※ Dnh Q&A/Problem thread II ※ for Danmakufu ph3 .1 preX(ry

[TresserT]

Is it possible to temporarily/permanently disable the backspace key? And by that, I mean the "start script from beginning" thing it does.

[Chronojet ⚙ Dragon]

It seems like that's something enforced by Danmakufu. If you were testing your own scripts, you could design a quick package script to run the singles/plurals/stages you are making, since the "retry script" hotkey is disabled in packages. It's a bit tedious, though, and I'm not sure whether it's worth the time and effort.

Unfortunately, backspace is not bound to a virtual key, so you can't force the key to a certain press state.

... This leads me to believe that Danmakufu has a "default package" that runs singles and plurals, which you can't edit. It seems that the resource folder only contains the stage codes for running plurals and singles, which are used by this hypothetical "hidden default package".

Of course, this is just my speculation, and may not actually be the case. However, what I can do is suggest to mkm a function that can change the key for retrying scripts, just like I had requested SetSkipModeKey.

Are you, perhaps, making a kind of typing game in Danmakufu?

[TresserT]

It seems like that's something enforced by Danmakufu. If you were testing your own scripts, you could design a quick package script to run the singles/plurals/stages you are making, since the "retry script" hotkey is disabled in packages. It's a bit tedious, though, and I'm not sure whether it's worth the time and effort.

Unfortunately, backspace is not bound to a virtual key, so you can't force the key to a certain press state.

... This leads me to believe that Danmakufu has a "default package" that runs singles and plurals, which you can't edit. It seems that the resource folder only contains the stage codes for running plurals and singles, which are used by this hypothetical "hidden default package".

Of course, this is just my speculation, and may not actually be the case. However, what I can do is suggest to mkm a function that can change the key for retrying scripts, just like I had requested SetSkipModeKey.

Are you, perhaps, making a kind of typing game in Danmakufu?

Haha no, I'm nowhere near that creative. I'm using CommonAreaDatas to try and make some a menu-like scripts. I have it so that when you hit backspace during a stage you restart that stage, which is fine, but if you hit backspace during stage select it returns you to the difficulty screen instead of the main menu, which is not fine. It doesn't sound that bad, but it screws up all the CommonAreaDatas. But if it can't be helped I'll just find another way to do menus =P

[Helepolis]

Semi related, speaking of CommonDataAreas, why is it that when I try get the data it returns me the default value instead?

For example SetAreaCommonData("BGM","Vol",100). If I want to GetAreaCommonData("BGM","Vol",0) it will always keep returning me the 0 instead of the 100 as stored. Am I missing something?

[arby26]

Semi related, speaking of CommonDataAreas, why is it that when I try get the data it returns me the default value instead?

For example SetAreaCommonData("BGM","Vol",100). If I want to GetAreaCommonData("BGM","Vol",0) it will always keep returning me the 0 instead of the 100 as stored. Am I missing something?

I believe you need to create the common data area first, with CreateCommonDataArea ("BGM").

[Helepolis]

* Helepolis facepalms

Seriously the scripting is starting to take its toll for going all out for last 3 weeks. But I have to....

Thanks.

Edit: Though that makes me realise, Ph3 isn't warning you nor throwing errors at you for this. It perfectly accepts the fact you're setting AreaCommonData and trying to get it. I guess something along the lines of strings and such?

[arby26]

Common datas seem to be one of those things that we have to go out of our way to handle. I try to treat use them as a last resort.

There's an IsCommonDataAreaExists() for checking the areas themselves.

I like that it returns the default value instead of erroring. In my projects there is at least some common data for game settings change in the package script, but the default values let me still test parts of the game as singles/plurals/stages.

[Helepolis]

I meant more like the Engine throwing errors because there is no such commondata defined but it seems like SetAreaCommonData already behaves like setting it but it is non existent or so 

Oh well.

