├── README.md
├── acb.py
├── assetbundle.py
├── faces.bash
├── genfacelists_remote.bash
├── iconsheet_new.py
├── kickoff
├── misc_utils
    ├── libahff
    │   ├── Makefile
    │   ├── ahff2png.c
    │   ├── lodepng.c
    │   ├── lodepng.h
    │   ├── pixel.c
    │   ├── rg_etc1.cpp
    │   ├── rg_etc1.h
    │   ├── rg_etc1wrap.cpp
    │   └── rg_etc1wrap.h
    ├── linux
    │   ├── libahff.dylib
    │   ├── make_icon_sheet
    │   ├── make_icon_sheet2
    │   └── scan_dec_hca
    └── macosx
    │   ├── libahff.dylib
    │   ├── make_icon_sheet
    │   ├── make_icon_sheet2
    │   └── scan_dec_hca
├── requirements.txt
├── rijndael.py
├── sbjk.py
├── simver
├── speculate.py
└── versioncheck.py


/README.md:
--------------------------------------------------------------------------------
  1 | # sbjk
  2 | 
  3 | The toolset used to maintain the image server for starlight.kirara.ca.
  4 | 
  5 | ### Getting started
  6 | 
  7 | Compile the LAME encoder and copy it into `misc_utils/<platform>`. The
  8 | program we want is the `lame` binary in `frontend`.
  9 | 
 10 | Then compile the HCA decoder (make sure to pre-bake the keys as
 11 | `scan_dec_hca` doesn't provide them on the CLI) and put `hca_decoder`
 12 | in `misc_utils/<platform>`.
 13 | 
 14 | Configure the environment as you did for sparklebox:
 15 | 
 16 |     export TRANSIENT_DIR_POINTER=TRANSIENT_DATA_DIR
 17 |     export TRANSIENT_DATA_DIR=_data/transient
 18 | 
 19 |     # the name of a folder in misc_utils. Platform binaries
 20 |     # needed by SBJK will be loaded from there.
 21 |     export PLATFORM=macosx
 22 | 
 23 |     # see sparklebox config
 24 |     export VC_ACCOUNT=...
 25 |     export VC_AES_KEY=...
 26 |     export VC_SID_SALT=...
 27 | 
 28 |     # truth version number if you don't have auto-update secrets
 29 |     export SIMULATED_VERSION=xxxxxxxx
 30 | 
 31 |     # rsync/scp format
 32 |     export SYNCREMOTE=cdn@ssh.static.com:/webroot/
 33 | 
 34 |     # generated with ssh-keygen; it's the big file, not the
 35 |     # small one that ends in .pub
 36 |     export SSH_PRIVATE_KEY=/path/to/some/ssh/key
 37 | 
 38 | Now you can set a cron job to run `kickoff` and we will automatically
 39 | update from the CDN.
 40 | 
 41 | ### Programs
 42 | 
 43 | `acb.py <path-to-acb-file> <path-to-out-dir>`
 44 | 
 45 | Extracts HCA tracks from ACB files with embedded data.
 46 | 
 47 | `assetbundle.py <path-to-unity3d-file> [output file]`
 48 | 
 49 | Extracts images from Unity assetbundles.
 50 | Note: Currently broken for CLI usage. We also use ctypes to call ahff2png
 51 | for image saving instead of the PIL, because it provides better compression.
 52 | 
 53 | `iconsheet_new.py <path-to-iconcache> <path-to-iconsheet-cache-folder> <path-to-output-folder>`
 54 | 
 55 | Incrementally combine the iconcache into sheets of 8x8 icons each. Uses a cache
 56 | to speed things up.
 57 | 
 58 | `kickoff`
 59 | 
 60 | Check the server version, and run SBJK, iconsheet, and rsync if it changed.
 61 | 
 62 | `sbjk.py <version>`
 63 | 
 64 | Downloads changed files from CDN and saves them to the `$WORKING_DIR`.
 65 | Requires the environment to be set.
 66 | 
 67 | `simver <version>`
 68 | 
 69 | If `$SIMULATED_VERSION` is set, print that. Otherwise, prints its first
 70 | argument. This lets you pin the server version to something when running
 71 | kickoff.
 72 | 
 73 | `speculate.py <hint>`
 74 | 
 75 | Check for new data that hasn't officially been released through the API
 76 | yet. Given a hint of 10014700, looks for 10014800, 10014710, and 10014750
 77 | in that order. Print the first one found.
 78 | 
 79 | `versioncheck.py <current-version>`
 80 | 
 81 | Check for new data through the official API. Requires `VC_*` be set.
 82 | Prints the current game version if it can, otherwise prints `current-version.`
 83 | 
 84 | ### Adding new things to SBJK
 85 | 
 86 | The first thing you need to do is write some code to run on your file(s).
 87 | You should have a callable objects that takes:
 88 | 
 89 | - the full URL to the resource
 90 | - the file name of the resource
 91 | - the resource flags (1 is the LZ4 compressed attribute).
 92 | 
 93 | This function downloads any assets given and saves it to the `/things` directory
 94 | on the image server. It makes use of the `get_resource`, which downloads the
 95 | asset and decompresses LZ4 if needed.
 96 | 
 97 | ```python
 98 |     def process_my_file(url, asset_name, flags):
 99 |         target_file = os.path.join(os.getenv("WORKING_DIR"), "root", "things", asset_name)
100 |         buf = get_resource(url, asset_name, flags)
101 | 
102 |         with open(target_file, "wb") as named_out_file:
103 |             named_out_file.write(buf)
104 | ```
105 | 
106 | To make SBJK use your function, add an entry to the ACTIONS list. Entries are a
107 | 2-tuple with a file pattern (similar to shell glob) and the callable.
108 | 
109 |     ("my_*.file", process_my_file)
110 | 
111 | ### Credits
112 | 
113 | - The guy who wrote the HCA decoder
114 | - @marcan for the [lightweight asset bundle decoder and open-source API client]
115 |   (https://github.com/marcan/deresuteme)
116 | 


--------------------------------------------------------------------------------
/acb.py:
--------------------------------------------------------------------------------
  1 | #!/usr/bin/env python3
  2 | # acb.py: For all your ACB extracting needs
  3 | 
  4 | # Copyright (c) 2016, The Holy Constituency of the Summer Triangle.
  5 | # Permission is hereby granted, free of charge, to any person obtaining a copy
  6 | # of this software and associated documentation files (the "Software"), to deal
  7 | # in the Software without restriction, including without limitation the rights
  8 | # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
  9 | # copies of the Software, and to permit persons to whom the Software is
 10 | # furnished to do so, subject to the following conditions:
 11 | #
 12 | # The above copyright notice and this permission notice shall be included in all
 13 | # copies or substantial portions of the Software.
 14 | #
 15 | # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 16 | # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 17 | # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 18 | # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 19 | # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 20 | # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 21 | # SOFTWARE.
 22 | 
 23 | # This program is based on code from VGMToolbox.
 24 | # Copyright (c) 2009
 25 | #
 26 | # Permission is hereby granted, free of charge, to any person
 27 | # obtaining a copy of this software and associated documentation
 28 | # files (the "Software"), to deal in the Software without
 29 | # restriction, including without limitation the rights to use,
 30 | # copy, modify, merge, publish, distribute, sublicense, and/or sell
 31 | # copies of the Software, and to permit persons to whom the
 32 | # Software is furnished to do so, subject to the following
 33 | # conditions:
 34 | #
 35 | # The above copyright notice and this permission notice shall be
 36 | # included in all copies or substantial portions of the Software.
 37 | #
 38 | # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 39 | # EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
 40 | # OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 41 | # NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
 42 | # HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
 43 | # WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
 44 | # FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
 45 | # OTHER DEALINGS IN THE SOFTWARE.
