This project contains multiple tools for reverse-engineering classic Mac OS applications and games.
The tools in this project are:
apt-get install zlibg1-dev
.)cmake .
, then make
.This project should build properly on sufficiently recent versions of macOS and Linux.
resource_dasm is a disassembler for classic Mac OS resource forks. It extracts resources from the resource fork of any file and converts many classic Mac OS resource formats (images, sounds, text, etc.) into modern formats.
Examples:
Export all resources from a specific file and convert them to modern formats (output is written to the \<filename>.out directory by default):
./resource_dasm files/Tesserae
Export all resources from all files in a folder, writing the output files into a parallel folder structure in the current directory:
./resource_dasm "files/Apeiron ƒ/" ./apeiron.out
Export a specific resource from a specific file, in both modern and original formats:
./resource_dasm "files/MacSki 1.7/MacSki Sounds" ./macski.out --target-type=snd --target-id=1023 --save-raw=yes
Export a PowerPC application's resources and disassemble its code:
./resource_dasm "files/Adventures of Billy" ./billy.out
./m68kdasm --pef "files/Adventures of Billy" ./billy.out/dasm.txt
Export all resources from a Mohawk archive:
./resource_dasm files/Riven/Data/a_Data.MHK ./riven_data_a.out --index-format=mohawk
Due to copying files across different types of filesystems, you might have a file's resource fork in the data fork of a separate file instead. To export resources from such a file:
./resource_dasm "windows/Realmz/Data Files/Portraits.rsf" ./portraits.out --data-fork
Create a new resource file, with a few TEXT and clut resources:
./resource_dasm --create --add-resource=TEXT:128@file128.txt --add-resource=TEXT:129@file129.txt --add-resource=clut:2000@clut.bin output.rsrc
Add a resource to an existing resource file:
./resource_dasm file.rsrc --add-resource=TEXT:128@file128.txt output.rsrc
Delete a resource from an existing resource file:
./resource_dasm file.rsrc --delete-resource=TEXT:128 output.rsrc
This isn't all resource_dasm can do. Run it without any arguments (or look at print_usage()
in src/resource_dasm.cc) for a full description of all the options.
resource_dasm can convert these resource types:
Type | Output format | Notes
------------------------------------------------------------------------
Text resources
bstr | .txt (one file per string) | *3
card | .txt |
finf | .txt (description of contents) |
FCMT | .txt | *3
FONT | .txt (description) and image (one image per glyph) | *E
lstr | .txt | *3
MACS | .txt | *3
minf | .txt | *3
mstr | .txt | *3
mst# | .txt (one file per string) | *3
NFNT | .txt (description) and image (one image per glyph) | *E
PSAP | .txt |
sfnt | .ttf (TrueType font) |
STR | .txt | *3
STR# | .txt (one file per string) | *3
styl | .rtf | *4
TEXT | .txt | *3
TwCS | .txt (one file per string) |
wstr | .txt |
------------------------------------------------------------------------
Image and color resources
actb | image (24-bit) | *E *8
acur | .txt (list of cursor frame IDs) |
cctb | image (24-bit) | *E *8
cicn | image (32-bit and monochrome) | *E
clut | image (24-bit) | *E *8
crsr | image (32-bit and monochrome) | *E *1
CTBL | image (24-bit) | *E
CURS | image (32-bit) | *E *1
dctb | image (24-bit) | *E *8
fctb | image (24-bit) | *E *8
icl4 | image (24 or 32-bit) and .icns | *E *0
icl8 | image (24 or 32-bit) and .icns | *E *0
icm# | image (32-bit) | *E
icm4 | image (24 or 32-bit) | *E *0
icm8 | image (24 or 32-bit) | *E *0
ICN# | image (32-bit) and .icns | *E
icns | image, .png, .jp2, .txt, .plist, .bin, etc. | *E *9
ICON | image (24-bit) | *E
ics# | image (32-bit) and .icns | *E
ics4 | image (24 or 32-bit) and .icns | *E *0
ics8 | image (24 or 32-bit) and .icns | *E *0
kcs# | image (32-bit) | *E
kcs4 | image (24 or 32-bit) | *E *0
kcs8 | image (24 or 32-bit) | *E *0
