Update java, make sure you don't put aibat in the actual osu folder (I couldn't run it in any folder which had a ! in the path).
That would make sense, since the program is configured to check from the first instance of "[Colours]" to the first instance of "[Hitobjects]". I think the bigger question is how the mapper managed to get 2 lines of that in the first placeblissfulyoshi wrote:
Oh yeah, this was a strange scenario. If you add a word under [Colours], (during the mod he had [Colours] printed twice), AIBat says the combo colors are inconsistent even though all of them were exactly the same. (It took me a while to realize what the issue was, but I severely doubt that this will happen very often)
Yeah, this problem isn't just for AIBat but comes up for all .jar's in any folder that has a "!" as the last character in the filename. Weird glitch or something, but I don't think it's something I can fix.ziin wrote:
Update java, make sure you don't put aibat in the actual osu folder (I couldn't run it in any folder which had a ! in the path).
I'm using Windows 7 too (64 bit). Hm... to try and control for other variables, try unzipping this test folder and putting it on your desktop, and trying to open that with AIBat. If that works then it probably has something to do with the external drive, if not then probably admin privileges. Additionally, what happens after you press the "Enter" button? Does the program freeze up (i.e. you can't click on anything/select any other folders) or just not respond?Card N'FoRcE wrote:
I swear that i can't get AIBat to work °°
I got the jar application, i run it and choose the beatmap folder.
After that i don't get anything else., whatever folder i choose it doesn't work.
I'm using Windows 7. If this helps, I have osu! sotred in a external drive, but i don't think it should be an issue °°
***Unused Image Files***
pippidonkiai3.png
pippidonkiai1.png
pippidonidle3.png
pippidonidle0.png
pippidonidle2.png
pippidonclear3.png
pippidonclear0.png
pippidonkiai0.png
pippidonclear6.png
pippidonclear1.png
pippidonfail0.png
pippidonfail1.png
pippidonclear4.png
pippidonidle1.png
pippidonclear2.png
pippidonkiai2.png
pippidonclear5.png
I already tried everything you said here: i tried opening the folder by both CTRL-O and manual folder selection, but when i press Enter it just does nothing °°akrolsmir wrote:
I'm using Windows 7 too (64 bit). Hm... to try and control for other variables, try unzipping this test folder and putting it on your desktop, and trying to open that with AIBat. If that works then it probably has something to do with the external drive, if not then probably admin privileges. Additionally, what happens after you press the "Enter" button? Does the program freeze up (i.e. you can't click on anything/select any other folders) or just not respond?
Also, on the off chance this works, try opening folders with the direct open function (ctrl+O, or File>Open). Shouldn't make a difference but might as well give it a shot.
I spent ten minutes trying to understand this part... did you mean AudioLeadIn? If so, that's a good suggestion, but is 1 sec long enough?blissfulyoshi wrote:
Another suggestion: Give a warning if the preview time + the time difference between the start of the song and the first note is less than 1s. This is to make sure the preview time is long enough.
Is Pippidon limited to exactly those frames, or can the animation be changed at will? (If it's the latter I'll have to implement that after I understand and properly account for SB animations. Which I should do anyways.)blissfulyoshi wrote:
Bugs:
Pippidon is not checked
selection-mod-xxxxxx should not be in a map skin (they don't show up) (there are actually quite a few skin files don't affect maps that are currently allowed right now, let me see if I can find a comprehensive list)
Taiko code printout for reference:
My guess is that Java somehow is denied access to reading files on your computer- that's the only reason I can think of for the window to open and yet not be able to open beatmaps. When I get some more spare time I'll try porting it to .exe to see if it makes a difference, but in the meantime you could try downloading eclipse and compiling the code yourself . I'm really sorry it doesn't work for you, if there are any maps you'd like to check you can always PM me and I'll send AIBat's results to you.Card N'FoRcE wrote:
I already tried everything you said here: i tried opening the folder by both CTRL-O and manual folder selection, but when i press Enter it just does nothing °°
The program doesn't freeze up when i press Enter, it just doesn't analyze the folder whatever the path is (even C:\AAAA).
I tried having a look at admin rights, but i basically have everything allowed already and UAC is disabled. I also removed some kind of "lock" under Properties because this is a downloaded file.
I dunno what else to do xD
No, this won't automatically rank your beatmapsSorry but this statement is still making me smile lololol
If AudioLeadIn + First note < 1500, output an error. It's more of an error check to make sure that you don't forget an AudioLeadIn if you need it.akrolsmir wrote:
I spent ten minutes trying to understand this part... did you mean AudioLeadIn? If so, that's a good suggestion, but is 1 sec long enough?