 46 | 
 47 | import struct
 48 | import functools
 49 | import itertools
 50 | import os
 51 | import io
 52 | import math
 53 | import sys
 54 | from collections import namedtuple as T
 55 | 
 56 | COLUMN_STORAGE_MASK = 0xF0
 57 | COLUMN_STORAGE_PERROW = 0x50
 58 | COLUMN_STORAGE_CONSTANT = 0x30
 59 | COLUMN_STORAGE_CONSTANT2 = 0x70
 60 | COLUMN_STORAGE_ZERO = 0x10
 61 | 
 62 | COLUMN_TYPE_MASK = 0x0F
 63 | COLUMN_TYPE_DATA   = 0x0B
 64 | COLUMN_TYPE_STRING = 0x0A
 65 | COLUMN_TYPE_FLOAT  = 0x08
 66 | COLUMN_TYPE_8BYTE  = 0x06
 67 | COLUMN_TYPE_4BYTE2 = 0x05
 68 | COLUMN_TYPE_4BYTE  = 0x04
 69 | COLUMN_TYPE_2BYTE2 = 0x03
 70 | COLUMN_TYPE_2BYTE  = 0x02
 71 | COLUMN_TYPE_1BYTE2 = 0x01
 72 | COLUMN_TYPE_1BYTE  = 0x00
 73 | 
 74 | WAVEFORM_ENCODE_TYPE_ADX          = 0
 75 | WAVEFORM_ENCODE_TYPE_HCA          = 2
 76 | WAVEFORM_ENCODE_TYPE_VAG          = 7
 77 | WAVEFORM_ENCODE_TYPE_ATRAC3       = 8
 78 | WAVEFORM_ENCODE_TYPE_BCWAV        = 9
 79 | WAVEFORM_ENCODE_TYPE_NINTENDO_DSP = 13
 80 | 
 81 | # string and data fields require more information
 82 | def promise_data(r):
 83 |     offset = r.uint32_t()
 84 |     size = r.uint32_t()
 85 |     return lambda h: r.bytes(size, at=h.data_offset + 8 + offset)
 86 | 
 87 | def promise_string(r):
 88 |     offset = r.uint32_t()
 89 |     return lambda h: r.string0(at=h.string_table_offset + 8 + offset)
 90 | 
 91 | column_data_dtable = {
 92 |     COLUMN_TYPE_DATA   : promise_data,
 93 |     COLUMN_TYPE_STRING : promise_string,
 94 |     COLUMN_TYPE_FLOAT  : lambda r: r.float32_t(),
 95 |     COLUMN_TYPE_8BYTE  : lambda r: r.uint64_t(),
 96 |     COLUMN_TYPE_4BYTE2 : lambda r: r.int32_t(),
 97 |     COLUMN_TYPE_4BYTE  : lambda r: r.uint32_t(),
 98 |     COLUMN_TYPE_2BYTE2 : lambda r: r.int16_t(),
 99 |     COLUMN_TYPE_2BYTE  : lambda r: r.uint16_t(),
100 |     COLUMN_TYPE_1BYTE2 : lambda r: r.int8_t(),
101 |     COLUMN_TYPE_1BYTE  : lambda r: r.uint8_t()}
102 | 
103 | column_data_stable = {
104 |     COLUMN_TYPE_DATA   : "8s",
105 |     COLUMN_TYPE_STRING : "4s",
106 |     COLUMN_TYPE_FLOAT  : "f",
107 |     COLUMN_TYPE_8BYTE  : "Q",
108 |     COLUMN_TYPE_4BYTE2 : "i",
109 |     COLUMN_TYPE_4BYTE  : "I",
110 |     COLUMN_TYPE_2BYTE2 : "h",
111 |     COLUMN_TYPE_2BYTE  : "H",
112 |     COLUMN_TYPE_1BYTE2 : "b",
113 |     COLUMN_TYPE_1BYTE  : "B"}
114 | 
115 | wave_type_ftable = {
116 |     WAVEFORM_ENCODE_TYPE_ADX          : ".adx",
117 |     WAVEFORM_ENCODE_TYPE_HCA          : ".hca",
118 |     WAVEFORM_ENCODE_TYPE_VAG          : ".at3",
119 |     WAVEFORM_ENCODE_TYPE_ATRAC3       : ".vag",
120 |     WAVEFORM_ENCODE_TYPE_BCWAV        : ".bcwav",
121 |     WAVEFORM_ENCODE_TYPE_NINTENDO_DSP : ".dsp"}
122 | 
123 | class R(object):
124 |     """ file reader based on types """
125 |     def __init__(self, file):
126 |         self.f = file
127 | 
128 |     def readfunc(fmt):
129 |         a = struct.Struct(fmt)
130 |         b = a.size
131 |         def f(f, at=None):
132 |             if at is not None:
133 |                 back = f.tell()
134 |                 f.seek(at)
135 |                 d = a.unpack(f.read(b))[0]
136 |                 f.seek(back)
137 |                 return d
138 |             else:
139 |                 return a.unpack(f.read(b))[0]
140 |         return f
141 | 
142 |     def latebinder(f):
143 |         return lambda s: f(s.f)
144 | 
145 |     int8_t    = latebinder(readfunc(">b"))
146 |     uint8_t   = latebinder(readfunc(">B"))
147 |     int16_t   = latebinder(readfunc(">h"))
148 |     uint16_t  = latebinder(readfunc(">H"))
149 |     int32_t   = latebinder(readfunc(">i"))
150 |     uint32_t  = latebinder(readfunc(">I"))
151 |     int64_t   = latebinder(readfunc(">q"))
152 |     uint64_t  = latebinder(readfunc(">Q"))
153 |     float32_t = latebinder(readfunc(">f"))
154 | 
155 |     le_int8_t    = latebinder(readfunc("<b"))
156 |     le_uint8_t   = latebinder(readfunc("<B"))
157 |     le_int16_t   = latebinder(readfunc("<h"))
158 |     le_uint16_t  = latebinder(readfunc("<H"))
159 |     le_int32_t   = latebinder(readfunc("<i"))
160 |     le_uint32_t  = latebinder(readfunc("<I"))
161 |     le_int64_t   = latebinder(readfunc("<q"))
162 |     le_uint64_t  = latebinder(readfunc("<Q"))
163 |     le_float32_t = latebinder(readfunc("<f"))
164 | 
165 | 
166 |     def seek(self, at, where=os.SEEK_SET):
167 |         self.f.seek(at, where)
168 | 
169 |     def struct(self, struct, at=None):
170 |         if at is not None:
171 |             back = self.f.tell()
172 |             self.f.seek(at)
173 |             d = self.struct(struct)
174 |             self.f.seek(back)
175 |             return d
176 | 
177 |         return struct.unpack(self.f.read(struct.size))
178 | 
179 |     def bytes(self, size, at=None):
180 |         if at is not None:
181 |             back = self.f.tell()
182 |             self.f.seek(at)
183 |             d = self.bytes(size)
184 |             self.f.seek(back)
185 |             return d
186 | 
187 |         return self.f.read(size)
188 | 
189 |     def string0(self, at=None):
190 |         if at is not None:
191 |             back = self.f.tell()
192 |             self.f.seek(at)
193 |             d = self.string0()
194 |             self.f.seek(back)
195 |             return d
196 | 
197 |         bk = self.f.tell()
198 |         tl = 0
199 |         sr = []
200 |         while 1:
201 |             b = self.f.read(16)
202 |             tl += len(b)
203 | 
204 |             if len(b) == 0:
205 |                 raise Exception("EOF")
206 | 
207 |             for c in b:
208 |                 if c != 0:
209 |                     sr.append(c)
210 |                 else:
211 |                     break
212 |             else:
213 |                 continue
214 |             break
215 |         string = bytes(sr)
216 |         self.f.seek(bk + len(string) + 1)
217 |         return string.decode("utf8")
218 | 
219 | class Struct(struct.Struct):
220 |     """ struct with an output filter (usually a namedtuple) """
221 |     def __init__(self, fmt, out_type):
222 |         super().__init__(fmt)
223 |         self.out_type = out_type
224 | 
225 |     def unpack(self, buf):
226 |         return self.out_type(* super().unpack(buf))
227 | 
228 | utf_header_t = Struct(">IHHIIIHHI",
229 |     T("utf_header_t", ("table_size", "u1", "row_offset", "string_table_offset",
230 |     "data_offset", "table_name_offset", "number_of_fields", "row_size", "number_of_rows")))
231 | 
232 | class UTFTable(object):
233 |     def __init__(self, file):
234 |         buf = R(file)
235 |         magic = buf.uint32_t()
236 |         if magic != 0x40555446:
237 |             raise ValueError("bad magic")
238 | 
239 |         self.header = buf.struct(utf_header_t)
240 |         self.name = buf.string0(at=self.header.string_table_offset + 8 + self.header.table_name_offset)
241 | 
242 |         buf.seek(0x20)
243 |         self.read_schema(buf)
244 | 
245 |         buf.seek(self.header.row_offset + 8)
246 |         self.rows = list(self.iter_rows(buf))
247 | 
248 |     def read_schema(self, buf):
249 |         buf.seek(0x20)
250 | 
251 |         dynamic_keys = []
252 |         format = ">"
253 |         constants = {}
254 | 
255 |         for _ in range(self.header.number_of_fields):
256 |             field_type = buf.uint8_t()
257 |             name_offset = buf.uint32_t()
258 | 
259 |             occurrence = field_type & COLUMN_STORAGE_MASK
260 |             type_key = field_type & COLUMN_TYPE_MASK
261 | 
262 |             if occurrence in (COLUMN_STORAGE_CONSTANT, COLUMN_STORAGE_CONSTANT2):
263 |                 name = buf.string0(at=self.header.string_table_offset + 8 + name_offset)
264 |                 val = column_data_dtable[type_key](buf)
265 |                 constants[name] = val
266 |             else:
267 |                 dynamic_keys.append(buf.string0(at=self.header.string_table_offset + 8 + name_offset))
268 |                 format += column_data_stable[type_key]
269 | 
270 |         for k in constants.keys():
271 |             if callable(constants[k]):
272 |                 constants[k] = constants[k](self.header)
273 | 
274 |         self.dynamic_keys = dynamic_keys
275 |         self.struct_format = format
276 |         self.constants = constants
