This topic was archived here:

https://web.archive.org/web/20230000000000fw_/https://www.zenhax.com/viewtopic.php?t=2732

TheUkrainianBard, posted Thu Jul 07, 2016 11:58 pm (15104)

---

Steam store

Archives .pac - Aluigi's script for Neptunia Re;Birth1.
Archives .cl3 - provided file.
Archives .arc - provided file.

No re-import.

UPD 2018-02-01: after a very long while I've finally made it for .arc files which are used in Megadimension Neptunia VII DLCs (.pac -> .arc -> .cl3)

neptunia_arc.zip

pc_megadimension_neptunia_vii_cl3.zip

Panzerdroid, posted Fri Jul 08, 2016 2:20 am (15108)

---

Reimport is impossible. Only neptunia decompression is available in QuickBMS.

Savage, posted Fri Jul 08, 2016 7:21 am (15116)

---

Panzerdroid wrote:

Reimport is impossible. Only neptunia decompression is available in QuickBMS.

Exactly i confirm this

Quote:

Error: unsupported compression 498 in reimport mode

Last script line before the error or that produced the error:
26 clog NAME OFFSET ZSIZE SIZE

Savage, posted Fri Jul 08, 2016 7:33 am (15118)

---

Here is some tools, in japanese i no have no idea how it works
https://steamcommunity.com/sharedfiles/ ... =409454600

https://mega.nz/#!ckAUlLQb!jZSDeRiE2KeP ... MUgORySzn8

TheUkrainianBard, posted Fri Jul 08, 2016 8:09 am (15121)

---

Panzerdroid, Savage
Thanks, I forgot how many factors there were in re-import.

There are tools named "kitserver" and "neptools", which re-route game's file requests (I dunno how to word it), but their confirmed compability is with Re;Birth games.

Savage, posted Fri Jul 08, 2016 8:43 am (15125)

---

Modding tool for the Hyperdimension Neptunia series games; enable packing/unpacking of resources from the games' custom archive format
https://github.com/agentOfChaos/oversized_syringe

Here is the code compression (i think)
https://github.com/agentOfChaos/oversiz ... ression.py

Code:

import struct
import os
from ovsylib.compresion_algos import yggdrasil
from ovsylib.aggressive_threading import Broker
from threading import Thread

intsize = 4


def checksize(binfile):
    old_file_position = binfile.tell()
    binfile.seek(0, os.SEEK_END)
    size = binfile.tell()
    binfile.seek(old_file_position, os.SEEK_SET)
    return size


class BadMagicNum(Exception):
    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)

class algo1:

    def __init__(self, id, size):
        self.uncomp_size = 0
        self.comp_size = 0
        self.rootaddress = 0
        self.header_write_back_offset = 0
        self.id = id
        self.size = size
        self.bitstream = None

    def loadSubHeaderFromFile(self, binfile):
        self.uncomp_size = struct.unpack("I", binfile.read(intsize))[0]
        self.comp_size = struct.unpack("I", binfile.read(intsize))[0]
        self.rootaddress = struct.unpack("I", binfile.read(intsize))[0]

    def decompress(self, binfile, metadata_offset, savefile, debuggy=False):
        binfile.seek(metadata_offset   self.rootaddress, 0)
        yggdrasil.uncompress(binfile, savefile, self.comp_size, self.uncomp_size, debuggy=debuggy)

    def writeSubHeader(self, savefile, dry_run=False):
        self.header_write_back_offset = savefile.tell()
        if not dry_run:
            savefile.write(struct.pack("I", self.uncomp_size))
            savefile.write(struct.pack("I", self.comp_size))
            savefile.write(struct.pack("I", self.rootaddress))

    def writeToFile(self, savefile, metadata_offset, dry_run=False):
        """ actually writes out the compressed data. Can be run only after compress """
        assert self.bitstream is not None
        self.rootaddress = savefile.tell() - metadata_offset
        savefile.write(self.bitstream.tobytes())
        afterwrite_offset = savefile.tell()
        savefile.seek(self.header_write_back_offset, 0)
        self.writeSubHeader(savefile, dry_run=dry_run)
        savefile.seek(afterwrite_offset, 0)
        self.bitstream = None  # free memory

    def compress(self, sourcefile, debuggy=False):
        """ this method can be parallelized, given different file handles to the various threads
        :return: self """
        total_filesize = checksize(sourcefile)
        self.uncomp_size = self.size
        if (self.size * (self.id   1)) > total_filesize:
            self.uncomp_size = total_filesize - (self.size * self.id)
        self.bitstream = yggdrasil.compress(sourcefile,
                                       self.size * self.id,
                                       (self.size * self.id)   self.uncomp_size,
                                       debuggy=debuggy)
        self.comp_size = self.bitstream.buffer_info()[1]
        return self