PAT | image (24-bit; pattern and 8x8 tiling) | *E
PAT# | image (24-bit; pattern and 8x8 tiling for each pattern) | *E
PICT | image (24-bit) or other format | *E *2
pltt | image (24-bit) | *E *8
ppat | image (24-bit; color, color 8x8, mono, mono 8x8) | *E
ppt# | image (24-bit; 4 images as above for each pattern) | *E
SICN | image (24-bit, one per icon) | *E
wctb | image (24-bit) | *E *8
------------------------------------------------------------------------
Sound and sequence resources
.mod | .mod (ProTracker module) |
ALIS | .txt (description of contents) |
cmid | .midi |
csnd | .wav or .mp3 | *5
ecmi | .midi |
emid | .midi |
esnd | .wav or .mp3 | *5
ESnd | .wav or .mp3 | *5
INST | .json | *6
MADH | .madh (PlayerPRO module) |
MADI | .madi (PlayerPRO module) |
MIDI | .midi |
Midi | .midi |
midi | .midi |
SMSD | .wav | *A
snd | .wav or .mp3 | *5
SONG | .json (smssynth) | *6
SOUN | .wav | *A
Tune | .midi | *7
Ysnd | .wav |
------------------------------------------------------------------------
Code resources
ADBS | .txt (68K assembly) | *C
adio | .txt (68K assembly) | *C
AINI | .txt (68K assembly) | *C
atlk | .txt (68K assembly) | *C
boot | .txt (68K assembly) | *C
CDEF | .txt (68K assembly) | *C
cdek | .txt (PPC32 assembly and header description) |
cdev | .txt (68K assembly) | *C
CDRV | .txt (68K assembly) | *C
cfrg | .txt (description of code fragments) | *D
citt | .txt (68K assembly) | *C
clok | .txt (68K assembly) | *C
cmtb | .txt (68K assembly) | *C
cmu! | .txt (68K assembly) | *C
CODE | .txt (68K assembly or import table description) | *B *C
code | .txt (68K assembly) | *C
dcmp | .txt (68K assembly) | *C
dcod | .txt (PPC32 assembly and header description) |
dem | .txt (68K assembly) | *C
dimg | .txt (68K assembly) | *C
drvr | .txt (68K assembly) | *C
DRVR | .txt (68K assembly) | *C
enet | .txt (68K assembly) | *C
epch | .txt (PPC32 assembly) |
expt | .txt (PPC32 assembly) |
FKEY | .txt (68K assembly) | *C
fovr | .txt (PPC32 assembly and header description) |
gcko | .txt (68K assembly) | *C
gdef | .txt (68K assembly) | *C
GDEF | .txt (68K assembly) | *C
gnld | .txt (68K assembly) | *C
INIT | .txt (68K assembly) | *C
krnl | .txt (PPC32 assembly) |
LDEF | .txt (68K assembly) | *C
lmgr | .txt (68K assembly) | *C
lodr | .txt (68K assembly) | *C
ltlk | .txt (68K assembly) | *C
MBDF | .txt (68K assembly) | *C
MDEF | .txt (68K assembly) | *C
mntr | .txt (68K assembly) | *C
ncmp | .txt (PPC32 assembly and header description) |
ndlc | .txt (PPC32 assembly and header description) |
ndmc | .txt (PPC32 assembly and header description) |
ndrv | .txt (PPC32 assembly and header description) |
nift | .txt (PPC32 assembly and header description) |
nitt | .txt (PPC32 assembly and header description) |
nlib | .txt (PPC32 assembly and header description) |
nsnd | .txt (PPC32 assembly and header description) |
nsrd | .txt (PPC32 assembly) |
ntrb | .txt (PPC32 assembly and header description) |
osl | .txt (68K assembly) | *C
otdr | .txt (68K assembly) | *C
otlm | .txt (68K assembly) | *C
PACK | .txt (68K assembly) | *C
pnll | .txt (68K assembly) | *C
ppct | .txt (PPC32 assembly and header description) |
proc | .txt (68K assembly) | *C
PTCH | .txt (68K assembly) | *C
ptch | .txt (68K assembly) | *C
pthg | .txt (68K or PPC32 assembly and header description) | *C
qtcm | .txt (PPC32 assembly and header description) |
ROvr | .txt (68K assembly) | *C
RSSC | .txt (68K assembly) | *C
scal | .txt (PPC32 assembly and header description) |
scod | .txt (68K assembly) | *C
SERD | .txt (68K assembly) | *C
sfvr | .txt (PPC32 assembly and header description) |
shal | .txt (68K assembly) | *C
sift | .txt (68K assembly) | *C
SMOD | .txt (68K assembly) | *C
snth | .txt (68K assembly) | *C
tdig | .txt (68K assembly) | *C
tokn | .txt (68K assembly) | *C
vdig | .txt (68K or PPC32 assembly and header description) | *C
wart | .txt (68K assembly) | *C
WDEF | .txt (68K assembly) | *C