Add in a "skin mode". Or add the skin specific stuff (like fruit-plate, selection mods, and circle highlight) in a "skin" tab, and apply its own custom skinset (like how hitcircle, hitcircleoverlay, and approach circle have to be together). If someone has a skin specific file in the beatmap, treat it like a skin, and put a notice at the top saying "if this is not a skin, you have an unnecessary skin file in your map and should remove it".akrolsmir wrote:
I left in selection-mod-xxxx and other similar files so that AIBat could be used to check skins as well beatmaps, but I guess that could cause problems with the overlap (i.e. someone accidentally includes a selection-mod-taiko in their beatmap folder and AIBat doesn't catch it). Perhaps I'll just create a short skin checker later and release it seperately.
You can try reading the id3 tag, but the best way is to figure out the song duration.akrolsmir wrote:
And the method for calculating bitrate seems great, the only thing stopping me is finding the length of an mp3 file (still).
akrolsmir wrote:
I spent ten minutes trying to understand this part... did you mean AudioLeadIn? If so, that's a good suggestion, but is 1 sec long enough?blissfulyoshi wrote:
Another suggestion: Give a warning if the preview time + the time difference between the start of the song and the first note is less than 1s. This is to make sure the preview time is long enough.
Yeah...Audio lead-in. Not sure about the time 1-1.5 seconds is probably enough though.Is Pippidon limited to exactly those frames, or can the animation be changed at will? (If it's the latter I'll have to implement that after I understand and properly account for SB animations. Which I should do anyways.) Sorry not sure about the details about the frames. I guess for now, just allow all frames.blissfulyoshi wrote:
Bugs:
Pippidon is not checked
selection-mod-xxxxxx should not be in a map skin (they don't show up) (there are actually quite a few skin files don't affect maps that are currently allowed right now, let me see if I can find a comprehensive list)
Taiko code printout for reference:
- Checks for snapping in inherited pointsWhy does this matter? The only green sections that need to be snapped are those that start kiai and those that end kiai, and even end kiai is questionable.
Actually technically they do for a few seconds:blissfulyoshi wrote:
selection-mod-xxxxxx should not be in a map skin (they don't show up) (there are actually quite a few skin files don't affect maps that are currently allowed right now, let me see if I can find a comprehensive list)
It doesnt matter yeah, however it's good to know which ones are unsnapped, because if they are unsnapped AHEAD of a note, then we may have a hitsound problem since for it to affect it it needs to be on the note or before, otherwise it doesnt matter besides kiai really.Lybydose wrote:
- Checks for snapping in inherited pointsWhy does this matter? The only green sections that need to be snapped are those that start kiai and those that end kiai, and even end kiai is questionable.
... I always miss the obvious, practical solutions.ziin wrote:
Add in a "skin mode". Or add the skin specific stuff (like fruit-plate, selection mods, and circle highlight) in a "skin" tab, and apply its own custom skinset (like how hitcircle, hitcircleoverlay, and approach circle have to be together). If someone has a skin specific file in the beatmap, treat it like a skin, and put a notice at the top saying "if this is not a skin, you have an unnecessary skin file in your map and should remove it".
Alright, from there I found a magical library that gets the bitrate directly for me; I might not even need song duration anymore! Also, does anyone have an mp3 file with a Variable Bit Rate and a known average to provide me for testing the bitrate that the library give me?ziin wrote:
You can try reading the id3 tag, but the best way is to figure out the song duration.
And: http://stackoverflow.com/questions/3046 ... me-in-java
You could try digging through the source code, it should be understandable. Maybe . Anyways, currently they are:ziin wrote:
Also, I'm not sure which skin elements you check, but these are the important ones:
approachcircle
hitcircle
hitcircleoverlay
cursor
hit0 etc...
Questionable:
fruits
One thing you can do is check each and make a note that "X" is (much) bigger than the template skin dimensions.
Alright, I think a warning for under 2 seconds would be good, and perhaps a stronger one if it's under 1?Sakura Hana wrote:
Audio Lead-in wise i think 1.5 - 2 seconds before the approach circle appears is fine enough, however that would need a human modder probably (unless you can calculate time with approach rate), so i'd say 2 seconds until the first note should be enough, any less and throw a warning
... tried to use != instead of !.equals() on a string. Anyways, it should be fixed now.blissfulyoshi wrote:
I forgot if this was reported already, but it seems that .png files inside a folder aren't reported correctly.