[Uruwi]

It seems like that's something enforced by Danmakufu. If you were testing your own scripts, you could design a quick package script to run the singles/plurals/stages you are making, since the "retry script" hotkey is disabled in packages. It's a bit tedious, though, and I'm not sure whether it's worth the time and effort.

Unfortunately, backspace is not bound to a virtual key, so you can't force the key to a certain press state.

... This leads me to believe that Danmakufu has a "default package" that runs singles and plurals, which you can't edit. It seems that the resource folder only contains the stage codes for running plurals and singles, which are used by this hypothetical "hidden default package".

Of course, this is just my speculation, and may not actually be the case. However, what I can do is suggest to mkm a function that can change the key for retrying scripts, just like I had requested SetSkipModeKey.

Are you, perhaps, making a kind of typing game in Danmakufu?

They're in the resource/script/ directory. Nothing about backspace, though.

[Chronojet ⚙ Dragon]

They're in the resource/script/ directory. Nothing about backspace, though.

It seems that the resource folder only contains the stage codes for running plurals and singles, which are used by this hypothetical "hidden default package".

Actually, it is possible to recreate Danmakufu's default script selection menu using these functions found here, and a bit of creativity:
http://www.geocities.co.jp/SiliconValley-Oakland/9951/pre/th_dnh_help_v3_data/func_common_std.html#sec_path

However, as I said it's probably not worth the time and effort required to create such a replication script.

I could always ask mkm about how script selection is handled (like a package or whatever) in order to support or refute my hypothesis, but, again...

[Sage Ω (Ultima)]

I meant more like the Engine throwing errors because there is no such commondata defined but it seems like SetAreaCommonData already behaves like setting it but it is non existent or so 

Oh well.

That is what the default value is for when getting the commondata. If there were no default data then it would error or crash like using Obj_GetValue instead of GetValueD on an object that doesn't have the variable you set to it defined.

EDIT: Also, the backspace restart function is a line of code found in the TStageScene task of the default package system, of course the package is not available for us to edit for the standard danmakufu GUI but you can find the line of code in that task for creating your own package system.

[Uruwi]

Why does StartStageScene throw an error about not finding the player? I've called SetStageIndex, SetStageMainScript, and SetStagePlayerScript, and don't know what I missed.

[Chronojet ⚙ Dragon]

Why does StartStageScene throw an error about not finding the player? I've called SetStageIndex, SetStageMainScript, and SetStagePlayerScript, and don't know what I missed.

#Player[] in the header of the script must contain the path to the player as well, (which I find to be kind of dumb, but eh.)

[Vampi]

Semi related, speaking of CommonDataAreas, why is it that when I try get the data it returns me the default value instead?

For example SetAreaCommonData("BGM","Vol",100). If I want to GetAreaCommonData("BGM","Vol",0) it will always keep returning me the 0 instead of the 100 as stored. Am I missing something?

U dont forget this?
CreateCommonDataArea
SetAreaCommonData
SaveCommonDataAreaA1
LoadCommonDataAreaA1
GetAreaCommonData

[Uruwi]

Do the text file object functions fail to handle UTF-16LE correctly? Of course, you can try encoding your file in ShiftJIS, until you want to translate your game into something such as Korean.

[Chronojet ⚙ Dragon]

Do the text file object functions fail to handle UTF-16LE correctly? Of course, you can try encoding your file in ShiftJIS, until you want to translate your game into something such as Korean.

Unicode little endian...? I'm sure it should work properly. All of my codes are encoded with UTF-8 without the Byte order mark, though.

[Uruwi]

Unicode little endian...? I'm sure it should work properly. All of my codes are encoded with UTF-8 without the Byte order mark, though.

I wasn't talking about source code. I was talking about reading text files, and apparently mkm hasn't taken the relevant functions into account when making the transition to Unicode.

(and under what conditions does ObjSound_Play fail?)

[Chronojet ⚙ Dragon]

I wouldn't be surprised if he did, I mean he didn't even implement GetFilePathList (or whatever that new function introduced in pre6 was callled) even though he listed it as a new addition.