277 | 
278 |     def resolve(self, buf, *args):
279 |         ret = []
280 |         for val in args:
281 |             if isinstance(val, bytes):
282 |                 if len(val) == 8:
283 |                     offset, size = struct.unpack(">II", val)
284 |                     ret.append(buf.bytes(size, at=self.header.data_offset + 8 + offset))
285 |                 else:
286 |                     offset = struct.unpack(">I", val)[0]
287 |                     ret.append(buf.string0(at=self.header.string_table_offset + 8 + offset))
288 |             else:
289 |                 ret.append(val)
290 |         return tuple(ret)
291 | 
292 |     def iter_rows(self, buf):
293 |         sfmt = Struct(self.struct_format, functools.partial(self.resolve, buf))
294 |         for n in range(self.header.number_of_rows):
295 |             values = buf.struct(sfmt)
296 |             ret = {k: v for k, v in zip(self.dynamic_keys, values)}
297 |             ret.update(self.constants)
298 |             yield ret
299 | 
300 | track_t = T("track_t", ("cue_id", "name", "wav_id", "enc_type", "is_stream"))
301 | 
302 | class TrackList(object):
303 |     def __init__(self, utf):
304 |         cue_handle = io.BytesIO(utf.rows[0]["CueTable"])
305 |         nam_handle = io.BytesIO(utf.rows[0]["CueNameTable"])
306 |         wav_handle = io.BytesIO(utf.rows[0]["WaveformTable"])
307 |         syn_handle = io.BytesIO(utf.rows[0]["SynthTable"])
308 | 
309 |         cues = UTFTable(cue_handle)
310 |         nams = UTFTable(nam_handle)
311 |         wavs = UTFTable(wav_handle)
312 |         syns = UTFTable(syn_handle)
313 | 
314 |         self.tracks = []
315 | 
316 |         name_map = {}
317 |         for row in nams.rows:
318 |             name_map[row["CueIndex"]] = row["CueName"]
319 | 
320 |         for ind, row in enumerate(cues.rows):
321 |             if row["ReferenceType"] not in {3, 8}:
322 |                 raise RuntimeError("ReferenceType {0} not implemented.".format(row["ReferenceType"]))
323 | 
324 |             r_data = syns.rows[row["ReferenceIndex"]]["ReferenceItems"]
325 |             a, b = struct.unpack(">HH", r_data)
326 | 
327 |             wav_id = wavs.rows[b]["Id"]
328 |             enc = wavs.rows[b]["EncodeType"]
329 |             is_stream = wavs.rows[b]["Streaming"]
330 | 
331 |             self.tracks.append(track_t(row["CueId"], name_map.get(ind, "UNKNOWN"), wav_id, enc, is_stream))
332 | 
333 | def align(n):
334 |     def _align(number):
335 |         return math.ceil(number / n) * n
336 |     return _align
337 | 
338 | afs2_file_ent_t = T("afs2_file_ent_t", ("cue_id", "offset", "size"))
339 | 
340 | class AFSArchive(object):
341 |     def __init__(self, file):
342 |         buf = R(file)
343 | 
344 |         magic = buf.uint32_t()
345 |         if magic != 0x41465332:
346 |             raise ValueError("bad magic")
347 | 
348 |         version = buf.bytes(4)
349 |         file_count = buf.le_uint32_t()
350 |         self.alignment = buf.le_uint32_t()
351 |         print("afs2:", file_count, "files in ar")
352 |         print("afs2: aligned to", self.alignment, "bytes")
353 | 
354 |         self.offset_size = version[1]
355 |         self.offset_mask = int("FF" * self.offset_size, 16)
356 |         print("afs2: a file offset is", self.offset_size, "bytes")
357 | 
358 |         self.files = []
359 |         self.create_file_entries(buf, file_count)
360 |         self.src = buf
361 | 
362 |     def create_file_entries(self, buf, file_count):
363 |         buf.seek(0x10)
364 |         read_cue_ids = struct.Struct("<" + ("H" * file_count))
365 |         read_raw_offs = struct.Struct("<" + ("I" * (file_count + 1)))
366 | 
367 |         # read all in one go
368 |         cue_ids = buf.struct(read_cue_ids)
369 |         raw_offs = buf.struct(read_raw_offs)
370 |         # apply the mask
371 |         unaligned_offs = tuple(map(lambda x: x & self.offset_mask, raw_offs))
372 |         aligned_offs = tuple(map(align(self.alignment), unaligned_offs))
373 |         offsets_for_length_calculating = unaligned_offs[1:]
374 |         lengths = itertools.starmap(
375 |             lambda my_offset, next_offset: next_offset - my_offset,
376 |             zip(aligned_offs, offsets_for_length_calculating))
377 | 
378 |         self.files = list(itertools.starmap(afs2_file_ent_t, zip(cue_ids, aligned_offs, lengths)))
379 | 
380 |     def file_data_for_cue_id(self, cue_id):
381 |         for f in self.files:
382 |             if f.cue_id == cue_id:
383 |                 return self.file_data(f)
384 |         else:
385 |             raise ValueError("id {0} not found in archive".format(cue_id))
386 | 
387 |     def file_data(self, ent):
388 |         return self.src.bytes(ent.size, at=ent.offset)
389 | 
390 | def extract_acb(acb_file, target_dir):
391 |     utf = UTFTable(acb_file)
392 |     cue = TrackList(utf)
393 |     embedded_awb = io.BytesIO(utf.rows[0]["AwbFile"])
394 |     data_source = AFSArchive(embedded_awb)
395 | 
396 |     for track in cue.tracks:
397 |         print(track)
398 |         name = "{0}{1}".format(track.name, wave_type_ftable.get(track.enc_type, track.enc_type))
399 |         with open(os.path.join(target_dir, name), "wb") as named_out_file:
400 |             named_out_file.write(data_source.file_data_for_cue_id(track.wav_id))
401 | 
402 | def main(invocation, acb_file, target_dir, *_):
403 |     with open(acb_file, "rb") as acb:
404 |         extract_acb(acb, target_dir)
405 | 
406 | if __name__ == '__main__':
407 |     main(*sys.argv)
408 | 


--------------------------------------------------------------------------------
/assetbundle.py:
--------------------------------------------------------------------------------
  1 | #!/usr/bin/python
  2 | # -!- coding: utf-8 -!-
  3 | #
  4 | # Copyright 2016 Hector Martin <marcan@marcan.st>
  5 | #
  6 | # Licensed under the Apache License, Version 2.0 (the "License");
  7 | # you may not use this file except in compliance with the License.
  8 | # You may obtain a copy of the License at
  9 | #
 10 | #     http://www.apache.org/licenses/LICENSE-2.0
 11 | #
 12 | # Unless required by applicable law or agreed to in writing, software
 13 | # distributed under the License is distributed on an "AS IS" BASIS,
 14 | # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 15 | # See the License for the specific language governing permissions and
 16 | # limitations under the License.
 17 | 
 18 | import struct, sys
 19 | import ctypes
 20 | 
 21 | xrange = range
 22 | 
 23 | baseStrings = {
 24 |     0:"AABB",
 25 |     5:"AnimationClip",
 26 |     19:"AnimationCurve",
 27 |     49:"Array",
 28 |     55:"Base",
 29 |     60:"BitField",
 30 |     76:"bool",
 31 |     81:"char",
 32 |     86:"ColorRGBA",
 33 |     106:"data",
 34 |     138:"FastPropertyName",
 35 |     155:"first",
 36 |     161:"float",
 37 |     167:"Font",
 38 |     172:"GameObject",
 39 |     183:"Generic Mono",
 40 |     208:"GUID",
 41 |     222:"int",
 42 |     241:"map",
 43 |     245:"Matrix4x4f",
 44 |     262:"NavMeshSettings",
 45 |     263:"MonoBehaviour",
 46 |     277:"MonoScript",
 47 |     299:"m_Curve",
 48 |     349:"m_Enabled",
 49 |     374:"m_GameObject",
 50 |     427:"m_Name",
 51 |     490:"m_Script",
 52 |     519:"m_Type",
 53 |     526:"m_Version",
 54 |     543:"pair",
 55 |     548:"PPtr<Component>",
 56 |     564:"PPtr<GameObject>",
 57 |     581:"PPtr<Material>",
 58 |     616:"PPtr<MonoScript>",
 59 |     633:"PPtr<Object>",
 60 |     688:"PPtr<Texture>",
 61 |     702:"PPtr<Texture2D>",
 62 |     718:"PPtr<Transform>",
 63 |     741:"Quaternionf",
 64 |     753:"Rectf",
 65 |     778:"second",
 66 |     795:"size",
 67 |     800:"SInt16",
 68 |     814:"int64",
 69 |     840:"string",
 70 |     847:"TextAsset",
 71 |     874:"Texture2D",
 72 |     884:"Transform",
 73 |     894:"TypelessData",
 74 |     907:"UInt16",
 75 |     928:"UInt8",
 76 |     934:"unsigned int",
 77 |     981:"vector",
 78 |     988:"Vector2f",
 79 |     997:"Vector3f",
 80 |     1006:"Vector4f",
 81 | }
 82 | 
 83 | class Stream(object):
 84 |     def __init__(self, d, p=0):
 85 |         self.d = d
 86 |         self.p = p
 87 |     def tell(self):
 88 |         return self.p
 89 |     def seek(self, p):