XCMD | .txt (68K assembly) | *C
XFCN | .txt (68K assembly) | *C
------------------------------------------------------------------------
MacApp resources
68k! | .txt (description of memory config for 680x0) |
CMNU | .txt (description of menu) |
cmnu | .txt (description of menu) |
errs | .txt (description of error ranges) |
mem! | .txt (description of memory config) |
ppc! | .txt (description of memory config for PPC) |
res! | .txt (string list of always resident segments) |
seg! | .txt (string list of segments) |
TxSt | .txt (description of text style) |
------------------------------------------------------------------------
Miscellaneous resources
ALRT | .txt (alert parameters) |
APPL | .txt (description of contents) |
audt | .txt (description of contents) |
BNDL | .txt (description of contents) |
CMDK | .txt (list of keys) |
CNTL | .txt (description of control) |
CTY# | .txt (description of cities) |
dbex | .txt (description of contents) |
DITL | .txt (dialog parameters) |
DLOG | .txt (dialog parameters) |
FBTN | .txt (description of buttons) |
FDIR | .txt (description of contents) |
fld# | .txt (description of folders) |
flst | .txt (description of font family list) |
fmap | .txt (description of finder icon mappings) |
FREF | .txt (description of file references) |
FRSV | .txt (list of font IDs) |
FWID | .txt (font parameters) |
gbly | .txt (description of Gibbly aka. System Enabler) |
GNRL | .txt (description of contents) |
hwin | .txt (description of help window) |
icmt | .txt (icon reference and comment) |
inbb | .txt (description of contents) |
indm | .txt (description of contents) |
infs | .txt (description of contents) |
inpk | .txt (description of contents) |
inra | .txt (description of contents) |
insc | .txt (description of contents) |
itl0 | .txt (international formatting information) |
ITL1 | .txt (short dates flag value) |
itlb | .txt (internationalization parameters) |
itlc | .txt (internationalization parameters) |
itlk | .txt (keyboard mappings) |
KBDN | .txt (keyboard name) |
LAYO | .txt (description of layout) |
mach | .txt (description of contents) |
MBAR | .txt (list of menu IDs) |
mcky | .txt (threshold values) |
MENU | .txt (description of menu) |
mitq | .txt (description of queue sizes) |
nrct | .txt (rectangle boundaries) |
PAPA | .txt (printer parameters) |
PICK | .txt (picker parameters) |
ppcc | .txt (description of contents) |
ppci | .txt (description of contents) |
PRC0 | .txt (description of contents) |
PRC3 | .txt (description of contents) |
pslt | .txt (description of Nubus pseudo-slot lists) |
ptbl | .txt (description of patch table) |
qrsc | .txt (description of queries) |
RECT | .txt (description of the rectangle) |
resf | .txt (list of fonts) |
RMAP | .txt (type mapping and list of ID exceptions) |
ROv# | .txt (list of overridden resource IDs) |
rtt# | .txt (list of database result handlers) |
RVEW | .txt (description of contents) |
scrn | .txt (screen device parameters) |
sect | .txt (description of contents) |
SIGN | .txt (description of contents) |
SIZE | .txt (description of parameters) |
slut | .txt (description of mapping) |
thn# | .txt (description of 'thng' mapping) |
TMPL | .txt (description of format) |
TOOL | .txt (description of contents) |
vers | .txt (version flags and strings) |
WIND | .txt (window parameters) |
Notes:
*0: Produces a 32-bit image if a corresponding monochrome resource exists
(ICN# for icl4/8, icm# for icm4/8, ics# for ics4/8, kcs# for kcs4/8). If
no monochrome resource exists, produces a 24-bit image instead. All
color information in the original resource is reproduced in the output,
even for fully-transparent pixels. If the icon was originally intended
to be used with a nonstandard compositing mode, the colors of fully-
transparent pixels may have been relevant, but most modern image viewers
and editors don't have a way to display this information.