    def printInfo(self):
        print("partial comp size: %d ; partial uncomp size: %d ; relative root address: %x"
              % (self.comp_size, self.uncomp_size, self.rootaddress))


class after_comp_callback:
    """ used as a callback to fill the 'compressed size' value in places where it was
    needed, but we didn't know it yet """

    def __init__(self, address, binfile):
        self.address = address
        self.binfile = binfile

    def compressed(self, length, dry_run=False):
        savedpos = self.binfile.tell()
        self.binfile.seek(self.address, 0)
        if not dry_run:
            self.binfile.write(struct.pack("I", length))
        self.binfile.seek(savedpos, 0)


# need to be standalone, for multiprocess to pickle it
def spawn(chunk, sourcefile_name, debuggy):
    with open(sourcefile_name, "rb") as sourcefile:
        return chunk.compress(sourcefile, debuggy=debuggy)


class chuunicomp:

    default_chunksize = 0x20000

    def __init__(self):
        self.magicseq = 0
        self.chunk_num = 0
        self.chunksize = 0
        self.header_size = 0
        self.chunks = []
        self.aftercompress_callback_obj = None

    def fromBinfile(self, binfile):
        self.magicseq = struct.unpack("I", binfile.read(intsize))[0]
        self.chunk_num = struct.unpack("I", binfile.read(intsize))[0]
        self.chunksize = struct.unpack("I", binfile.read(intsize))[0]
        self.header_size = struct.unpack("I", binfile.read(intsize))[0]
        self.chunks = self.prepareChunks(binfile)
        if self.magicseq != 0x1234:
            raise BadMagicNum(self.magicseq)

    def fromFutureImport(self, chunksnum):
        self.magicseq = 0x1234
        self.chunk_num = chunksnum
        self.chunksize = self.default_chunksize
        self.chunks = self.prepareChunks(None)
        self.header_size = (4   (3 * chunksnum)) * 4

    def prepareChunks(self, binfile=None):
        body = []
        for i in range(self.chunk_num):
            piece = algo1(i, self.chunksize)
            if binfile is not None:
                piece.loadSubHeaderFromFile(binfile)
            body.append(piece)
        return body

    def decompress(self, binfile, savefile, debuggy=False):
        """ we don't read the header at decompression time, since we have already acquired the data
        via fromBinfile(). Hence we simply skip the header """
        binfile.seek(self.header_size, 1)
        metadata_offset = binfile.tell()
        for chunk in self.chunks:
            chunk.decompress(binfile, metadata_offset, savefile, debuggy=debuggy)

    def writeHeader(self, savefile, dry_run=False):
        if not dry_run:
            savefile.write(struct.pack("I", self.magicseq))
            savefile.write(struct.pack("I", self.chunk_num))
            savefile.write(struct.pack("I", self.chunksize))
            savefile.write(struct.pack("I", self.header_size))

    def compress(self, sourcefile_name, savefile, dry_run=False, debuggy=False):
        """ run the compression algorithm for each chunk, in parallel. Note that the
        caller has the duty to set the file seek in the right position before calling this. """
        totalcomp = 0
        threads = Broker(len(self.chunks), debugmode=debuggy, greed=1)

        def spawnAll(chunklist):
            for chunk in chunklist:
                threads.appendNfire(spawn, (chunk, sourcefile_name, debuggy))

        self.writeHeader(savefile, dry_run=dry_run)
        for chunk in self.chunks:
            chunk.writeSubHeader(savefile, dry_run=dry_run)
        metadata_offset = savefile.tell()

        # this thread will feed the broker with tasks
        Thread(target=spawnAll, args=(self.chunks,)).start()

        # gather all the results, write them in sequence
        collected = 0
        while collected < len(self.chunks):
            for partial in threads.collect():
                totalcomp  = partial.bitstream.buffer_info()[1]
                partial.writeToFile(savefile, metadata_offset, dry_run=dry_run)
                collected  = 1
        threads.stop()
        if self.aftercompress_callback_obj is not None:
            self.aftercompress_callback_obj.compressed(totalcomp   self.header_size, dry_run=dry_run)
        return totalcomp   4

    def printInfo(self):
        print("     Compression chunks: %d , header size: x" % (self.chunk_num, self.header_size))
        for chunk in self.chunks:
            print("     chunk #d: " % (self.chunks.index(chunk),), end="")
            chunk.printInfo()