Did you use ObjSound_Load? Beyond that I think I'll have to take a look in your code for details

[Baron_Blade]

I swear, each problem I have is worse than the last. I'm trying to make a Last Word package with a script that allows you to play through them all. But, I have two big problems. I can't load sounds or images through the bosses' scripts except for the boss herself, and when I try to play the plural script, Danmakufu crashes. No error, just a crash. I've compared everything in my function_lib to my other plurals and it should work just fine.

[Uruwi]

I wouldn't be surprised if he did, I mean he didn't even implement GetFilePathList (or whatever that new function introduced in pre6 was callled) even though he listed it as a new addition.

Can I contact him about the problem then, or should I leave that for someone else?

Did you use ObjSound_Load? Beyond that I think I'll have to take a look in your code for details

Yes, I did. However, the sound object is created in a package script and stored into common data.

[Chronojet ⚙ Dragon]

You can contact him. Anyone can contact him. Over in LOCAA, we tend to just have one person collect bug reports from the other users and put it all under his name. Usually, that person is Ultima.

---

The way I do it in Digital Earth: Contest is like so:

I use User Events to notify the package script from the stage scripts.
Each time the package script receives a notification, the sound will be played. That means, the package script's @Event will contain the code for calling ObjSound_Play, instead of the stage script.

[Uruwi]

You can contact him. Anyone can contact him. Over in LOCAA, we tend to just have one person collect bug reports from the other users and put it all under his name. Usually, that person is Ultima.

---

The way I do it in Digital Earth: Contest is like so:

I use User Events to notify the package script from the stage scripts.
Each time the package script receives a notification, the sound will be played. That means, the package script's @Event will contain the code for calling ObjSound_Play, instead of the stage script.

Thanks for your explanation of what you do. Anyway, I'll wait until there's an option for which encoding to choose when reading text files. Barring that, I'll just have to use a Python script to convert all the text files into a binary format.

[Sage Ω (Ultima)]

I already contacted mkm about my GetFilePathList() request not actually being implemented. He said that the function was named GetFileList() and that it will get its intended name in the next update.

[Uruwi]

I already contacted mkm about my GetFilePathList() request not actually being implemented. He said that the function was named GetFileList() and that it will get its intended name in the next update.

I was asking about the inability to choose an encoding for reading from text file objects.

[Sage Ω (Ultima)]

I was asking about the inability to choose an encoding for reading from text file objects.

Code: [Select]

ObjFileB_SetByteOrder()
　　ENDIAN_LITTLE：リトルエンディアン
　　ENDIAN_BIG：ビッグエンディアン

ObjFileB_SetCharacterCode()
　　CODE_ACP：ANSI (Shift-JIS)
　　CODE_UTF8：UTF8
　　CODE_UTF16LE：UTF16LE
　　CODE_UTF16BE：UTF16BE

I'm pretty sure these two functions can do what you're asking.

[Uruwi]

Code: [Select]

ObjFileB_SetByteOrder()
　　ENDIAN_LITTLE：リトルエンディアン
　　ENDIAN_BIG：ビッグエンディアン

ObjFileB_SetCharacterCode()
　　CODE_ACP：ANSI (Shift-JIS)
　　CODE_UTF8：UTF8
　　CODE_UTF16LE：UTF16LE
　　CODE_UTF16BE：UTF16BE

I'm pretty sure there's no difference between a text file and a binary file.

I believe those apply only to binary file objects. In fact, using these functions on a text file object doesn't even raise an error - it simply fails silently.

[Sage Ω (Ultima)]

Indeed they do, currently there is no way to change the reading for normal text files. You can try asking mkm about it however, it may not be implemented right away. Requests like those tend to get added to what we in locaa call "the list". Its pretty big and contains many requested functions that have not been implemented, or bugs that have not been fixed yet. Informing mkm about it is still worth a try, if it wasn't for people requesting specific visual effect capabilities (all added to "the list") we may not have had HLSL implemented.