*1: The hotspot coordinates are appended to the output filename. As in *0,
resource_dasm faithfully reproduces the color values of transparent
pixels in the output file, but most modern image editors won't show
these "transparent" pixels.
*2: resource_dasm implements multiple PICT decoders. It will first attempt
to decode the PICT using its internal decoder, which usually produces
correct results but fails on PICTs that contain complex drawing opcodes.
This decoder can handle basic QuickTime images as well (e.g. embedded
JPEGs and PNGs), but can't do any drawing under or over them, or
matte/mask effects. PICTs that contain embedded JPEGs or PNGs will
result in a JPEG or PNG file rather than the format specified by
--image-format (which is BMP by default). If the internal decoder fails,
resource_dasm will fall back to a decoder that uses picttoppm, which is
part of NetPBM. There is a rare failure mode in which picttoppm hangs
forever; resource_dasm gives it 10 seconds to do its job before killing
it and giving up. If picttoppm is not installed, fails to decode the
PICT, or is killed due to a timeout, resource_dasm will prepend the
necessary header and save the data as a PICT file instead.
*3: Text is assumed to use the Mac OS Roman encoding. It is converted to
UTF-8, and line endings (\r) are converted to Unix style (\n).
*4: Some rare style options may not be translated correctly. styl resources
provide styling information for the TEXT resource with the same ID, so
such a resource must be present to properly decode a styl.
*5: RMF archives can contain snd resources that are actually in MP3 format;
in this case, the exported sound will be a .mp3 file. Otherwise, the
exported sound is an uncompressed WAV file, even if the resource's data
is compressed. resource_dasm can decompress IMA 4:1, MACE 3:1, MACE 6:1,
A-law, and mu-law (ulaw) compression.
*6: JSON files from SoundMusicSys SONG resources can be played with smssynth
(http://www.github.com/fuzziqersoftware/gctools). The JSON file refers
to the instrument sounds and MIDI sequence by filename and does not
include directory names, so if you want to play these, you'll have to
manually put the sounds and MIDI files in the same directory as the JSON
file if you're using --filename-format.
*7: Tune decoding is experimental and will likely produce unplayable MIDIs.
*8: For color table resources, the raw data is always saved even if it is
decoded properly, since the original data contains 16-bit values for
each channel and the output image file has less-precise 8-bit channels.
*9: icns resources are decoded into many different file types depending on
the contents of the resource. For subfields that have split alpha
channels (that is, the transparency data is in a different subfield),
resource_dasm produces an original image and one with transparency
applied. Some icns resources also contain metadata, which is exported as
.bin, .txt, and .plist files, except for the Icon Composer version used
to create the file, which is ignored. If you want the result in Icon
Composer format, use --save-raw=yes and resource_dasm will save it as a
.icns file.
*A: These resources appear to have a fixed format, with a constant sample
rate, sample width and channel count. You may have to adjust these
parameters in the output if it turns out that these are configurable.
*B: The disassembler attempts to find exported functions by parsing the jump
table in the CODE 0 resource, but if this resource is missing or not in
the expected format, it skips this step and does not fail. Generally, if
any "export_X:" labels appear in the disassembly, then export resolution
succeeded and all of the labels should be correct (otherwise they will
all be missing).
*C: Some coprocessor and floating-point opcodes (F-class) are not
implemented and will disassemble with the comment "// unimplemented".
*D: Most PowerPC applications have their executable code in the data fork.
To disassemble it, use m68kdasm (example above).
*E: The output image format can be specified using --image-format. The
default output format is bmp (Windows bitmap); other supported formats
are png and ppm.
If resource_dasm fails to convert a resource, or doesn't know how to, it will attempt to decode the resource using the corresponding TMPL (template) resource if it exists. If there's no appropriate TMPL, the TMPL is corrupt, or the TMPL can't decode the resource, resource_dasm will produce the resource's raw data instead.