 90 |         self.p = p
 91 |     def skip(self, off):
 92 |         self.p += off
 93 |     def read(self, cnt):
 94 |         self.skip(cnt)
 95 |         return self.d[self.p-cnt:self.p]
 96 |     def align(self, n):
 97 |         self.p = (self.p + n - 1) & ~(n - 1)
 98 |     def read_str(self):
 99 |         s = self.d[self.p:].split(b"\0")[0]
100 |         self.skip(len(s)+1)
101 |         return s
102 | 
103 | 
104 | class Def(object):
105 |     TYPEMAP = {
106 |         "int": "<i",
107 |         "int64": "<q",
108 |         "char": "<1s",
109 |         "bool": "<B",
110 |         "float": "<f"
111 |     }
112 |     def __init__(self, name, type_name, size, flags, array=False):
113 |         self.children = []
114 |         self.name = name
115 |         self.type_name = type_name
116 |         self.size = size
117 |         self.flags = flags
118 |         self.array = array
119 | 
120 |     def read(self, s):
121 |         if self.array:
122 |             #print "a", self.name
123 |             size = self.children[0].read(s)
124 |             assert size < 10000000
125 |             if self.children[1].type_name in ("UInt8","char"):
126 |                 #print "s", size
127 |                 return s.read(size)
128 |             else:
129 |                 return [self.children[1].read(s) for i in xrange(size)]
130 |         elif self.children:
131 |             #print "o", self.name
132 |             v = {}
133 |             for i in self.children:
134 |                 v[i.name] = i.read(s)
135 |             if len(v) == 1 and self.type_name == "string":
136 |                 return v["Array"]
137 |             return v
138 |         else:
139 |             x = s.tell()
140 |             s.align(min(self.size,4))
141 |             d = s.read(self.size)
142 |             if self.type_name in self.TYPEMAP:
143 |                 d = struct.unpack(self.TYPEMAP[self.type_name], d)[0]
144 |             #print hex(x), self.name, self.type_name, repr(d)
145 |             return d
146 | 
147 |     def __getitem__(self, i):
148 |         return self.children[i]
149 | 
150 |     def append(self, d):
151 |         self.children.append(d)
152 | 
153 | class Asset(object):
154 |     def __init__(self, fd):
155 |         self.s = Stream(fd.read())
156 | 
157 |         self.s.seek(0x70)
158 |         self.off = self.s.tell()
159 | 
160 |         self.table_size, self.data_end, self.file_gen, self.data_offset = struct.unpack(">IIII", self.s.read(16))
161 |         self.s.read(4)
162 |         self.version = self.s.read_str()
163 |         self.platform = struct.unpack("<I", self.s.read(4))
164 |         self.defs = self.decode_defs()
165 |         self.objs = self.decode_data()
166 | 
167 | 
168 |     def decode_defs(self):
169 |         are_defs, count = struct.unpack("<BI", self.s.read(5))
170 |         return dict(self.decode_attrtab() for i in xrange(count))
171 | 
172 |     def decode_data(self):
173 |         count = struct.unpack("<I", self.s.read(4))[0]
174 |         objs = []
175 |         assert count < 1024
176 |         for i in xrange(count):
177 |             self.s.align(4)
178 |             pathId, off, size, t1, t2, unk = struct.unpack("<QIIIH2xB", self.s.read(25))
179 |             save = self.s.tell()
180 | 
181 |             self.s.seek(off + self.data_offset + self.off)
182 |             objs.append(self.defs[t1].read(self.s))
183 | 
184 |             self.s.seek(save)
185 |         return objs
186 | 
187 |     def decode_attrtab(self):
188 |         code, ident, attr_cnt, stab_len = struct.unpack("<I16sII", self.s.read(28))
189 |         #print "%08x %s" % (code, ident.encode("hex"))
190 |         attrs = self.s.read(attr_cnt*24)
191 |         stab = self.s.read(stab_len)
192 | 
193 |         defs = []
194 |         assert attr_cnt < 1024
195 |         for i in xrange(attr_cnt):
196 |             a1, a2, level, a4, type_off, name_off, size, idx, flags = struct.unpack("<BBBBIIIII", attrs[i*24:i*24+24])
197 |             if name_off & 0x80000000:
198 |                 name = baseStrings[name_off & 0x7fffffff]
199 |             else:
200 |                 name = stab[name_off:].split(b"\0")[0].decode("utf8")
201 |             if type_off & 0x80000000:
202 |                 type_name = baseStrings[type_off & 0x7fffffff]
203 |             else:
204 |                 type_name = stab[type_off:].split(b"\0")[0].decode("utf8")
205 |             d = defs
206 |             assert level < 16
207 |             for i in range(level):
208 |                 d = d[-1]
209 |             if size == 0xffffffff:
210 |                 size = None
211 |             d.append(Def(name, type_name, size, flags, array=bool(a4)))
212 |             #print "%2x %2x %2x %20s %8x %8x %2d: %s%s" % (a1, a2, a4, type_name, size or -1, flags, idx, "  " * level, name)
213 | 
214 |         assert len(defs) == 1
215 |         return code, defs[0]
216 | 
217 | def load_image(fd):
218 |     d = Asset(fd)
219 |     texs = [i for i in d.objs if "image data" in i]
220 |     # assert len(tex) == 1
221 |     for tex in texs:
222 |         def _fulfill_promise():
223 |             data = tex["image data"]
224 |             width, height, fmt = tex["m_Width"], tex["m_Height"], tex["m_TextureFormat"]
225 |             if fmt == 7: # BGR565
226 |                 im = Image.frombytes("RGB", (width, height), data, "raw", "BGR;16")
227 |             elif fmt == 3:
228 |                 im = Image.frombytes("RGB", (width, height), data, "raw", "RGB")
229 |             elif fmt == 4:
230 |                 im = Image.frombytes("RGBA", (width, height), data, "raw", "RGBA")
231 |             elif fmt == 13: # ABGR4444
232 |                 im = Image.frombytes("RGBA", (width, height), data, "raw", "RGBA;4B")
233 |                 r, g, b, a  = im.split()
234 |                 im = Image.merge("RGBA", (a, b, g, r))
235 |             else:
236 |                 raise Exception("Unsupported format %d" % fmt)
237 |             im = im.transpose(Image.FLIP_TOP_BOTTOM)
238 |             return im
239 |         yield (tex["m_Name"], _fulfill_promise)
240 | 
241 | if __name__ == "__main__":
242 |     im = load_image(open(sys.argv[1]))
243 |     if len(sys.argv) > 2:
244 |         im.save(sys.argv[2])
245 |     else:
246 |         im.show()
247 | 


--------------------------------------------------------------------------------
/faces.bash:
--------------------------------------------------------------------------------
 1 | #!/bin/bash
 2 | shopt -s extglob
 3 | shopt -s nullglob
 4 | 
 5 | for CHARA_ID in "$1"/*; do
 6 |     for FILE in "$CHARA_ID/"+([^_]).png; do
 7 |         if [[ "x$(echo ${FILE%.png}_*.png)" == "x" ]] ; then
 8 |             mogrify -trim $FILE
 9 |         else
10 |             IFS=" " read -r -a ORIGINAL_DIMENSIONS <<< $(identify -format "%[fx:w] %[fx:h]" $FILE)
11 |             convert $FILE -gravity South-East -background white -splice 1x0 -background black -splice 1x0 -trim -chop 1x0 +repage -background white -splice 0x1 -background black -splice 0x1 -trim -chop 0x1 +repage $(dirname $FILE)/__tmp_$(basename $FILE)
12 |             IFS=" " read -r -a TRIMMED_DIMENSIONS <<< $(identify -format "%[fx:w] %[fx:h]" $(dirname $FILE)/__tmp_$(basename $FILE))
13 | 
14 |             BN=$(basename $FILE)
15 |             printf "%d, -%d, -%d, %d, %d" \
16 |                 ${BN%.png} \
17 |                 $(expr ${ORIGINAL_DIMENSIONS[0]} - ${TRIMMED_DIMENSIONS[0]}) \
18 |                 $(expr ${ORIGINAL_DIMENSIONS[1]} - ${TRIMMED_DIMENSIONS[1]}) \
19 |                 ${ORIGINAL_DIMENSIONS[0]} \
20 |                 ${ORIGINAL_DIMENSIONS[1]} \
21 |                 > ${FILE%.png}.adjustment
22 | 
23 |             convert $(dirname $FILE)/__tmp_$(basename $FILE) -trim +repage $FILE
24 |             rm $(dirname $FILE)/__tmp_$(basename $FILE)
25 |         fi
26 |     done
27 | done
28 | 


--------------------------------------------------------------------------------
/genfacelists_remote.bash:
--------------------------------------------------------------------------------
 1 | #!/bin/bash
 2 | shopt -s extglob
 3 | shopt -s nullglob
 4 | 
 5 | # This file must only use bash builtins.
 6 | # Nothing is available in the remote chroot.
