14 replies to this topic

[annullus]

Been playing around with Lego Harry Potter years 1-4 on pc trying to see the limits of file extraction. I can get all the files extracted using the Quickbms Script however it seems to be very glitched in the content that i end up with. For Example in game the harry potter minifig is bugged and his legs wont move, and other minifigs have graphical corruptions. I just want to know if anyone else has tried playing around with this games extracted files and if they have gotten better results than I.

[McJobless]

IIRC the LEGO Harry Potter games were made after the most recent version of the script. So, all the files might get extracted, but you most likely will get that corruption. Quick question, did you actually change anything yet, or have you just extracted the files and removed the .dat files to see if it runs?

[annullus]

I have tried both, btw i am using the v5.0 script which seems to get all the files out properly.

# Traveller's Tales games DAT files extractor (script 0.5)
# LEGO Harry Potter
# LEGO Star Wars III
# LEGO Pirates of the Caribbean
#
# this script is expansion to the original script from aluigi
# original script for QuickBMS http://aluigi.org/papers.htm#quickbms
comtype lz2k
get INFO_OFF long
if INFO_OFF & 0x80000000
math INFO_OFF ^= 0xffffffff
math INFO_OFF <<= 8
math INFO_OFF += 0x100
endif
get INFO_SIZE long
log MEMORY_FILE INFO_OFF INFO_SIZE
get TYPE_BOH long MEMORY_FILE
get FILES long MEMORY_FILE
savepos INFO_OFF MEMORY_FILE
math TMP = FILES
math TMP *= 16
math NAME_INFO = INFO_OFF
math NAME_INFO += TMP
goto NAME_INFO MEMORY_FILE
get NAMES long MEMORY_FILE
savepos NAME_INFO MEMORY_FILE
math TMP = NAMES
math TMP *= 8
math NAME_OFF = NAME_INFO
math NAME_OFF += TMP
goto NAME_OFF MEMORY_FILE
get NAMECRC_OFF long MEMORY_FILE
savepos NAME_OFF MEMORY_FILE
math NAMECRC_OFF += NAME_OFF
goto NAMECRC_OFF MEMORY_FILE
for i = 0 < FILES
get CRC long MEMORY_FILE
putarray 0 i CRC
next i
if TYPE_BOH <= -2
get DUMMY1 long MEMORY_FILE
get DUMMY2 long MEMORY_FILE
if DUMMY1 > 0
	 print "Error: not implemented yet"
	 cleanexit
endif
endif
# print "files:	 %FILES%"
# print "names:	 %NAMES%"
# print "info_off: %INFO_OFF%"
# print "info_size: %INFO_SIZE%"
# print "name_info: %NAME_INFO%"
# print "name_off: %NAME_OFF%"
# print "namecrc_off: %NAMECRC_OFF%"
set NAMEZ long 0
set FULLNAME string ""
set FULLPATH string ""
for i = 0 < FILES
do
	 goto NAME_INFO MEMORY_FILE
	 get NEXT short MEMORY_FILE
	 get PREV short MEMORY_FILE
	 get OFF long MEMORY_FILE
	 savepos NAME_INFO MEMORY_FILE
	 math OFF += NAME_OFF
	 goto OFF MEMORY_FILE
	 get NAME string MEMORY_FILE
	 if NEXT & 0x8000 # 16bit sign conversion
		 math NEXT |= 0xffff0000
	 endif
	 if PREV != 0
		 getarray FULLPATH 1 PREV
	 endif
	 putarray 1 NAMEZ FULLPATH
	 if NEXT > 0 # folder
		 if NAME != ""
			 string FULLPATH += NAME
			 string FULLPATH += \
		 endif
	 else #file
	 endif
	 math NAMEZ += 1
while NEXT > 0
set FULLNAME string FULLPATH
string FULLNAME += NAME
	 math CRC = 0x811c9dc5
	 strlen NAMESZ FULLNAME
	 for j = 0 < NAMESZ
		 getvarchr CHR FULLNAME j
		 if CHR >= 0x61
			 if CHR <= 0x7a
				 math CHR -= 0x20
			 endif
		 endif
		 math CRC ^= CHR
		 math CRC *= 0x199933
	 next j
	 for j = 0 < FILES
		 getarray TMP 0 j
		 if CRC == TMP
			 break
		 endif
	 next j
	 if j >= FILES
		 print "Alert: the crc of the file %FULLNAME% has not been found, it can't be extracted!"
		 #cleanexit
		 # all wrongs!
		 math j = i
		 #math TMP = NEXT
		 #math j = 0
		 #math j -= NEXT
		 #math j -= 1
	 endif
	 math TMP = j
	 math TMP *= 16
	 math TMP += INFO_OFF
	 goto TMP MEMORY_FILE
	 get OFFSET long MEMORY_FILE
	 math OFFSET <<= 8
	 get ZSIZE long MEMORY_FILE
	 get SIZE long MEMORY_FILE
	 get PACKED long MEMORY_FILE
	 goto OFFSET
	 getdstring SIGN 4
	 if SIGN == "LZ2K"
		 set PACKED long 2
	 else
		 set PACKED long 0
	 endif
	 if PACKED & 2
		 callfunction UNLZ2K
		 log FULLNAME 0 SIZE MEMORY_FILE2
	 else
		 if SIZE != 0
		 if ZSIZE != 0
		 if SIZE != ZSIZE
			 print "SIZE (%SIZE%) and ZSIZE (%ZSIZE%) differ, contact me"
			 cleanexit
		 endif
		 endif
		 endif
		 log FULLNAME OFFSET SIZE
	 endif
next i
startfunction UNLZ2K
putvarchr MEMORY_FILE2 SIZE 0
log MEMORY_FILE2 0 0
append
for TMPSZ = 0 < ZSIZE
	 goto OFFSET
	 getdstring SIGN 4
	 if SIGN != "LZ2K"
		 print "Error: the signature at offset %OFFSET% (%SIGN%) is not LZ2K, contact me"
		 cleanexit
	 endif
	 get LZ2K_SIZE long
	 get LZ2K_ZSIZE long
	 savepos OFFSET
	 if LZ2K_ZSIZE == LZ2K_SIZE
		 log MEMORY_FILE2 OFFSET LZ2K_SIZE
	 else
		 clog MEMORY_FILE2 OFFSET LZ2K_ZSIZE LZ2K_SIZE
	 endif
	 math OFFSET += LZ2K_ZSIZE
	 savepos TMP MEMORY_FILE2
	 math TMPSZ += 12
	 math TMPSZ += LZ2K_ZSIZE
next
append
endfunction