Most of the decoder implementations in resource_dasm are based on reverse-engineering existing software and pawing through the dregs of old documentation, so some rarer types of resources probably won't work yet. However, I want this project to be as complete as possible, so if you have a resource that you think should be decodable but resource_dasm can't decode it, send it to me (perhaps by attaching to a GitHub issue) and I'll try my best to make resource_dasm understand it.
resource_dasm transparently decompresses resources that are marked by the resource manager as compressed.
The resource manager compression scheme was never officially documented by Apple or made public, so the implementation of these decompressors is based on reverse-engineering ResEdit and other classic Mac OS code. In summary, resources are decompressed by executing 68K or PowerPC code from a dcmp or ncmp resource, which is looked up at runtime in the chain of open resource files like most other resources are. (In practice, the relevant dcmp/ncmp is usually contained in either the same file as the compressed resource or in the System file.) There are two different formats of compressed resources and two corresponding formats of 68K decompressors; resource_dasm implements support for both formats.
resource_dasm contains native implementations of all four decompressors built into the Mac OS System file. Specifically:
resource_dasm has built-in 68K and PowerPC emulators to run non-default decompressors. These emulators can also run the default decompressors, which are included with resource_dasm. Current status of emulated decompressors:
There may be other decompressors out there that I haven't seen, which may not work. If you see "warning: failed to decompress resource" when using resource_dasm, please create a GitHub issue and upload the exported compressed resource (.bin file) that caused the failure, and all the dcmp and ncmp resources from the same source file.
Run sudo make install
to copy the header files and library to the relevant paths after building. After installation, you can #include <resource_file/IndexFormats/ResourceFork.hh>
(for example) and link with -lresource_file
. There is no documentation for this library beyond what's written in the header files.
The library contains the following useful functions and classes:
Using m68kdasm is fairly straightforward. Run m68kdasm --help
for a full list of options.
Currently m68kdasm can disassemble these types of data:
Some of these executable formats support CPU architectures that m68kdasm does not support; if it encounters one of these, it prints the code segments as data segments instead.
m68kdasm can also assemble PowerPC, x86, and SH-4 assembly into raw binary. (It does not support assembling M68K text into binary, but this will be implemented in the future.) The expected input syntax for each architecture matches the disassembly syntax; for PowerPC and SH-4, this is not the standard syntax used by most other tools.
m68kexec is a CPU emulator and debugger for the Motorola 68000, 32-bit PowerPC, and x86 architectures. I often use it to help understand what some archaic code is trying to do, or to compare the behavior of code that I've transcribed to a modern language with the original code's behavior. For use cases like this, you generally will want to set up one or more input regions containing the data you're testing with, and one or more output regions for the emulated code to write to.
Perhaps this is best explained by example. This command is used to execute the encryption context generation function from Phantasy Star Online Blue Burst, to compare it with the same function as implemented in newserv:
./m68kexec --x86 --trace \
--mem=A0000000/2AC43585C46A6366188889BCE3DB88B15C2B3C751DB6757147E7E9390598275CC79547B2E5C00DD145002816B59C067C \
--mem=A1000000:1048 \
--load-pe=files/windows/pso/psobb.exe \
--pc=00763FD0 \
--reg=ecx:A1000000 \
--push=00000030 \
--push=A0000000 \
--push=FFFFFFFF \
--breakpoint=FFFFFFFF
The --mem
options set up the input regions; the A0000000 region contains the encryption seed (0x30 bytes) and the A1000000 region will contain the generated encryption context when the function returns. The --load-pe
option loads the code to be executed and --pc
tells the emulator where to start. (By default, it will start at the entrypoint defined in the executable, if any is given; here, we want to call a specific function instead.) The --reg
option sets the this
pointer in the function to the space we allocated for it. The --push
options set the function's arguments and return address. It will return to FFFFFFFF, which has no allocated memory, but we've also set a --breakpoint
at that address which will stop emulation just before an exception is thrown.
Since we used --trace
, the emulator prints the registers' state after every opcode, so we can trace through its behavior and compare it with our external implementation of the same function. When the function returns and triggers the breakpoint, we can use r A1000000 1048
in the shell to see the data that it generated, and also compare that to our external function's result.
render_bits is useful to answer the question "might this random-looking binary data actually be an image or 2-D array?" Give it a color format and some binary data, and it will produce a full-color BMP file that you can look at with your favorite image viewer or editor. You can also give a color table (.bin file produced by resource_dasm from a clut resource) if you think the input is indexed color data. If the output looks like garbage, play around with the width and color format until you figure out the right parameters.