 7 | 
 8 | for CHARA_ID in "$1"/*; do
 9 |     for POSE in "$CHARA_ID/"+([^_]).png; do
10 |         read ADJUST_ARGLIST < ${POSE%.png}.adjustment
11 |         echo "SVX_APPLY_ADJUSTMENT($ADJUST_ARGLIST);" > ${POSE%.png}.json
12 | 
13 |         FACE_LIST="["
14 |         for FILE in "${POSE%.png}"_*.png; do
15 |             X="${POSE%.png}_"
16 |             X=${FILE#$X}
17 |             FACE_LIST="$FACE_LIST${X%.png},"
18 |         done
19 |         X=${POSE%.png}
20 |         echo "SVX_APPLY_FACE_LIST(${X#$CHARA_ID/}, $FACE_LIST]);" >> ${POSE%.png}.json
21 |     done
22 | done


--------------------------------------------------------------------------------
/iconsheet_new.py:
--------------------------------------------------------------------------------
 1 | #!/usr/bin/env python3
 2 | import sys
 3 | import os
 4 | import itertools
 5 | import subprocess
 6 | import json
 7 | import binascii
 8 | import re
 9 | 
10 | libexec = os.path.join(os.path.dirname(os.path.realpath(__file__)), "misc_utils", os.getenv("PLATFORM"))
11 | 
12 | w = 48
13 | h = 48
14 | 
15 | matrix_w = 8
16 | matrix_h = 8
17 | 
18 | def coords_for_position(ind):
19 |     return (ind % matrix_w) * w, (ind // matrix_h) * h
20 | 
21 | def composite_images(images, cache, out):
22 |     try:
23 |         with open(cache, "r") as manifest:
24 |             l = json.load(manifest)
25 |     except (IOError, ValueError):
26 |         l = None
27 | 
28 |     if l == images:
29 |         print("looks fine to me:", out)
30 |         return
31 | 
32 |     args = [os.path.join(libexec, "make_icon_sheet"), out]
33 |     args.extend(images)
34 |     subprocess.call(args, stdout=subprocess.DEVNULL)
35 | 
36 |     args[0] = os.path.join(libexec, "make_icon_sheet2")
37 |     args[1] = args[1].replace(".png", "@2x.jpg")
38 |     subprocess.call(args, stdout=subprocess.DEVNULL)
39 | 
40 |     with open(cache, "w") as manifest:
41 |         json.dump([i for i in images], manifest)
42 | 
43 | def do(this_sect, composite_args, cache, odir, image_count, css, css2):
44 |     composite_images(composite_args,
45 |         os.path.join(cache, "icons_{0}.png.cache".format(image_count)),
46 |         os.path.join(odir, "icons_{0}.png".format(image_count)))
47 |     random = binascii.hexlify(os.urandom(16)).decode("utf8")
48 |     for i, id in enumerate(this_sect):
49 |         px, py = coords_for_position(i)
50 |         css.write(".icon.icon_{0} {{ background-image:url(\"icons_{3}.png?{4}\"); background-position:-{1}px -{2}px; }}\n".format(
51 |             id, px, py, image_count, random))
52 |         css2.write(".icon.icon_{0} {{ background:url(\"icons_{3}@2x.jpg?{4}\") -{1}px -{2}px/384px 384px; }}\n".format(
53 |             id, px, py, image_count, random))
54 |     del this_sect[:]
55 |     del composite_args[:]
56 |     return image_count + 1
57 | 
58 | def do_if_full(this_sect, composite_args, cache, odir, image_count, css, css2):
59 |     if len(this_sect) == matrix_h * matrix_w:
60 |         return do(this_sect, composite_args, cache, odir, image_count, css, css2)
61 |     return image_count
62 | 
63 | HIDPI_CSS_HEADER = """\
64 | @media only screen and (-webkit-min-device-pixel-ratio: 1.3), 
65 |     only screen and (-o-min-device-pixel-ratio: 13/10), 
66 |     only screen and (min-resolution: 120dpi) {"""
67 | def gen_icon_sheets(root, cache, odir, css, css2):
68 |     css2.write(HIDPI_CSS_HEADER + "\n")
69 |     this_sect = []
70 |     composite_args = []
71 |     image_count = 0
72 | 
73 |     gex = re.compile("^([0-9]+).png$")
74 |     for image in filter(lambda y: gex.match(y), os.listdir(root)):
75 |         this_sect.append(gex.match(image).group(1))
76 |         composite_args.append(os.path.join(root, image))
77 |         image_count = do_if_full(this_sect, composite_args, cache, odir, image_count, css, css2)
78 | 
79 |     do(this_sect, composite_args, cache, odir, image_count, css, css2)
80 |     css2.write("}")
81 | 
82 | def main(root, cache, odir):
83 |     with open(os.path.join(odir, "icons.css"), "w") as css, open(os.path.join(odir, "icons@2x.css"), "w") as css2:
84 |         gen_icon_sheets(root, cache, odir, css, css2)
85 | 
86 | if __name__ == "__main__":
87 |     import plac
88 |     plac.call(main)
89 | 


--------------------------------------------------------------------------------
/kickoff:
--------------------------------------------------------------------------------
 1 | #!/bin/bash
 2 | cd "$(echo $0 | sed -E s/[^/]+$//g)"
 3 | 
 4 | eval WORKING_DIR=\$$TRANSIENT_DIR_POINTER
 5 | export WORKING_DIR
 6 | 
 7 | if [ ! -f "$WORKING_DIR/last_run" ]; then
 8 |     echo "VERSION=0" > "$WORKING_DIR/last_run"
 9 | fi
10 | source "$WORKING_DIR/last_run"
11 | 
12 | function kc_sync_now() {
13 |     echo "[-] Notice: syncing now."
14 |     rsync -e "ssh -oStrictHostKeyChecking=no -i $SSH_PRIVATE_KEY" -arvzL "$WORKING_DIR/root/" $SYNCREMOTE
15 | }
16 | 
17 | function kc_postupdate_tasks_fast() {
18 |     echo "[-] Notice: updating the iconsheets."
19 |     for d in "$WORKING_DIR/iconcache" "$WORKING_DIR/iconsheetcache" "$WORKING_DIR/root/icons2"; do
20 |         mkdir -p $d
21 |     done
22 | 
23 |     ./iconsheet_new.py "$WORKING_DIR/iconcache" "$WORKING_DIR/iconsheetcache" "$WORKING_DIR/root/icons2"
24 |     kc_sync_now
25 | }
26 | 
27 | function kc_postupdate_tasks_ex() {
28 |     echo "[-] Notice: cropping poses."
29 |     ./faces.bash "$WORKING_DIR/root/chara2"
30 |     cp genfacelists_remote.bash "$WORKING_DIR/root"
31 | 
32 |     kc_sync_now
33 |     read -r -a REMOTE_PARAM <<< $(tr : " " <<< $SYNCREMOTE)
34 |     echo "[-] Notice: executing genfacelists on remote."
35 |     ssh -oStrictHostKeyChecking=no -i $SSH_PRIVATE_KEY ${REMOTE_PARAM[0]} "${REMOTE_PARAM[1]}/genfacelists_remote.bash ${REMOTE_PARAM[1]}/chara2"
36 | }
37 | 
38 | function kc_restore_from_checkpoint() {
39 |     cp $1 "$WORKING_DIR/__manifestloader_cache/A_SyncCache.db"
40 | }
41 | 
42 | function kc_update() {
43 |     local UPGRADE_VERSION=$1; shift
44 |     echo "[-] kc_update $UPGRADE_VERSION"
45 | 
46 |     CHECKPOINT_PATH="$WORKING_DIR/__manifestloader_cache/A_SyncCache_checkpoint.db"
47 |     cp "$WORKING_DIR/__manifestloader_cache/A_SyncCache.db" $CHECKPOINT_PATH
48 | 
49 |     mkdir -p "$WORKING_DIR/root/"
50 |     ln -s "$WORKING_DIR/iconcache" "$WORKING_DIR/root/icon_card"
51 | 
52 |     SBJK_EXEC_MODE="fast" ./sbjk.py $UPGRADE_VERSION
53 |     if [[ $? != 0 ]]; then
54 |         kc_restore_from_checkpoint $CHECKPOINT_PATH
55 |         echo "[!] SBJK crashed, will not update lastrun."
56 |         rm -rf "$WORKING_DIR/root"
57 |         return
58 |     fi
59 |     kc_postupdate_tasks_fast $UPGRADE_VERSION
60 | 
61 |     SBJK_EXEC_MODE="slow" ./sbjk.py $UPGRADE_VERSION
62 |     if [[ $? != 0 ]]; then
63 |         kc_restore_from_checkpoint $CHECKPOINT_PATH
64 |         echo "[!] SBJK crashed, will not update lastrun."
65 |         rm -rf "$WORKING_DIR/root"
66 |         return
67 |     fi
68 |     kc_postupdate_tasks_ex $UPGRADE_VERSION
69 | 
70 |     echo "[-] SBJK complete. Writing new version to lastrun."
71 |     echo "VERSION=$UPGRADE_VERSION" > "$WORKING_DIR/last_run"
72 | 
73 |     echo "[-] All done. Removing temproot."
74 |     rm -rf "$WORKING_DIR/root"
75 | }
76 | 
77 | function versioncheck() {
78 |     if [[ "x$VERSION" == "x" ]]
79 |         then VERSION=0
80 |     fi
81 | 
82 |     echo "[-] Last version is $VERSION. Checking for something new..."
83 |     for CHECKER in ./simver ./versioncheck.py ./speculate.py; do
84 |         SERVER_VERSION=$($CHECKER $VERSION)
85 |         if [[ $SERVER_VERSION != $VERSION ]]; then
86 |             echo "[-] $CHECKER reports new version $SERVER_VERSION, continuing."
87 |             kc_update $SERVER_VERSION
88 |             break
89 |         fi
90 |     done
91 | }
92 | 
93 | versioncheck


--------------------------------------------------------------------------------
/misc_utils/libahff/Makefile:
--------------------------------------------------------------------------------
1 | libahff.dylib: ahff2png.c pixel.c rg_etc1.o rg_etc1wrap.o lodepng.c
2 | 	${CXX} ${CXXFLAGS} -c rg_etc1.cpp rg_etc1wrap.cpp
3 | 	${CC} ${CFLAGS} -shared -o $@ ahff2png.c lodepng.c rg_etc1.o rg_etc1wrap.o -lz -g
4 | 


--------------------------------------------------------------------------------
/misc_utils/libahff/ahff2png.c:
--------------------------------------------------------------------------------
  1 | /*
  2 |  * ahff2png.c - disunityed file tool
  3 |  * Copyright (c) 2015 The Holy Constituency of the Summer Triangle.
  4 |  * All rights reserved.
