of this file. Do either of you have a suggested "error" I could introduce in the "diplomacy" segment of the descr_strat.txt so I could ensure that ENTIRE file is working?

2. The other place I know is very sensitive to mistakes is the export_descr_buildings.txt. It's the other place where if I'm going to make a mistake... it'll be there. What type of "error" should I introduce there into the last temple entry (the final building chain) to verify that one? I might find that the error is before this file too.

3. Since we eliminated the Ostrogoth "family tree" segment as suspect with that test, that leaves the 2 name files as the obvious suspects for the CTD error. How do I introduce an error to test that?! All name errors as far as I know are CTD errors... and therefore there is no means of testing the name files. Unless... do either of you know what file the game accesses AFTER the name files? I might be able to introduce an error there (and maybe the file before the names ones?) so I can be 100% sure the name files have the error.

Thanks guys! I know it's a tiny step, but it's the first step forward I've had in a week! What's next?

[MasterOfNone]

If the diplomacy part of DS is wrongly coded it will not report an error (or CTD or KTM) - but if you add a campaign script reference line after it then you can test it because the script line won't trigger. But this is likely not where your error is.

[Crusader Vanguard]

Okay. If the diplomacy segment will not give me an error, then that's a bust. Also, the campaign script reference line... it sounds like the campaign must be WORKING in order for me to know whether it triggered or not. True or false? I'm guessing that's what you mean, and my campaign currently doesn't function.

If the diplomacy area don't produce any errors, that's actually good. There is nothing after it, and I know from my last working mod version that the diplomacy is correct.

Question: I tired to create a KTM error in the last "slave" army listing (since the "slave army" listings are the last thing before the diplomacy, the final segment of the file I have, this way I can check everything before diplomacy), but, based on results, I'm not sure if I did this correctly... or discovered my descr_strat.txt isn't as clean as I thought. My question: are the 2 things I attempted below sound KTM error testing, or are will they create CTD errors themselves?

1. I tried to replace the sub-faction name of the last general to "Error" to test that. I remember (I think) getting a KTM error from that before. Message was strange, but it worked to produce a message error. The last "slave army" is Hun based, with the name "Yabgu_Roila" attached. I changed the name to "Error"... and I got a CTD error. Did my changing of the name cause that, because I mistakened this for a KTM error type, or... did I narrow down the CTD area?

2. The last "unit" listed in the above mentioned slave army is "merc vandal raiders". I deleted the "s" off "raiders". Again, I got a CTD error. Sound testing or did I create a CTD again?

EDIT: Never mind. I double-checked my current descr_strat.txt against the one I backed up before editing names, and the only differences between the files are the Ostrogoth's "family tree" segment, which I just ruled out as suspect in the test from my last post. So... I must have just caused 2 more CTD errors that I have fixed. MasterOfNone is right, the error probably isn't there. However, to ensure it isn't elsewhere:

What about the 2nd and 3rd questions of my previous post? Any ideas on how to test those files?

[Crusader Vanguard]

A week and no one has anything? Does anyone have any suggestions for the questions at the bottom of Post #61? I need to narrow down the "error field" a little more, any suggestions?