Run render_bits without any options for usage information.
render_text lets you see what actual text would look like when rendered with a bitmap font (FONT/NFNT resource). To use it, get a .bin file from a FONT or NFNT resource (e.g. with resource_dasm --save-raw). Then run render_text with no arguments to see how to use it.
Sometimes in the course of reverse-engineering you'll end up with an image that has the right content and structure, but the colors are completely wrong. Chances are it was rendered with the wrong color table; to fix this, you can use replace_clut to map all of the image's pixels from one colorspace to another.
Run replace_clut without any options for usage information.
Some games store large images split up into a set of smaller textures; assemble_images can programmatically combine them into a a single large image again. Run assemble_images without any options to see how to use it.
dupe_finder finds duplicate resources of the same type in one or several resource files.
Run dupe_finder without any options for usage information.
hypercard_dasm stack_file [output_dir]
, or just hypercard_dasm
to see all optionsicon_dearchiver archive_file [output_dir]
unpacks the icons to .icns files.vrfs_dump VRFS_file [output_dir]
decode_data can decode and decompress a few custom encoding formats used by various games. Specifically:
Game/App/Library | Encoding | CLI option | Notes
-------------------------------------------------------------
DinoPark Tycoon | LZSS | --dinopark | %0
DinoPark Tycoon | RLE | --dinopark |
Flashback | LZSS | --presage | %0
MacSki | COOK | --macski |
MacSki | CO2K | --macski |
MacSki | RUN4 | --macski |
PackBits (compress) | PackBits | --pack-bits |
PackBits (decompress) | PackBits | --unpack-bits |
Pathways Into Darkness | Pathways | --unpack-pathways |
SoundMusicSys | LZSS | --sms | %0
Notes:
%0: Although these are all variants of LZSS (and are indeed very similar to
each other), they are mutually incompatible formats.
decode_data can be used on its own to decompress data, or can be used as an external preprocessor via resource_dasm to transparently decompress some formats. For example, to use decode_data for MacSki resources, you can run a command like resource_dasm --external-preprocessor="./decode_data --macski" input_filename ...
render_sprite can render several custom game sprite formats. For some formats listed below, you'll have to provide a color table resource in addition to the sprite resource. A .bin file produced by resource_dasm from a clut, pltt, or CTBL resource will suffice; usually these can be found in the same file as the sprite resources or in the game application. Run render_sprite with no arguments for usage information.
Supported formats:
Game | Type | CLI option | Need color table | Notes
---------------------------------------------------------------------------
Beyond Dark Castle | PBLK | --PBLK | No |
Beyond Dark Castle | PPCT | --PPCT | No |
Beyond Dark Castle | PSCR | --PSCR-v2 | No |
Blobbo | BTMP | --BTMP | No |
Blobbo | PMP8 | --PMP8 | Yes | $9
BodyScope | Imag | --Imag | Yes | $2 $3
Bonkheads | Sprt | --Sprt | Yes |
Bubble Trouble | btSP | --btSP | Yes |
Dark Castle (color) | DC2 | --DC2 | No | $4
Dark Castle (monochrome) | PPCT | --PPCT | No |
Dark Castle (monochrome) | PSCR | --PSCR-v1 | No |
DinoPark Tycoon | BMap | --BMap | No |
DinoPark Tycoon | XBig | --XBig | No | $2
DinoPark Tycoon | XMap | --XMap | Yes | $2 $7
Dr. Quandary | Imag | --Imag | Sometimes | $1 $2 $3
Factory | 1img | --1img | No |
Factory | 4img | --4img | Yes |
Factory | 8img | --8img | Yes |
Flashback | PPSS | --PPSS | Yes | $2 $8
Fraction Munchers | Imag | --Imag-fm | Sometimes | $1 $2 $3
Greebles | GSIF | --GSIF | Yes |
Harry the Handsome Executive | HrSp | --HrSp | Yes | $9
Lemmings | SHPD | --SHPD-v1 | Sometimes | $0 $1 $2 $5
Marathon | .256 | --.256-m | No | $2
Mario Teaches Typing | Pak | --Pak | Sometimes | $1 $2
Mars Rising | btSP | --btSP | Yes |