You may have to find some clever workaround in order to achieve what you want, what that is, I can't help there.

[Uruwi]

Indeed they do, currently there is no way to change the reading for normal text files. You can try asking mkm about it however, it may not be implemented right away. Requests like those tend to get added to what we in locaa call "the list". Its pretty big and contains many requested functions that have not been implemented, or bugs that have not been fixed yet. Informing mkm about it is still worth a try, if it wasn't for people requesting specific visual effect capabilities (all added to "the list") we may not have had HLSL implemented.

You may have to find some clever workaround in order to achieve what you want, what that is, I can't help there.

I can't seem to post to the official forums...

Sorry for annoying you.

YAE (Yet Another Edit): Those two functions don't even work on binary files.
YaeMk.II: I realized that it was due to my Python script reading text data in ANSI and not UTF-8.
YaeMk.III: Now the text won't display the proper characters. The guillemets appear as boxed question marks, and the letters with carons appear as unboxed question marks.
YaeMk.IV: Using the default font instead of Consolas solves the problems with guillemets, but the accented letters still appear incorrectly.
I Hope This Is The Last One I Have To Make: Apparently another bug on mkm's part. ObjFileB_ReadString apparently can't return a string with any characters outside the Shift-JIS character set, regardless of the encoding set for it.

[Sparen]

OK. So here's a series of questions I really want to have answered for future use.

Firstly, what exactly does 'ref' mean in the Log Window? Is it how many copies of the file have been saved to RAM? Is it how many objects are using the file? Additionally, is the memory address the RAM memory address

Secondly, how does loading of assets such as sound and images actually work? If you do not load a file with LoadImage or LoadImageInThread, does it load automatically? If so, does RemoveTexture actually purge a loaded file from memory? I ask because purging large files from memory after they have been used for the last time is a reliable way to lower the amount of stress that is placed on ones computer. I ask because I load all of my main images and sounds during a loading phase of the package, and I am experimenting with a way to purge those files until they are needed again. More specifically, I plan to LoadImage in @Loading at the start of the stage for images that have already been loaded (in case the stage is replayed), and then RemoveTexture in @Finalize, so that the files are removed. Doing this, however, leaves references to the file, so I cannot be sure that it has been purged from Danmakufu's memory.

Finally, a question about archive files. I have noticed that archive files do not work based on a directory structure in that if you archive all the contents of a folder called 'img' and leave the archive in the folder img, Danmakufu cannot find it, but if placed outside of the directory 'img', Danmakufu will be able to read the files inside the archive. The path after loading still shows that the folder 'img' is in the file path despite the images' location on the computer not being in img but rather in an archive located one directory above img. How does this affect multiple-archive systems where some filenames are the same, and how does Danmakufu's archive system actually WORK?

Thank you.

[Sage Ω (Ultima)]

1: ref is the amount of times something was used while it is loaded. (the number is abit weird and doesn't always add up correctly)

2: Textures load when used regardless unless it already exists. RemoveTexture does exactly what it says. (I'm very disappointed that you asked this. You are no stranger to danmakufu at all, and the second question could've been answered on your own by testing it.)

Removing stuff after every script ends is kind of pointless unless its menu related things, you're still gonna have many textures hanging around during the stage; the performance increase is insignificant and won't weigh over the other more important things that will slow dmf down. Things that will have the biggest impact on danmakufu are: excessive running tasks(800+), excessive heavy math calculations, too many shots on screen with enabled hitboxes, too many shots on screen with more than one hitbox enabled(the cause of lag for curvy lasers), rendering the screen to a rendertarget. Those things are what you should be worrying about, not loading a bunch of textures and sound effects.

3: Archive files save the file path to the directory as if the files were not compressed, and the archive file was a folder. Danmakufu doesn't check by filename alone, it checks the file path exactly. You cannot however, compress different archives into one big archive. The tool itself works as a lock to prevent others from touching resources, there is some compression but it isn't anything big.