I have both taken all the files out to see if it runs, which indeed it does, however the graphical glitches seem to still be there..

[McJobless]

The v5 script is new to me. If you could contact the author and let him know he should be able to chase it up. For now, uninstall the game and delete all the extracted files, and try again. Also, watch out for any errors during extraction.

[annullus]

do you know if any progress has been made in making a recompiler which works for the Lego games ?

[McJobless]

None so far, but presumably one could just reverse the steps in that BMS Script and it would work...

Don't literally reverse it

[le717]

I haven't seen that QuickBMS script either, but if you are having issues with extraction, take origimaiguy's extractor out for a spin:

http://www.rockraide...ing/#entry73294

As for recreating the DATs, didn't you say that LIJ2 could run without them, Extreme? If al the games are kinda alike, HP should be able to do the same thing.

[McJobless]

True, that is correct, although, I was thinking in the case of WAD files, if someone builds an entire overhaul they might like to repack the game files.

[annullus]

I tried that extractor it doesnt bring any of the textfiles out properly, instead they all have bits of audio data stashed in them

[McJobless]

Because it was built before Harry Potter. Remember, they change the file formats with every game. Completely reinstall your game, make sure it runs, extract the files and delete the *.dat files, run again to check it's working.

[annullus]

As i installed it using steam i wiped the .dats and am going to to try extracting it now, will post back my results. Also i had that same problem when using the extractor on starwars saga

[McJobless]

Ah, make sure you right click the item in Steam, select "Properties" and look around. You need to disable Steam from Automatically Downloading Updates.

EDIT: Can you please tell me your computers specs?

[annullus]

Ah i merely have a Phenom II 1090T, with 16Gb ram, and an 7950. Also the steam updating is irrelevant to the extracted archives, as it patches into the archives, if they are not present it will then download the archived data again and seek to patch. As this is not happening all i can conclude is that it must be an error with the extracted file data. Btw you dont need to extract all the files to test, only the files containing common data as that also stores the Hub area

[McJobless]

No, that's not the problem. The problem is the same with KotOR II and that when you mod the game and change the files, Steam fails a checksum, and will download new files, which will likely corrupt them.

And yes, you need to extract EVERYTHING because some of the other important files are split right across the .dat files.

If I can run my modded copy of Star Wars Saga with no graphical errors, this is an issue with your setup. So, you're definately sure there is no graphic errors prior to extraction, and that the extracted files all have names and are in the correct folder layout?

[annullus]

an update, Extracting all the files it still bugs, however in a wierd issue if i start a level in freeplay the character who is bugged changes. My only hypothesis at the moment is that the graphical errors are caused by a lack of optimization and that the compressed archives are able to be read faster. That is the crux of the issue at the moment, the corruption seems to be limited to animations, the legs and jumping are animated properly on some minifigs. I was thinking that maybe the model animation file had been corrupted but that doesnt seem to be the case due to the findings of freeplay