Number Munchers | Imag | --Imag-fm | Sometimes | $1 $2 $3
Odell Down Under | Imag | --Imag | Sometimes | $1 $2 $3
Oh No! More Lemmings | SHPD | --SHPD-v2 | Sometimes | $0 $1 $2 $5
Pathways Into Darkness | .256 | --.256-pd | No | $2
Prince of Persia | SHPD | --SHPD-p | Sometimes | $0 $1 $2 $5
Prince of Persia 2 | SHAP | --SHAP | Yes |
SimCity 2000 | SPRT | --SPRT | Yes | $2
SimTower | | | No | $A
Slithereens | SprD | --SprD | Yes | $2
SnapDragon | Imag | --Imag | Sometimes | $1 $2 $3
Spectre | shap | --shap | No | $6
Step On It! | sssf | --sssf | Yes | $2
Super Munchers | Imag | --Imag-fm | Sometimes | $1 $2 $3
Swamp Gas | PPic | --PPic | Sometimes | $0 $2 $3
The Amazon Trail | Imag | --Imag | Sometimes | $2 $3
The Oregon Trail | Imag | --Imag | Sometimes | $1 $2 $3
TheZone | Spri | --Spri | Yes |
Word Munchers | Imag | --Imag-fm | Sometimes | $1 $2 $3
Notes:
$0: render_sprite can't tell from the contents of the resource whether it is
color or monochrome, so it assumes the resource is color if you give a
color table on the command line. If decoding fails with a color table,
try decoding without one (or vice versa).
$1: These games contain some color and some monochrome graphics. It should
be obvious which are which (usually color graphics are in a separate
file), but if not, you can give a clut anyway in these cases and
render_sprite will ignore it if the image is monochrome.
$2: These sprite formats contain multiple images, so render_sprite will
produce multiple image files.
$3: Resources of this type can contain embedded color tables; if you're
rendering a color image that doesn't have a color table, you'll have to
provide one via a command-line option. If the resource (or individual
images therein) contain their own color tables or are monochrome, no
color table is required on the command line, and any provided color
table via the command line will be ignored.
$4: You can get DC2 sprites from the DC Data file with
`resource_dasm --index-format=dc-data "DC Data"`.
$5: The graphics files contain resources that refer to segments of the data
fork in the same file. So, this option expects the original Graphics or
BW Graphics or Persia file (with both data and resource forks present),
not an already-extracted resource.
$6: shap resources contain 3D models and 2D top-down projections of them.
When given a shap resource, render_sprite produces an STL file and an
OBJ file for the 3D model, and an SVG file for the 2D top-down view.
$7: Some XMap resources are stored inside CBag archives. You can extract
them with `resource_dasm --index-format=cbag <CBAG_file.bin>`.
$8: This game has only one clut and it's huge - far longer than the usual
256 entries. It seems PPSS image sets are meant to be rendered with a
subset of this clut, but I haven't been able to figure out (yet) how the
game chooses what subset of it to use.
$9: The game doesn't contain any color tables. You can use a 256-color clut
resource from the Mac OS System file, or use the --default-clut option.
$A: The game stores its sprites in normal PICT resources with an incorrect
type. Use `resource_dasm --copy-handler=PICT:%89%E6%91%9C` to decode
them instead of using render_sprite.
icon_dearchiver unpacks the icons in an Icon Archiver (by Alessandro Levi Montalcini) archive to .icns. Run it with no options for usage information.
blobbo_render <Blev-file.bin> <PMP8-128.bmp>
ferazel_render
in the directory with the data files, or ferazel_render --help
to see all the options (there are many!)gamma_zee_render gamma_zee_application levels_filename
harry_render --clut-file=clut.bin
, or just harry_render
to see all the options (there are many!)infotron_render
in the Info Datafiles directorylemmings_render --clut-file=clut.bin
, or lemmings_render --help
to see all the optionsmshines_render world_file [output_directory]
lemmings_render
as for original Lemmings, but also use the --v2
optionrealmz_dasm global_data_dir [scenario_dir] out_dir
(if scenario_dir is not given, disassembles the shared data instead)