  5 |  */
  6 | 
  7 | #include <stdio.h>
  8 | #include <stdlib.h>
  9 | #include <unistd.h>
 10 | #include <assert.h>
 11 | #include <fcntl.h>
 12 | 
 13 | #include "lodepng.h"
 14 | 
 15 | typedef union {
 16 |     unsigned char b[4];
 17 |     uint32_t n;
 18 | } byte_addressable_uint32;
 19 | typedef unsigned char byte;
 20 | 
 21 | // the Ad-Hoc File Format is intentionally very simple so i don't have
 22 | // to write so much java
 23 | // packed should have no effect, this is just for safety reasons
 24 | typedef union {
 25 |     byte b[16];
 26 |     struct __attribute__((packed)) {
 27 |         uint32_t width;
 28 |         uint32_t height;
 29 |         uint32_t datsize;
 30 |         uint32_t pixel_format;
 31 |     } s;
 32 | } ahff_header_t;
 33 | 
 34 | enum /* pixel_format_t */ {
 35 |     A8 = 1,
 36 |     RGB24 = 3,
 37 |     RGBA32 = 4,
 38 |     ARGB32 = 5,
 39 |     RGB565 = 7,
 40 |     RGBA4444 = 13,
 41 |     ETC1RGB = 34,
 42 | };
 43 | 
 44 | #define READ_FULLY(fd, buf, size) do { \
 45 |     size_t _eval_one = (size); \
 46 |     size_t nread = read(fd, buf, _eval_one); \
 47 |     assert(nread == _eval_one || !"READ_FULLY did not read fully."); \
 48 | } while(0)
 49 | 
 50 | #include "pixel.c"
 51 | 
 52 | void flip_image_sideways(byte *buf, uint32_t width, uint32_t height) {
 53 |     byte *work = malloc(width * 4);
 54 |     byte *worp = work;
 55 | 
 56 |     for (int row = 0; row < height; ++row) {
 57 |         byte *crow = buf + (row * width * 4);
 58 |         worp = work;
 59 | 
 60 |         for (size_t i = (width - 1) * 4; i > 0; i -= 4) {
 61 |             memcpy(worp, crow + i, 4);
 62 |             worp += 4;
 63 |         }
 64 | 
 65 |         memcpy(crow, work, width * 4);
 66 |     }
 67 | 
 68 |     free(work);
 69 | }
 70 | 
 71 | void flip_image_upside_down(byte *buf, uint32_t width, uint32_t height) {
 72 |     byte *work = malloc(width * 4);
 73 | 
 74 |     for (int row = 0, target_row = height - 1; row < (height / 2); ++row, --target_row) {
 75 |         memcpy(work, buf + (target_row * width * 4), width * 4);
 76 |         memcpy(buf + (target_row * width * 4), buf + (row * width * 4), width * 4);
 77 |         memcpy(buf + (row * width * 4), work, width * 4);
 78 |     }
 79 | 
 80 |     free(work);
 81 | }
 82 | 
 83 | int ahff_encode_texdata(int fmt,
 84 |                         int width, int height,
 85 |                         size_t len, const uint8_t *data,
 86 |                         const char *out_path) {
 87 |     unsigned char *out = calloc(width * height, 4);
 88 |     uint32_t point_count = width * height;
 89 |     uint32_t expect_size = 0;
 90 |     int has_alpha = 1;
 91 | 
 92 |     switch (fmt) {
 93 |         case A8:
 94 |             expect_size = point_count;
 95 |             copy_1bpp_alpha(data, expect_size, out);
 96 |             break;
 97 |         case RGB24:
 98 |             expect_size = point_count * 3;
 99 |             copy_3bpp_rgb(data, expect_size, out);
100 |             has_alpha = 0;
101 |             break;
102 |         case RGB565:
103 |             expect_size = point_count * 2;
104 |             copy_2bpp_rgb565(data, expect_size, out);
105 |             has_alpha = 0;
106 |             break;
107 |         case RGBA4444:
108 |             expect_size = point_count * 2;
109 |             copy_2bpp_rgba4444(data, expect_size, out);
110 |             break;
111 |         case RGBA32:
112 |         case ARGB32: {
113 |             expect_size = point_count * 4;
114 |             memcpy(out, data, expect_size);
115 | 
116 |             if (fmt == ARGB32) {
117 |                 for (uint8_t *pbase = out; pbase < (out + expect_size); pbase += 4) {
118 |                     uint8_t tmp = pbase[0];
119 |                     pbase[0] = pbase[1];
120 |                     pbase[1] = pbase[2];
121 |                     pbase[2] = pbase[3];
122 |                     pbase[3] = tmp;
123 |                 }
124 |             }
125 | 
126 |             break;
127 |         }
128 |         case ETC1RGB:
129 |             /* ETC1 encodes 4x4 blocks.
130 |              * So w and h must be multiples of 4. */
131 |             assert(width % 4 == 0);
132 |             assert(height % 4 == 0);
133 |             expect_size = point_count / 2;
134 |             copy_etc1_rgb(data, expect_size, out, width);
135 |             has_alpha = 0;
136 |             break;
137 |         default:
138 |             fprintf(stderr, "unknown itype %d\n", fmt);
139 |             goto end;
140 |     }
141 | 
142 |     // flip_image_sideways(out, info.width, info.height);
143 |     flip_image_upside_down(out, width, height);
144 |     lodepng_encode32_file(out_path, out, width, height);
145 | 
146 |   end:
147 |     free(out);
148 |     return 0;
149 | }
150 | 


--------------------------------------------------------------------------------
/misc_utils/libahff/lodepng.h:
--------------------------------------------------------------------------------
   1 | /*
   2 | LodePNG version 20140801
   3 | 
   4 | Copyright (c) 2005-2014 Lode Vandevenne
   5 | 
   6 | This software is provided 'as-is', without any express or implied
   7 | warranty. In no event will the authors be held liable for any damages
   8 | arising from the use of this software.
   9 | 
  10 | Permission is granted to anyone to use this software for any purpose,
  11 | including commercial applications, and to alter it and redistribute it
  12 | freely, subject to the following restrictions:
  13 | 
  14 |     1. The origin of this software must not be misrepresented; you must not
  15 |     claim that you wrote the original software. If you use this software
  16 |     in a product, an acknowledgment in the product documentation would be
  17 |     appreciated but is not required.
  18 | 
  19 |     2. Altered source versions must be plainly marked as such, and must not be
  20 |     misrepresented as being the original software.
  21 | 
  22 |     3. This notice may not be removed or altered from any source
  23 |     distribution.
  24 | */
  25 | 
  26 | #ifndef LODEPNG_H
  27 | #define LODEPNG_H
  28 | 
  29 | #include <string.h> /*for size_t*/
  30 | 
  31 | #ifdef __cplusplus
  32 | #include <vector>
  33 | #include <string>
  34 | #endif /*__cplusplus*/
  35 | 
  36 | /*
  37 | The following #defines are used to create code sections. They can be disabled
  38 | to disable code sections, which can give faster compile time and smaller binary.
  39 | The "NO_COMPILE" defines are designed to be used to pass as defines to the
  40 | compiler command to disable them without modifying this header, e.g.
  41 | -DLODEPNG_NO_COMPILE_ZLIB for gcc.
  42 | */
  43 | /*deflate & zlib. If disabled, you must specify alternative zlib functions in
  44 | the custom_zlib field of the compress and decompress settings*/
  45 | #ifndef LODEPNG_NO_COMPILE_ZLIB
  46 | #define LODEPNG_COMPILE_ZLIB
  47 | #endif
  48 | /*png encoder and png decoder*/
  49 | #ifndef LODEPNG_NO_COMPILE_PNG
  50 | #define LODEPNG_COMPILE_PNG
  51 | #endif
  52 | /*deflate&zlib decoder and png decoder*/
  53 | #ifndef LODEPNG_NO_COMPILE_DECODER
  54 | #define LODEPNG_COMPILE_DECODER
  55 | #endif
  56 | /*deflate&zlib encoder and png encoder*/
  57 | #ifndef LODEPNG_NO_COMPILE_ENCODER
  58 | #define LODEPNG_COMPILE_ENCODER
  59 | #endif
  60 | /*the optional built in harddisk file loading and saving functions*/
  61 | #ifndef LODEPNG_NO_COMPILE_DISK
  62 | #define LODEPNG_COMPILE_DISK
  63 | #endif
  64 | /*support for chunks other than IHDR, IDAT, PLTE, tRNS, IEND: ancillary and unknown chunks*/
  65 | #ifndef LODEPNG_NO_COMPILE_ANCILLARY_CHUNKS
  66 | #define LODEPNG_COMPILE_ANCILLARY_CHUNKS
  67 | #endif
  68 | /*ability to convert error numerical codes to English text string*/
  69 | #ifndef LODEPNG_NO_COMPILE_ERROR_TEXT
  70 | #define LODEPNG_COMPILE_ERROR_TEXT
  71 | #endif
  72 | /*Compile the default allocators (C's free, malloc and realloc). If you disable this,
  73 | you can define the functions lodepng_free, lodepng_malloc and lodepng_realloc in your
  74 | source files with custom allocators.*/
  75 | #ifndef LODEPNG_NO_COMPILE_ALLOCATORS
  76 | #define LODEPNG_COMPILE_ALLOCATORS
  77 | #endif
  78 | /*compile the C++ version (you can disable the C++ wrapper here even when compiling for C++)*/
  79 | #ifdef __cplusplus
  80 | #ifndef LODEPNG_NO_COMPILE_CPP
  81 | #define LODEPNG_COMPILE_CPP
  82 | #endif
  83 | #endif
  84 | 
  85 | #ifdef LODEPNG_COMPILE_PNG
  86 | /*The PNG color types (also used for raw).*/
  87 | typedef enum LodePNGColorType
  88 | {
  89 |   LCT_GREY = 0, /*greyscale: 1,2,4,8,16 bit*/
  90 |   LCT_RGB = 2, /*RGB: 8,16 bit*/
  91 |   LCT_PALETTE = 3, /*palette: 1,2,4,8 bit*/
  92 |   LCT_GREY_ALPHA = 4, /*greyscale with alpha: 8,16 bit*/
  93 |   LCT_RGBA = 6 /*RGB with alpha: 8,16 bit*/
  94 | } LodePNGColorType;
  95 | 
  96 | #ifdef LODEPNG_COMPILE_DECODER
  97 | /*
  98 | Converts PNG data in memory to raw pixel data.
