I don't know if this is possible, but would be nice if Validator raised an error if some .txt/.csv files have non-ANSI encoding, as it screws up in game
By "non-ANSI" he means Windows-1252 encoding.
The Validator - Find errors quickly and with minimal pain!
- Thread starter Jamie550
- Start date
-
We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
By "non-ANSI" he means Windows-1252 encoding.
The first one is added, as for the second not sure how this is supposed to work with things like mercenaries etc. The first doesn't seem to work that well either but I will put it in and see what people have to say.
I don't know if this is possible, but would be nice if Validator raised an error if some .txt/.csv files have non-ANSI encoding, as it screws up in game
The validator does read using the correct encoding but unfortuanetly it can't really tell when the text is not of that encoding.
Version v1.9 RC14 has been released: https://www.dropbox.com/s/b7292s3kkznl5q6/Audax.Validator v1.9 RC14.rar?dl=0
If I did not reply to you, then you issue should be fixed, but of course report again if it doesn't seem to be.
Please keep reporting false positives with the Validator, and they will likely get fixed. If you are reporting a false positive, it is very useful to copy the error message into the post - you can just select the error message, then press Ctrl+C to copy it.
If no issues are reported on either the Eu4 or Ck2 forums for a week, then a final 1.9 will be released and it will be the last version until a new patch comes out.
If I did not reply to you, then you issue should be fixed, but of course report again if it doesn't seem to be.
Please keep reporting false positives with the Validator, and they will likely get fixed. If you are reporting a false positive, it is very useful to copy the error message into the post - you can just select the error message, then press Ctrl+C to copy it.
If no issues are reported on either the Eu4 or Ck2 forums for a week, then a final 1.9 will be released and it will be the last version until a new patch comes out.
The second would be cool, as it's a pain to find manually ! You're right about mercenaries, in fact if title doesn't have any nested title inside (i.e. it's titular), then there should be no checks on capital province.
I've noticed a false positive:
Code:
--- Error 1 of 1 ---
At <mod>\decisions\ACR_hellenic_decisions.txt [decisions\recreate_pontificate\potential\primary_title\higher_real_tier_than] (Line 78, column 22):
Invalid node "higher_real_tier_than" in scope TitleTrigger (value is: DUKE)I've tested it works fine, both on TitleTrigger and CharTrigger (similar to higher_tier_than).
Haven't tested lower_real_tier_than, but I would assume it's the same.
I've had some troubles with history/technology with a missing year definition for a group of duchies, causing CTD on game applying history :closedeyes:
Basically if defines DONT_EXECUTE_TECH_BEFORE is set to a given year (ex: 475), then that year entry MUST exist in all history/technology group of duchies.
I don't know if it's possible to check that earliest year is the same in all groups ? For instance raise some error in the following case:
Code:
technology = {
titles = {
d_duchy1
}
475 = {
military = 0.5
economy = 0.5
culture = 0.5
}
1337 = {
military = 3.5
economy = 3.5
culture = 3.5
}
}
technology = {
titles = {
d_duchy2
}
#Missing year 475, game will CTD if DONT_EXECUTE_TECH_BEFORE is set to any year before 769
769 = {
military = 0.5
economy = 0.5
culture = 0.5
}
1337 = {
military = 3.5
economy = 3.5
culture = 3.5
}
}
Can you do add a check that the event (but probably is a problem for decisions, localization and so on) does not contains the character "non breakable space" or in unicode '\u00A0'? I lost two days of my life trying to understand why a correctly event does not triggered luckily when put in the code tag the ugly truth is shown:
in the form of that '*' that is not, apparently, present in my text!
I don't know if there are other not printable character that the engine does not like if so please add them too.
Thanks!
Code:
namespace = eog
character_event = {
id = eog.1
desc = *EVTDESCeog.1
[...]in the form of that '*' that is not, apparently, present in my text!
I don't know if there are other not printable character that the engine does not like if so please add them too.
Thanks!
No. The exact opposite. By "non-ANSI" he means NOT Windows-1252 encoding.
@Romulien
If you have 479 as the don't execute tech date you can as well have a tech definition for 470 and it will work. Only if you have no tech for the defined year or after you'll get a problem
@Jamie
Would it be possible if you could introduce a second type of error? Let's call them error and warning. If errors are only things that will crash your game on load, on resign and on save/load as well as bugs that cause blank events and make all the others warnings your tool(s) would be a lot more useful for trouble shooting. You could look at the list and fix the errors that could be sorted to the top first and then fix some of the probably thousands of warning the Validator prints out.
There's plenty of things that will crash the game that are hidden among inconsequential bugs or not reported at all.
I love your tool but it is frustrating to use it to find deeply troubling bugs. A friend of mine keeps claiming that he found the great bug that haunts the Warhammer mod by fixing 100 Validator errors while the big, unfathomable bug still looms in the dark. Maybe the Validator has found the cause but if it did it is buried beneath thousands of other errors (vanilla event files, disconnected dynasties, ...). I am not saying that this things are not worth fixing but it would take us like a months to connect all those disconnected dynasties and it would not bring us closer to fixing the greater, underlying bug.
It would just be great if the Validator could visualize things that crash the game if it found any. If I forgot a brace on a character's birth clause the game will break. If I did such a mistake I'd love to see this right away.
Last edited:
Would be great indeed
For instance invalid node type or duplicated ID should be error, while missing localization or duplicated names in list should be warning.
Would be great indeed
For instance invalid node type or duplicated ID should be error, while missing localization or duplicated names in list should be warning.
Duplicated ID? Care to elaborate? I like to duplicate my landed titles and would love to not be confronted with the errors that the Validator prints out due to that. It is a nice and clean way to separate things in landed titles. This file has all the cultural renamings, the main landed_titles only has the core things.