Aww... that would involve changing how Kiai is checked- I was lazy and just compared the entire thing as strings, which was easier. Which is probably bad coding. Put under TODO.blissfulyoshi wrote:
Also, if some but not all of the kiai timings match up exactly in every diff, could you not post them. (so if only one kiai section is inconsistent among all the diffs, could you only post that one section?)
I couldn't reproduce this, could you tell me which folder you were using when this happened?blissfulyoshi wrote:
Bug: AIBat suggest there should be an epilepsy warning even if epilepsy warning is on.
Yeah, I've heard that. Besides what Sakura said, I just kind of like maps to be nice and neat and snapped (I once popped a bubble to fix one of my maps in that regard). If other's don't care about such things and don't want to change it they can just ignore those warnings, I guess.Lybydose wrote:
Why does this matter? The only green sections that need to be snapped are those that start kiai and those that end kiai, and even end kiai is questionable.
akrolsmir wrote:
Alright, from there I found a magical library that gets the bitrate directly for me; I might not even need song duration anymore! Also, does anyone have an mp3 file with a Variable Bit Rate and a known average to provide me for testing the bitrate that the library give me? Yay, let's use my map. http://osu.ppy.sh/s/32318 Bitrate is in the metadata (164kbps). More exactly, It is probably somewhere between 164-165.I couldn't reproduce this, could you tell me which folder you were using when this happened?blissfulyoshi wrote:
Bug: AIBat suggest there should be an epilepsy warning even if epilepsy warning is on.
As for this, I was checking http://osu.ppy.sh/s/26860, and if you look in the file, the epilepsy warning is on, but it warned me anyway. (Reminder: you can only see the visual effect of the epilepsy warning when playing standard, not through the editor)
I kind of already knew that, but thanks for your contribution. Thankfully the magical library I mentioned earlier finds both for me, directly .Maav wrote:
File length is filesize divided by bitrate.
Hope that helps.
Edit:
True for both ogg and mp3.
Cool, the check returned 164 as well. I guess that means I'll be implementing the bitrate check very soon.blissfulyoshi wrote:
Yay, let's use my map. http://osu.ppy.sh/s/32318 Bitrate is in the metadata (164kbps). More exactly, It is probably somewhere between 164-165.
Thanks for the sample map! I've caught and fixed the bug- I forgot to account for the space in between "Epilepsy Warning:" and "1".blissfulyoshi wrote:
As for this, I was checking http://osu.ppy.sh/s/26860, and if you look in the file, the epilepsy warning is on, but it warned me anyway. (Reminder: you can only see the visual effect of the epilepsy warning when playing standard, not through the editor)
About the new guidelines, it seems there needs to be a fair amount of discussion before those are finished and in addition AIMod is fairly low on the priority list, so I wouldn't worry too much AIMod having duplicate functionality.akrolsmir wrote:
Anyone know if peppy is planning on updating AIMod to reflect these new guidelines? If so, I might drop AIBat or rework it to include less overlap. It's not that I mind working on it, I just don't want to waste my effort on duplicating functionality.
Hm... could you provide a sample folder for me to look at? Thanks!Equivalent wrote:
Minor bug:
Hitcircle.png not reported missing even if none in beatmap as the other elements of hitcirles are there.
Ah, okay. I'll primarily be focusing on adding checks in the next iteration, since I've already implemented most of the features I wanted to. I guess I can start looking through the suggested new guidelines and putting checks for those as well. Thanks for the update!blissfulyoshi wrote:
About the new guidelines, it seems there needs to be a fair amount of discussion before those are finished and in addition AIMod is fairly low on the priority list, so I wouldn't worry too much AIMod having duplicate functionality.
Pereira006 wrote:
Hello
I have question here
I just feeling problem aibat or i wrong
Aibat is working my pc ok but sometime i click folder will not show nothing
I just test aibat and i click folder from my map but still not show and i remove diff insane, normal now Aibat is working and show all. So i put again normal and insane and remove hard now aibat is not working again
I just test another folder map is same problem
I can't understand very well what problem Aibat...
Sometime Aibat is not working somehere diff...
or I wrong?
i use Windows XP SP3
and sorry my bad english
Yes, that would be very helpful, since it sounds like AIBat is only breaking on certain maps, not everything (Like it does for CnF)Sakura Hana wrote:
Probably you should include a link to the beatmap/osz in question so he can work it out and see what's the problem so he can fix it
That's true, and hasn't really been fixed yet, so it's possibly the cause. I should probably make AIBat throw out a notice about that if it ever happens.lolcubes wrote:
From what I remember in earlier testing, you can't check mapsets which have empty maps. So for example, if you have a normal diff which is completely empty without a note, you won't be able to open the mapset at all. I haven't tested the latest version yet though.