  99 | out: Output parameter. Pointer to buffer that will contain the raw pixel data.
 100 |      After decoding, its size is w * h * (bytes per pixel) bytes larger than
 101 |      initially. Bytes per pixel depends on colortype and bitdepth.
 102 |      Must be freed after usage with free(*out).
 103 |      Note: for 16-bit per channel colors, uses big endian format like PNG does.
 104 | w: Output parameter. Pointer to width of pixel data.
 105 | h: Output parameter. Pointer to height of pixel data.
 106 | in: Memory buffer with the PNG file.
 107 | insize: size of the in buffer.
 108 | colortype: the desired color type for the raw output image. See explanation on PNG color types.
 109 | bitdepth: the desired bit depth for the raw output image. See explanation on PNG color types.
 110 | Return value: LodePNG error code (0 means no error).
 111 | */
 112 | unsigned lodepng_decode_memory(unsigned char** out, unsigned* w, unsigned* h,
 113 |                                const unsigned char* in, size_t insize,
 114 |                                LodePNGColorType colortype, unsigned bitdepth);
 115 | 
 116 | /*Same as lodepng_decode_memory, but always decodes to 32-bit RGBA raw image*/
 117 | unsigned lodepng_decode32(unsigned char** out, unsigned* w, unsigned* h,
 118 |                           const unsigned char* in, size_t insize);
 119 | 
 120 | /*Same as lodepng_decode_memory, but always decodes to 24-bit RGB raw image*/
 121 | unsigned lodepng_decode24(unsigned char** out, unsigned* w, unsigned* h,
 122 |                           const unsigned char* in, size_t insize);
 123 | 
 124 | #ifdef LODEPNG_COMPILE_DISK
 125 | /*
 126 | Load PNG from disk, from file with given name.
 127 | Same as the other decode functions, but instead takes a filename as input.
 128 | */
 129 | unsigned lodepng_decode_file(unsigned char** out, unsigned* w, unsigned* h,
 130 |                              const char* filename,
 131 |                              LodePNGColorType colortype, unsigned bitdepth);
 132 | 
 133 | /*Same as lodepng_decode_file, but always decodes to 32-bit RGBA raw image.*/
 134 | unsigned lodepng_decode32_file(unsigned char** out, unsigned* w, unsigned* h,
 135 |                                const char* filename);
 136 | 
 137 | /*Same as lodepng_decode_file, but always decodes to 24-bit RGB raw image.*/
 138 | unsigned lodepng_decode24_file(unsigned char** out, unsigned* w, unsigned* h,
 139 |                                const char* filename);
 140 | #endif /*LODEPNG_COMPILE_DISK*/
 141 | #endif /*LODEPNG_COMPILE_DECODER*/
 142 | 
 143 | 
 144 | #ifdef LODEPNG_COMPILE_ENCODER
 145 | /*
 146 | Converts raw pixel data into a PNG image in memory. The colortype and bitdepth
 147 |   of the output PNG image cannot be chosen, they are automatically determined
 148 |   by the colortype, bitdepth and content of the input pixel data.
 149 |   Note: for 16-bit per channel colors, needs big endian format like PNG does.
 150 | out: Output parameter. Pointer to buffer that will contain the PNG image data.
 151 |      Must be freed after usage with free(*out).
 152 | outsize: Output parameter. Pointer to the size in bytes of the out buffer.
 153 | image: The raw pixel data to encode. The size of this buffer should be
 154 |        w * h * (bytes per pixel), bytes per pixel depends on colortype and bitdepth.
 155 | w: width of the raw pixel data in pixels.
 156 | h: height of the raw pixel data in pixels.
 157 | colortype: the color type of the raw input image. See explanation on PNG color types.
 158 | bitdepth: the bit depth of the raw input image. See explanation on PNG color types.
 159 | Return value: LodePNG error code (0 means no error).
 160 | */
 161 | unsigned lodepng_encode_memory(unsigned char** out, size_t* outsize,
 162 |                                const unsigned char* image, unsigned w, unsigned h,
 163 |                                LodePNGColorType colortype, unsigned bitdepth);
 164 | 
 165 | /*Same as lodepng_encode_memory, but always encodes from 32-bit RGBA raw image.*/
 166 | unsigned lodepng_encode32(unsigned char** out, size_t* outsize,
 167 |                           const unsigned char* image, unsigned w, unsigned h);
 168 | 
 169 | /*Same as lodepng_encode_memory, but always encodes from 24-bit RGB raw image.*/
 170 | unsigned lodepng_encode24(unsigned char** out, size_t* outsize,
 171 |                           const unsigned char* image, unsigned w, unsigned h);
 172 | 
 173 | #ifdef LODEPNG_COMPILE_DISK
 174 | /*
 175 | Converts raw pixel data into a PNG file on disk.
 176 | Same as the other encode functions, but instead takes a filename as output.
 177 | NOTE: This overwrites existing files without warning!
 178 | */
 179 | unsigned lodepng_encode_file(const char* filename,
 180 |                              const unsigned char* image, unsigned w, unsigned h,
 181 |                              LodePNGColorType colortype, unsigned bitdepth);
 182 | 
 183 | /*Same as lodepng_encode_file, but always encodes from 32-bit RGBA raw image.*/
 184 | unsigned lodepng_encode32_file(const char* filename,
 185 |                                const unsigned char* image, unsigned w, unsigned h);
 186 | 
 187 | /*Same as lodepng_encode_file, but always encodes from 24-bit RGB raw image.*/
 188 | unsigned lodepng_encode24_file(const char* filename,
 189 |                                const unsigned char* image, unsigned w, unsigned h);
 190 | #endif /*LODEPNG_COMPILE_DISK*/
 191 | #endif /*LODEPNG_COMPILE_ENCODER*/
 192 | 
 193 | 
 194 | #ifdef LODEPNG_COMPILE_CPP
 195 | namespace lodepng
 196 | {
 197 | #ifdef LODEPNG_COMPILE_DECODER
 198 | /*Same as lodepng_decode_memory, but decodes to an std::vector. The colortype
 199 | is the format to output the pixels to. Default is RGBA 8-bit per channel.*/
 200 | unsigned decode(std::vector<unsigned char>& out, unsigned& w, unsigned& h,
 201 |                 const unsigned char* in, size_t insize,
 202 |                 LodePNGColorType colortype = LCT_RGBA, unsigned bitdepth = 8);
 203 | unsigned decode(std::vector<unsigned char>& out, unsigned& w, unsigned& h,
 204 |                 const std::vector<unsigned char>& in,
 205 |                 LodePNGColorType colortype = LCT_RGBA, unsigned bitdepth = 8);
 206 | #ifdef LODEPNG_COMPILE_DISK
 207 | /*
 208 | Converts PNG file from disk to raw pixel data in memory.
 209 | Same as the other decode functions, but instead takes a filename as input.
 210 | */
 211 | unsigned decode(std::vector<unsigned char>& out, unsigned& w, unsigned& h,
 212 |                 const std::string& filename,
 213 |                 LodePNGColorType colortype = LCT_RGBA, unsigned bitdepth = 8);
 214 | #endif //LODEPNG_COMPILE_DISK
 215 | #endif //LODEPNG_COMPILE_DECODER
 216 | 
 217 | #ifdef LODEPNG_COMPILE_ENCODER
 218 | /*Same as lodepng_encode_memory, but encodes to an std::vector. colortype
 219 | is that of the raw input data. The output PNG color type will be auto chosen.*/
 220 | unsigned encode(std::vector<unsigned char>& out,
 221 |                 const unsigned char* in, unsigned w, unsigned h,
 222 |                 LodePNGColorType colortype = LCT_RGBA, unsigned bitdepth = 8);
 223 | unsigned encode(std::vector<unsigned char>& out,
 224 |                 const std::vector<unsigned char>& in, unsigned w, unsigned h,
 225 |                 LodePNGColorType colortype = LCT_RGBA, unsigned bitdepth = 8);
 226 | #ifdef LODEPNG_COMPILE_DISK
 227 | /*
 228 | Converts 32-bit RGBA raw pixel data into a PNG file on disk.
 229 | Same as the other encode functions, but instead takes a filename as output.
 230 | NOTE: This overwrites existing files without warning!