I don't know if can be implemented in the next version of quickbms for reimport

Curxe, posted Sun Jul 31, 2016 5:50 am (15969)

---

Savage wrote:

Modding tool for the Hyperdimension Neptunia series games; enable packing/unpacking of resources from the games' custom archive format
https://github.com/agentOfChaos/oversized_syringe

Here is the code compression (i think)
https://github.com/agentOfChaos/oversiz ... ression.py

Code:

import struct
import os
from ovsylib.compresion_algos import yggdrasil
from ovsylib.aggressive_threading import Broker
from threading import Thread

intsize = 4


def checksize(binfile):
    old_file_position = binfile.tell()
    binfile.seek(0, os.SEEK_END)
    size = binfile.tell()
    binfile.seek(old_file_position, os.SEEK_SET)
    return size


class BadMagicNum(Exception):
    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)

class algo1:

    def __init__(self, id, size):
        self.uncomp_size = 0
        self.comp_size = 0
        self.rootaddress = 0
        self.header_write_back_offset = 0
        self.id = id
        self.size = size
        self.bitstream = None

    def loadSubHeaderFromFile(self, binfile):
        self.uncomp_size = struct.unpack("I", binfile.read(intsize))[0]
        self.comp_size = struct.unpack("I", binfile.read(intsize))[0]
        self.rootaddress = struct.unpack("I", binfile.read(intsize))[0]

    def decompress(self, binfile, metadata_offset, savefile, debuggy=False):
        binfile.seek(metadata_offset   self.rootaddress, 0)
        yggdrasil.uncompress(binfile, savefile, self.comp_size, self.uncomp_size, debuggy=debuggy)

    def writeSubHeader(self, savefile, dry_run=False):
        self.header_write_back_offset = savefile.tell()
        if not dry_run:
            savefile.write(struct.pack("I", self.uncomp_size))
            savefile.write(struct.pack("I", self.comp_size))
            savefile.write(struct.pack("I", self.rootaddress))

    def writeToFile(self, savefile, metadata_offset, dry_run=False):
        """ actually writes out the compressed data. Can be run only after compress """
        assert self.bitstream is not None
        self.rootaddress = savefile.tell() - metadata_offset
        savefile.write(self.bitstream.tobytes())
        afterwrite_offset = savefile.tell()
        savefile.seek(self.header_write_back_offset, 0)
        self.writeSubHeader(savefile, dry_run=dry_run)
        savefile.seek(afterwrite_offset, 0)
        self.bitstream = None  # free memory

    def compress(self, sourcefile, debuggy=False):
        """ this method can be parallelized, given different file handles to the various threads
        :return: self """
        total_filesize = checksize(sourcefile)
        self.uncomp_size = self.size
        if (self.size * (self.id   1)) > total_filesize:
            self.uncomp_size = total_filesize - (self.size * self.id)
        self.bitstream = yggdrasil.compress(sourcefile,
                                       self.size * self.id,
                                       (self.size * self.id)   self.uncomp_size,
                                       debuggy=debuggy)
        self.comp_size = self.bitstream.buffer_info()[1]
        return self

    def printInfo(self):
        print("partial comp size: %d ; partial uncomp size: %d ; relative root address: %x"
              % (self.comp_size, self.uncomp_size, self.rootaddress))


class after_comp_callback:
    """ used as a callback to fill the 'compressed size' value in places where it was
    needed, but we didn't know it yet """

    def __init__(self, address, binfile):
        self.address = address
        self.binfile = binfile

    def compressed(self, length, dry_run=False):
        savedpos = self.binfile.tell()
        self.binfile.seek(self.address, 0)
        if not dry_run:
            self.binfile.write(struct.pack("I", length))
        self.binfile.seek(savedpos, 0)


# need to be standalone, for multiprocess to pickle it
def spawn(chunk, sourcefile_name, debuggy):
    with open(sourcefile_name, "rb") as sourcefile:
        return chunk.compress(sourcefile, debuggy=debuggy)


class chuunicomp:

    default_chunksize = 0x20000

    def __init__(self):
        self.magicseq = 0
        self.chunk_num = 0
        self.chunksize = 0
        self.header_size = 0
        self.chunks = []
        self.aftercompress_callback_obj = None

    def fromBinfile(self, binfile):
        self.magicseq = struct.unpack("I", binfile.read(intsize))[0]
        self.chunk_num = struct.unpack("I", binfile.read(intsize))[0]
        self.chunksize = struct.unpack("I", binfile.read(intsize))[0]
        self.header_size = struct.unpack("I", binfile.read(intsize))[0]
        self.chunks = self.prepareChunks(binfile)
        if self.magicseq != 0x1234:
            raise BadMagicNum(self.magicseq)

