from struct import unpack as U
import csv
import sys

big_endian = False

def readint(f):
    if big_endian:
        return U(">i", f.read(4))[0]
    else:
        return U("<i", f.read(4))[0]

def dumpy(f, mapping):
    f.seek(0, 2)
    fsize = f.tell()
    f.seek(0, 0)

    something = readint(f)
    assert something == 0x10000, something

    size = readint(f)
    assert size == fsize, size

    unk = readint(f)
    if big_endian:
        assert unk == 0x01FF0000, unk
    else:
        assert unk == 1, unk

    count = readint(f)

    offset_count = readint(f)

    somecount1 = readint(f)  # still unknown!
    something = readint(f)  # still unknown!

    starts = {}
    lengths = {}
    ids = []
    cumulative_length = 0
    previous_end = None

    for i in range(count):
        if big_endian:
            a, b, c = U(">iii", f.read(4 * 3))
        else:
            a, b, c = U("<iii", f.read(4 * 3))
        #print(f"{a:10}: {b:10} to {c:10}")
        length = c - b + 1

        assert a not in starts
        if previous_end is not None:
            assert a == previous_end

        starts[a] = b
        lengths[a] = length

        for i in range(length):
            ids.append(b + i)

        cumulative_length += length
        previous_end = a + length

    assert offset_count == cumulative_length

    offsets = []
    for i in range(offset_count):
        offsets.append(readint(f))

    for id, offset in zip(ids, offsets):
        if offset == 0:
            #mapping[id] = ""
            continue

        f.seek(offset)
        string = ""
        while True:
            char = f.read(2)
            if char == b"\0\0":
                break
            if big_endian:
                string += char.decode("utf-16be")
            else:
                string += char.decode("utf-16le")
        mapping[id] = string

fp = sys.argv[1]
fpo = sys.argv[2]

if len(sys.argv) > 3:
    big_endian = sys.argv[3] == "big"

en_mapping = {}
jp_mapping = {}

with open(fp, "rb") as f:
    dumpy(f, en_mapping)

with open(fp.replace("ENGLISH", "JAPANESE"), "rb") as f:
    dumpy(f, jp_mapping)

from collections import defaultdict
mappings = defaultdict(lambda: ["", ""])

for k, v in en_mapping.items():
    mappings[k][0] = v

for k, v in jp_mapping.items():
    mappings[k][1] = v

with open(fpo, "w", newline="", encoding="utf-8") as f:
    cw = csv.writer(f, dialect="excel-tab")
    for k in sorted(mappings.keys()):
        en_v, jp_v = mappings[k]
        cw.writerow([k, en_v, jp_v])