string-gecko.py - generate Wii gecko/ocarina codes to replace ASCII strings in game executables

string-gecko.py

# string-gecko.py by github.com/InvoxiPlayGames
# command usage: python string-gecko.py [dol|dump] [filename] [oldstring] [newstring] 
# code licensing: don't be a dick with it, credit for general usage would be nice but not required
#                 but if you use parts of this python script in your own project, credit is required

import struct
import sys

memorybase = 0x80000000
memorysize = 0x01800000 # size of Wii MEM1, i think
memory = [0] * memorysize

if sys.argv[1] == "dol":
    dolfilename = sys.argv[2]
    print("loading DOL file '%s'..." % (dolfilename))
    f = open(dolfilename, "rb")
    
    # read header
    f.seek(0)
    textoffsets = struct.unpack(">IIIIIII", f.read(7 * 4))
    dataoffsets = struct.unpack(">IIIIIIIIIII", f.read(11 * 4))
    textloadaddr = struct.unpack(">IIIIIII", f.read(7 * 4))
    dataloadaddr = struct.unpack(">IIIIIIIIIII", f.read(11 * 4))
    textsizes = struct.unpack(">IIIIIII", f.read(7 * 4))
    datasizes = struct.unpack(">IIIIIIIIIII", f.read(11 * 4))
    [bssaddr] = struct.unpack(">I", f.read(4))
    [bsssize] = struct.unpack(">I", f.read(4))
    [entrypoint] = struct.unpack(">I", f.read(4))
    # read data
    f.seek(0)
    doldata = f.read()
    # we don't need the file anymore
    f.close()

    print()
    if entrypoint >= memorybase and entrypoint < memorybase + memorysize:
        print("entrypoint:")
        print("  address: 0x%08x" % (entrypoint))
    else:
        print("entrypoint unused or invalid")
    print()
    if bssaddr >= memorybase and bssaddr < memorybase + memorysize:
        print("bss:")
        print("  address: 0x%08x" % (bssaddr))
        print("  length:  0x%x" % (bsssize))
    else:
        print("bss unused or invalid")
    
    print()
    for x in range(7):
        if textloadaddr[x] >= memorybase and textloadaddr[x] < memorybase + memorysize - textsizes[x]:
            print("text%i:" % (x+1))
            print("  address: 0x%08x" % (textloadaddr[x]))
            print("  length:  0x%x" % (textsizes[x]))
            print("  offset:  0x%x" % (textoffsets[x]))
            memory[textloadaddr[x]-memorybase:textloadaddr[x]+textsizes[x]-memorybase] = doldata[textoffsets[x]:textoffsets[x]+textsizes[x]]
        else:
            print("text%i unused or invalid" % (x+1))
    
    print()
    for x in range(11):
        if dataloadaddr[x] >= memorybase and dataloadaddr[x] < memorybase + memorysize - datasizes[x]:
            print("data%i:" % (x+1))
            print("  address: 0x%08x" % (dataloadaddr[x]))
            print("  length:  0x%x" % (datasizes[x]))
            print("  offset:  0x%x" % (dataoffsets[x]))
            memory[dataloadaddr[x]-memorybase:dataloadaddr[x]+datasizes[x]-memorybase] = doldata[dataoffsets[x]:dataoffsets[x]+datasizes[x]]
        else:
            print("data%i unused or invalid" % (x+1))
    
    print("loaded '%s' successfully" % (dolfilename))
elif sys.argv[1] == "dump":
    dumpfilename = sys.argv[2]
    print("loading memory dump file '%s'..." % (dumpfilename))
    f = open(dumpfilename, "rb")
    memory = f.read()
    f.close()
    print("loaded '%s' successfully" % (dumpfilename))

memory = bytearray(memory)

oldstring = sys.argv[3]
newstring = sys.argv[4]
codelength = len(oldstring)

if len(newstring) > len(oldstring):
    print("new string is larger than old string, you will overwrite memory!!")
    codelength = len(newstring)
elif len(oldstring) > len(newstring):
    # might add automatically replacing substrings later, but not right now
    print("new string is shorter than old string, will fill in the blank with null bytes")

if len(newstring) % 8 != 0:
    newstring = newstring + ('\0' * 0x8) # lazy padding to avoid problems

noinstance = False
lastinstance = 0
instances = []

print("searching for instances of %s" % (oldstring))
while not noinstance:
    lastinstance = memory.find(bytearray(oldstring, "utf8"), lastinstance + len(oldstring), memorysize)
    if lastinstance > 0:
        print("found string at %08x" % (lastinstance))
        instances.append(lastinstance)
    else:
        noinstance = True

print()
print("code:")
for instance in instances:
    print("%08x %08x" % (instance | 0x06000000, codelength))
    x = 0
    while x < (codelength - 1):
        
        print("%02x%02x%02x%02x %02x%02x%02x%02x" % ( ord(newstring[x]), ord(newstring[x+1]), ord(newstring[x+2]), ord(newstring[x+3]), ord(newstring[x+4]), ord(newstring[x+5]), ord(newstring[x+6]), ord(newstring[x+7]) ))
        x += 8