Unpacked XML Missing (count="#") Attribute -> unable to import in EoF

[ForgeOn]

Hello fabianosan and the other Toolkit developers! (or anyone who might have a grasp on this situation)

 

I have made another post in the EoF forum regarding this already:

http://customsforge.com/topic/2529-importing-unpacked-xml-failure-check-log/?do=findComment&comment=13423

 

The gist of it is as it says in the title of this post:

 

I unpacked a working Custom DLC using the latest toolkit (v2.5.0.0 - Nightly Build 3d3bec6a7567).

Was unable to import the instrument XMLs with an error about count tags.

On closer inspection the unpacked XML differs from EoF (r1304) generated XMLs in that it has no count="#" attributes. (Among other smaller differences.)

 

 

I've been trying to edit the XML myself to get a working import but while I have successfully gotten the XML to import (with a lot of time consuming edits) the import still has chaotic bpm changes.

 

Any help on resolving this for the custom I am working on and for future unpacking/importing would be greatly appreciated!

 

Thanks so much!

 

 

(please check the post in the EoF forum for more details!)

[fabianosan]

First of all. EoF have not support for rs2 xml import at this time, only for rs1, maybe @@raynebc can help you.

 

About count tag it's not necessary for toolkit and will not be generated, the only thing that we need is the correct imported data.

 

The lot of differences that you have is not used for rs2 too, toolkit make a clean xml when convert from sng, only with used tags for rs2.

 

Thanks.

[ForgeOn]

Thank you so much for the reply, @@fabianosan!

If I understand you correctly it would appear to be an EoF issue then so I will continue the thread in that section!

Thanks again! :)

[raynebc]

The beat timing format hasn't changed between RS1 and RS2, you would want to compare those between the original and unpacked XML files.

[fabianosan]

Is not a EoF issue, it's a feature not support at this time.

 

And have no issue with beat time, only if have 0 at right is not added. Have not problem with time, the value is the same.

[ForgeOn]

The CDLC I unpacked is a RS2 CDLC, and I don't know what you mean by not an issue? I get an error when trying to import. That seems like an issue. Anyway, raynebc has said in the other thread that he will remove the count requirement; that may well fix my problem! :)

 

Thank you both so much for your hard work!

[raynebc]

I've been working on the changes to EOF's Rocksmith import, and I may have stumbled onto an actual bug with the toolkit. I've been testing with some unpacked XML files that Kent0348 posted here:

http://customsforge.com/topic/1529-latest-eof-releases-r1304/page-3?do=findComment&comment=10046

 

In muse-toada_lead.xml, it looks like all chord tags (starting at line 2197) reference invalid chord IDs. There are only 27 chord templates in the XML, but almost referenced chord IDs are number 28 and higher. Many of the handshape tags also reference invalid chord IDs.

[fabianosan]

@ForgeOn, issue is when a software have a feature and you try to use and can't, you need to wait, when EoF supports rs2 xml import, not only rs1 and you try to use and can't, so you will report the error.

 

Until this, just wait.

 

@raynebc, if you or user can, post here or in the github the original and unpacked xml for analysis.

 

Thanks.

[ForgeOn]

@@fabianosan, I appreciate what you are saying. That does seem like a feature request. If it makes any difference, the XML I was importing doesn't have any special techniques specific to RS2, it is all single notes. Are there other differences between RS! and RS2 XMLs?

 

In any case, I did not realize EoF doesn't do RS2 XML Importing yet so I will be patient!

 

Thanks! :)

[Alex360]

well we need also escape unallowed xml chars, just saying that we don't do that..

[raynebc]

if you or user can, post here or in the github the original and unpacked xml for analysis.

I'll ask if he can post the files, but if anybody else is able to test this by packing and unpacking an arrangement, it should let us know if the toolkit's unpack is writing incorrect chord IDs.

[raynebc]

Unpacked official files don't seem to have the chord ID glitch:

http://customsforge.com/topic/2574-unpacked-xml-from-official-dlc-wont-import-into/?do=findComment&comment=13832

 

This just leaves the testing of custom DLC being packed and unpacked.

[ForgeOn]

Hey reynebc! I'm happy to help if I can. What would I be looking for exactly in the unpacked xml?

 

Edit:

I only see the xml files linked. Can't pack it without the audio can I?

[raynebc]

The XML files that Kent0348 posted are already broken. This test would need to be done with a clean XML file.

[ForgeOn]

Ahh ok, well I can't do it right now but if it still needs doing tomorrow, I'm your guy!

[raynebc]

Kent0348 sent me the original XML file that seemed to unpack incorrectly:

http://www.sendspace.com/file/9ubvyq

 

This one doesn't seem to have any errors and EOF is able to import it. If somebody can test packing this arrangement and unpacking it, it would be worth seeing if it still unpacks with errors like invalid chord ID numbers.

[fabianosan]

I'll take a look.

 

I downloaded song: "Muse - Thoughts Of A Dying Atheist", take the DDC generated XML and did unpack on .SNG and the only differences I see is:

 

- Decimal places (make no difference to toolkit and the game);

- Count attributes (make no difference to toolkit and the game);

- The order of some attributes comparing with DDC XML (but the order is equal to EoF, because I did changes to make equal);

- LinkedDiff (is not used on RS2014);

- The TranscriptionTrack element (sng don't provide this info, but can be calculated, but is not required for toolkit and the game to work);

- "-arp" sufix in chord display name;

 

Only the last item make difference and is simple to fix, I'll fix it.

 

Here is the comparisson:

 

Have another example of invalid unpacked XML?

[raynebc]

I don't have any other examples at this time.

[fabianosan]

Missing "-arp" and "-nop" fixed in displayName of chords.

[raynebc]

What does the "-nop" suffix do for chords?

[fabianosan]

You can see in this lessons:

 

ge_e11_rhopo3.xml

ge_ldoublestops2_lsn19.xml

 

with "_nop"

 

I believe that is chords with "HOPO" technique, HammerOn or PullOf.

[raynebc]

Do chords where all chordnotes have hammer on or pull off technique still score correctly? Is there a need for EOF to try to write "-nop" chord templates?

[fabianosan]

You can try to write this tags, but I believe that we need some testes before.

 

When all notes have hammeron or pulloff you need to mark "hopo" attribute in chord element, so you can add "-nop" to displayName, but as I wrote.

 

We need to do some tests before to view the beharvior.