Im pretty much guessing Bliss mean inherited time and unherited time on the same spot. It is rarely oocurs, if it only 1 offset maybe it easier to check without this app, but what if it got like 20 redlines? A good features I sayblissfulyoshi wrote:
Feature Request xD:
Check to make sure there are no timing sections at the same point in time, unless the timing section is there to change the slider speed. (If you are lazy, you can just check to see if there are timing sections with the same offset xD)
Well sometimes ppl put 2 uninherited sections at the same time (usually by accident....), so I want to avoid that from happening. Besides, of couse I want to prevent silly things like 2 inherited sections at the same time too.Equivalent wrote:
Im pretty much guessing Bliss mean inherited time and unherited time on the same spot. It is rarely oocurs, if it only 1 offset maybe it easier to check without this app, but what if it got like 20 redlines? A good features I sayblissfulyoshi wrote:
Feature Request xD:
Check to make sure there are no timing sections at the same point in time, unless the timing section is there to change the slider speed. (If you are lazy, you can just check to see if there are timing sections with the same offset xD)
package javaapplication4;
import java.io.BufferedReader;
import java.io.InputStreamReader;
import java.net.URL;
/**
*
* @author Alex
*/
public class WebRetrieve {
public static void main(String[] args) {
try {
URL url = new URL("http://osu.ppy.sh/p/beatmaplist");
BufferedReader br = new BufferedReader(new InputStreamReader(url.openConnection().getInputStream()));
String thisline = "";
while((thisline = br.readLine()) != null) {
System.out.println(thisline);
}
br.close();
} catch (Exception ex) {
System.err.println(ex);
}
}
}
It'd probably be relatively easy to configure this as checking for any two sections that are within, say, 5 ms of each other. Good idea, I'll work on it!blissfulyoshi wrote:
Well sometimes ppl put 2 uninherited sections at the same time (usually by accident....), so I want to avoid that from happening. Besides, of couse I want to prevent silly things like 2 inherited sections at the same time too.Equivalent wrote:
Im pretty much guessing Bliss mean inherited time and unherited time on the same spot. It is rarely oocurs, if it only 1 offset maybe it easier to check without this app, but what if it got like 20 redlines? A good features I say
Released, and thanks for this code! (I had no idea reading online stuff was that easy o.O) Don't know if I'll update beatmap url formatter yet, since I think AIBat has a lot more utility and potential as of right now. Also, if someone can find something really useful to do by reading the source code of osu! pages, I'd be glad to try and implement it.Zhanger wrote:
just wondering-- could you release the source for beatmap url formatter and the osudistancechecker as well? Just curious P:
also --
This would allow you to read the page source for your beatmap url formatter therefore speeding up the process :3
Thanks! These are good suggestions, and I really appreciate feedback about the layout and how to improve usage- I find that I spend as much time on that as getting the program to work in the first place.ziin wrote:
a few suggestions:
I'm starting to include an AIBat post in my mods. If possible could you do this stuff:
*Stuff*
And as for coding this in java, I don't think there's a problem I rather like having separate programs from osu since you can't have 2 osu windows open.
Glad it's being used! Development is slowing down since it's implemented most of the interesting-to-code features I wanted (e.g. searching), and putting in checks isn't as fun, not to mention working full-time hours for the first time ever . Hearing that it helps people keeps me working at it, though, so thank you for the encouragement!Sakura Hana wrote:
This is becoming a quite helpful tool, it helps me save a lot of time, keep it up!
[Storyboard]It would be nice if you put them in alphabetical/numerical order.
Unused Image Files:
comboburst-1.png
comboburst-5.png
comboburst-3.png
comboburst-0.png
comboburst-2.png
comboburst-4.png
package javaapplication4;
import javax.script.*;
import java.io.*;
/**
*
* @author Alex
*/
public class scripttest {
static ScriptEngineManager mgr = new ScriptEngineManager();
public static void main(String[] args) {
try {
scriptexec();
} catch (ScriptException ex) {
ex.printStackTrace();
}
}
static void scriptexec() throws ScriptException {
ScriptEngine se = mgr.getEngineByName("javascript");
se.eval("println('hello world');");
}
}