    def fromFutureImport(self, chunksnum):
        self.magicseq = 0x1234
        self.chunk_num = chunksnum
        self.chunksize = self.default_chunksize
        self.chunks = self.prepareChunks(None)
        self.header_size = (4   (3 * chunksnum)) * 4

    def prepareChunks(self, binfile=None):
        body = []
        for i in range(self.chunk_num):
            piece = algo1(i, self.chunksize)
            if binfile is not None:
                piece.loadSubHeaderFromFile(binfile)
            body.append(piece)
        return body

    def decompress(self, binfile, savefile, debuggy=False):
        """ we don't read the header at decompression time, since we have already acquired the data
        via fromBinfile(). Hence we simply skip the header """
        binfile.seek(self.header_size, 1)
        metadata_offset = binfile.tell()
        for chunk in self.chunks:
            chunk.decompress(binfile, metadata_offset, savefile, debuggy=debuggy)

    def writeHeader(self, savefile, dry_run=False):
        if not dry_run:
            savefile.write(struct.pack("I", self.magicseq))
            savefile.write(struct.pack("I", self.chunk_num))
            savefile.write(struct.pack("I", self.chunksize))
            savefile.write(struct.pack("I", self.header_size))

    def compress(self, sourcefile_name, savefile, dry_run=False, debuggy=False):
        """ run the compression algorithm for each chunk, in parallel. Note that the
        caller has the duty to set the file seek in the right position before calling this. """
        totalcomp = 0
        threads = Broker(len(self.chunks), debugmode=debuggy, greed=1)

        def spawnAll(chunklist):
            for chunk in chunklist:
                threads.appendNfire(spawn, (chunk, sourcefile_name, debuggy))

        self.writeHeader(savefile, dry_run=dry_run)
        for chunk in self.chunks:
            chunk.writeSubHeader(savefile, dry_run=dry_run)
        metadata_offset = savefile.tell()

        # this thread will feed the broker with tasks
        Thread(target=spawnAll, args=(self.chunks,)).start()

        # gather all the results, write them in sequence
        collected = 0
        while collected < len(self.chunks):
            for partial in threads.collect():
                totalcomp  = partial.bitstream.buffer_info()[1]
                partial.writeToFile(savefile, metadata_offset, dry_run=dry_run)
                collected  = 1
        threads.stop()
        if self.aftercompress_callback_obj is not None:
            self.aftercompress_callback_obj.compressed(totalcomp   self.header_size, dry_run=dry_run)
        return totalcomp   4

    def printInfo(self):
        print("     Compression chunks: %d , header size: x" % (self.chunk_num, self.header_size))
        for chunk in self.chunks:
            print("     chunk #d: " % (self.chunks.index(chunk),), end="")
            chunk.printInfo()

I don't know if can be implemented in the next version of quickbms for reimport

Ok, how do I use the code? This is my first time breaking down game file.

TheUkrainianBard, posted Sun Jul 31, 2016 7:52 am (15973)

---

Heck if I know, I have no experience with Python whatsoever.

TheUkrainianBard, posted Wed Jan 31, 2018 11:34 pm (32093)

---

After a very long while I've finally made it for .arc files which are used in Megadimension Neptunia VII DLCs

Razzee, posted Fri Mar 29, 2019 7:45 pm (46189)

---

Hello. Is this add-on compatible with Neptunia VII only? I'm trying to extract .cl3 files from Trinity Universe folder, a precursor of the Neptunia franchise.



Found a add-on similar to yours, but it gives the same error. Tried NepTools, and it didn't work either. I'm out of ideas.

The file:
https://mega.nz/#!Q893HaAC!tcCRuwNsHAVV ... nHcZ45nDHw

TheUkrainianBard, posted Fri Mar 29, 2019 8:21 pm (46190)

---

PS3 is a Big Endian platform, try this.

neptunia_cl3.zip

Razzee, posted Sat Mar 30, 2019 4:02 pm (46203)

---

Yes, it worked. Thanks for the fast response.

Yesterday I edited a main.dat (which I didn't have to extract from a .cl3) and once I added it to the game, every text inside that file disappeared.

[spoiler]

[/spoiler]

The font probably does not support accents, but I wrote that way to check. I believe it happened because this file has a limited amount of characters as almost every console game, am I right?

TheUkrainianBard, posted Sat Mar 30, 2019 4:17 pm (46204)

---

NepTools should work here