Duplicated event and char ids? Duplicating things for the most part does not break the game. Even if you mistakenly use the same id for several chars the game will load up and play just fine. Even if then a person has a man as a father.
If error levels would be possible to have 3 levels would probably make most sense: FATAL, ERROR and WARNING (dark red, red and yellow (?)).
That's what I meant. We had some problems with encoding in AR and I had to do some research to figure out what people meant by "ANSI" (because the American National Standards Institute is not a character encoding
).Duplicate event and character ids should definitely be an error.
Otherwise they are an invisible failure and I have no idea what happened.
Yeah, would need to review exactly what works and what doesn't, and agree on what is an error and what should be a warning...
Duplication for things loaded from folders is actually an important topic to improve compatibility between mods and between mod and vanilla !
Decisions: error (duplicate decisions between mod and vanilla does not work, effect is not called
Events: error (between 2 events in the mod) or warning ? (between mod event and vanilla event - but does the override actually work ?).
Titles: error (duplicate de jure structure in landed_titles can cause holy sites to be referenced twice in religion, same barony referenced in multiple counties causes bug, ...), warning in some cases (barony title override, ...).
Cultures: error (culture group appears twice in ruler designer I think)
Religions: warning (override works)
Characters: need to agree
GFX: warning (override works)
on_actions: warning (works)
localisation: warning (works)
CB, Nicknames, minor_titles, traits: ? (not tested)
Yeah, would need to review exactly what works and what doesn't, and agree on what is an error and what should be a warning...
Duplication for things loaded from folders is actually an important topic to improve compatibility between mods and between mod and vanilla !
Decisions: error (duplicate decisions between mod and vanilla does not work, effect is not called
Events: error (between 2 events in the mod) or warning ? (between mod event and vanilla event - but does the override actually work ?).
Titles: error (duplicate de jure structure in landed_titles can cause holy sites to be referenced twice in religion, same barony referenced in multiple counties causes bug, ...), warning in some cases (barony title override, ...).
Cultures: error (culture group appears twice in ruler designer I think)
Religions: warning (override works)
Characters: need to agree
GFX: warning (override works)
on_actions: warning (works)
localisation: warning (works)
CB, Nicknames, minor_titles, traits: ? (not tested)
For me it is: landed_titles duplicate definitely not an error. I have all kinds of odd incomprehensible errors unless I delete my title renamings file, my title color-change file, etc.
Someone could even make a holy_sites.txt and put only those there. You could see with a glance where each holy site without having to scroll through a gigantic landed titles file.
Localisation duplicates should probably not be reported at all unless seperately enabled. You see if you make a new map you need to rename all provinces. PROV1, PROV2, .... --> tons of "duplicates"
If you just replace_path localisations or replace a single you'd have to drag vanilla localisations along which is also bad.
character, trait and decision ids are of course bad most of the time.
You could, however, make a perfectly legitimate mod that only adds traits to certain important characters only by overriding things. That way it would be compatible with mods and would be very easy to fix for patches.
A really great tool, have used it quite extensively to fix my own personal mod when the new CK2 versions come out and break it. Only problem is, is that for some reason notepad++ only shows a blank page when I click the *check mark* next to the error in the Validator. Does anyone happen to know why it's doing this?
The Validator is often case-sensitive (on purpose), so "duke". Though there were a few missing ones that were fixed.
Can you post a copy of the bad file? So I can check if the checks work right. Note since things should be ASCII, the validator would check for things like \x00 and other things that aren't normal characters.
One person's error is another person's warning. (There are some things that are definitely warnings, but most are debatable). In some cases if you forget braces the game will still run just fine. If you want you can always copy out the error list and write a script to filter it to what you find interesting.
What do you have in File -> Options -> Text Editor?
Version v1.9 RC15 has been released: https://www.dropbox.com/s/vnwn0db6pzs21m7/Audax.Validator v1.9 RC15.rar?dl=0
If I did not reply to you, then you issue should be fixed, but of course report again if it doesn't seem to be.
Please keep reporting false positives with the Validator, and they will likely get fixed. If you are reporting a false positive, it is very useful to copy the error message into the post - you can just select the error message, then press Ctrl+C to copy it.
If no issues are reported on either the Eu4 or Ck2 forums for a week, then a final 1.9 will be released and it will be the last version until a new patch comes out.
If I did not reply to you, then you issue should be fixed, but of course report again if it doesn't seem to be.
Please keep reporting false positives with the Validator, and they will likely get fixed. If you are reporting a false positive, it is very useful to copy the error message into the post - you can just select the error message, then press Ctrl+C to copy it.
If no issues are reported on either the Eu4 or Ck2 forums for a week, then a final 1.9 will be released and it will be the last version until a new patch comes out.
So no chance for a tiered display of errors because some for some it is debatable if they are errors or warning? How disappointing.
So I've downloaded the validator and I've got the Microsoft .NET Framework 4 Client Profile thing. But every time I try to fire up the Validator I get the message "Audax.Validator.Boot has stopped working". Is there a way I can fix this?
It would be nice to have a filter for fatal and non fatal errors at the very least. While what is a warning could be considered debatable, if it crashes the game or not should be easy enough to figure out.
Sorry for the late reply, I originally had the notepad++.exe, but I removed it, and NotePad++ still showed up, and was blank. So I'm not sure what is going on. Maybe because I set NotePad++ as the deafault text editor on my System?
Hmm if Notepad++ is the default editor then it should at least not open blank, but open to the file. Can you post the contents of Data\save.txt, and also report on what happens when you open the run dialog with WindowsKey+R, and type in the path of a text file, then press enter?