 231 | */
 232 | unsigned encode(const std::string& filename,
 233 |                 const unsigned char* in, unsigned w, unsigned h,
 234 |                 LodePNGColorType colortype = LCT_RGBA, unsigned bitdepth = 8);
 235 | unsigned encode(const std::string& filename,
 236 |                 const std::vector<unsigned char>& in, unsigned w, unsigned h,
 237 |                 LodePNGColorType colortype = LCT_RGBA, unsigned bitdepth = 8);
 238 | #endif //LODEPNG_COMPILE_DISK
 239 | #endif //LODEPNG_COMPILE_ENCODER
 240 | } //namespace lodepng
 241 | #endif /*LODEPNG_COMPILE_CPP*/
 242 | #endif /*LODEPNG_COMPILE_PNG*/
 243 | 
 244 | #ifdef LODEPNG_COMPILE_ERROR_TEXT
 245 | /*Returns an English description of the numerical error code.*/
 246 | const char* lodepng_error_text(unsigned code);
 247 | #endif /*LODEPNG_COMPILE_ERROR_TEXT*/
 248 | 
 249 | #ifdef LODEPNG_COMPILE_DECODER
 250 | /*Settings for zlib decompression*/
 251 | typedef struct LodePNGDecompressSettings LodePNGDecompressSettings;
 252 | struct LodePNGDecompressSettings
 253 | {
 254 |   unsigned ignore_adler32; /*if 1, continue and don't give an error message if the Adler32 checksum is corrupted*/
 255 | 
 256 |   /*use custom zlib decoder instead of built in one (default: null)*/
 257 |   unsigned (*custom_zlib)(unsigned char**, size_t*,
 258 |                           const unsigned char*, size_t,
 259 |                           const LodePNGDecompressSettings*);
 260 |   /*use custom deflate decoder instead of built in one (default: null)
 261 |   if custom_zlib is used, custom_deflate is ignored since only the built in
 262 |   zlib function will call custom_deflate*/
 263 |   unsigned (*custom_inflate)(unsigned char**, size_t*,
 264 |                              const unsigned char*, size_t,
 265 |                              const LodePNGDecompressSettings*);
 266 | 
 267 |   const void* custom_context; /*optional custom settings for custom functions*/
 268 | };
 269 | 
 270 | extern const LodePNGDecompressSettings lodepng_default_decompress_settings;
 271 | void lodepng_decompress_settings_init(LodePNGDecompressSettings* settings);
 272 | #endif /*LODEPNG_COMPILE_DECODER*/
 273 | 
 274 | #ifdef LODEPNG_COMPILE_ENCODER
 275 | /*
 276 | Settings for zlib compression. Tweaking these settings tweaks the balance
 277 | between speed and compression ratio.
 278 | */
 279 | typedef struct LodePNGCompressSettings LodePNGCompressSettings;
 280 | struct LodePNGCompressSettings /*deflate = compress*/
 281 | {
 282 |   /*LZ77 related settings*/
 283 |   unsigned btype; /*the block type for LZ (0, 1, 2 or 3, see zlib standard). Should be 2 for proper compression.*/
 284 |   unsigned use_lz77; /*whether or not to use LZ77. Should be 1 for proper compression.*/
 285 |   unsigned windowsize; /*must be a power of two <= 32768. higher compresses more but is slower. Default value: 2048.*/
 286 |   unsigned minmatch; /*mininum lz77 length. 3 is normally best, 6 can be better for some PNGs. Default: 0*/
 287 |   unsigned nicematch; /*stop searching if >= this length found. Set to 258 for best compression. Default: 128*/
 288 |   unsigned lazymatching; /*use lazy matching: better compression but a bit slower. Default: true*/
 289 | 
 290 |   /*use custom zlib encoder instead of built in one (default: null)*/
 291 |   unsigned (*custom_zlib)(unsigned char**, size_t*,
 292 |                           const unsigned char*, size_t,
 293 |                           const LodePNGCompressSettings*);
 294 |   /*use custom deflate encoder instead of built in one (default: null)
 295 |   if custom_zlib is used, custom_deflate is ignored since only the built in
 296 |   zlib function will call custom_deflate*/
 297 |   unsigned (*custom_deflate)(unsigned char**, size_t*,
 298 |                              const unsigned char*, size_t,
 299 |                              const LodePNGCompressSettings*);
 300 | 
 301 |   const void* custom_context; /*optional custom settings for custom functions*/
 302 | };
 303 | 
 304 | extern const LodePNGCompressSettings lodepng_default_compress_settings;
 305 | void lodepng_compress_settings_init(LodePNGCompressSettings* settings);
 306 | #endif /*LODEPNG_COMPILE_ENCODER*/
 307 | 
 308 | #ifdef LODEPNG_COMPILE_PNG
 309 | /*
 310 | Color mode of an image. Contains all information required to decode the pixel
 311 | bits to RGBA colors. This information is the same as used in the PNG file
 312 | format, and is used both for PNG and raw image data in LodePNG.
 313 | */
 314 | typedef struct LodePNGColorMode
 315 | {
 316 |   /*header (IHDR)*/
 317 |   LodePNGColorType colortype; /*color type, see PNG standard or documentation further in this header file*/
 318 |   unsigned bitdepth;  /*bits per sample, see PNG standard or documentation further in this header file*/
 319 | 
 320 |   /*
 321 |   palette (PLTE and tRNS)
 322 | 
 323 |   Dynamically allocated with the colors of the palette, including alpha.
 324 |   When encoding a PNG, to store your colors in the palette of the LodePNGColorMode, first use
 325 |   lodepng_palette_clear, then for each color use lodepng_palette_add.
 326 |   If you encode an image without alpha with palette, don't forget to put value 255 in each A byte of the palette.
 327 | 
 328 |   When decoding, by default you can ignore this palette, since LodePNG already
 329 |   fills the palette colors in the pixels of the raw RGBA output.
 330 | 
 331 |   The palette is only supported for color type 3.
 332 |   */
 333 |   unsigned char* palette; /*palette in RGBARGBA... order. When allocated, must be either 0, or have size 1024*/
 334 |   size_t palettesize; /*palette size in number of colors (amount of bytes is 4 * palettesize)*/
 335 | 
 336 |   /*
 337 |   transparent color key (tRNS)
 338 | 
 339 |   This color uses the same bit depth as the bitdepth value in this struct, which can be 1-bit to 16-bit.
 340 |   For greyscale PNGs, r, g and b will all 3 be set to the same.
 341 | 
 342 |   When decoding, by default you can ignore this information, since LodePNG sets
 343 |   pixels with this key to transparent already in the raw RGBA output.
 344 | 
 345 |   The color key is only supported for color types 0 and 2.
 346 |   */
 347 |   unsigned key_defined; /*is a transparent color key given? 0 = false, 1 = true*/
 348 |   unsigned key_r;       /*red/greyscale component of color key*/
 349 |   unsigned key_g;       /*green component of color key*/
 350 |   unsigned key_b;       /*blue component of color key*/
 351 | } LodePNGColorMode;
 352 | 
 353 | /*init, cleanup and copy functions to use with this struct*/
 354 | void lodepng_color_mode_init(LodePNGColorMode* info);
 355 | void lodepng_color_mode_cleanup(LodePNGColorMode* info);
 356 | /*return value is error code (0 means no error)*/
 357 | unsigned lodepng_color_mode_copy(LodePNGColorMode* dest, const LodePNGColorMode* source);
 358 | 
 359 | void lodepng_palette_clear(LodePNGColorMode* info);
 360 | /*add 1 color to the palette*/
 361 | unsigned lodepng_palette_add(LodePNGColorMode* info,
 362 |                              unsigned char r, unsigned char g, unsigned char b, unsigned char a);
 363 | 
 364 | /*get the total amount of bits per pixel, based on colortype and bitdepth in the struct*/
 365 | unsigned lodepng_get_bpp(const LodePNGColorMode* info);
 366 | /*get the amount of color channels used, based on colortype in the struct.
 367 | If a palette is used, it counts as 1 channel.*/
 368 | unsigned lodepng_get_channels(const LodePNGColorMode* info);
 369 | /*is it a greyscale type? (only colortype 0 or 4)*/
 370 | unsigned lodepng_is_greyscale_type(const LodePNGColorMode* info);
 371 | /*has it got an alpha channel? (only colortype 2 or 6)*/
 372 | unsigned lodepng_is_alpha_type(const LodePNGColorMode* info);
 373 | /*has it got a palette? (only colortype 3)*/
 374 | unsigned lodepng_is_palette_type(const LodePNGColorMode* info);
 375 | /*only returns true if there is a palette and there is a value in the palette with alpha < 255.
 376 | Loops through the palette to check this.*/
 377 | unsigned lodepng_has_palette_alpha(const LodePNGColorMode* info);
 378 | /*
 379 | Check if the given color info indicates the possibility of having non-opaque pixels in the PNG image.
 380 | Returns true if the image can have translucent or invisible pixels (it still be opaque if it doesn't use such pixels).
 381 | Returns false if the image can only have opaque pixels.
 382 | In detail, it returns true only if it's a color type with alpha, or has a palette with non-opaque values,
 383 | or if "key_defined" is true.
 384 | */
 385 | unsigned lodepng_can_have_alpha(const LodePNGColorMode* info);
 386 | /*Returns the byte size of a raw image buffer with given width, height and color mode*/
 387 | size_t lodepng_get_raw_size(unsigned w, unsigned h, const LodePNGColorMode* color);
 388 | 
 389 | #ifdef LODEPNG_COMPILE_ANCILLARY_CHUNKS
 390 | /*The information of a Time chunk in PNG.*/
 391 | typedef struct LodePNGTime
 392 | {
 393 |   unsigned year;    /*2 bytes used (0-65535)*/
 394 |   unsigned month;   /*1-12*/
 395 |   unsigned day;     /*1-31*/
 396 |   unsigned hour;    /*0-23*/
 397 |   unsigned minute;  /*0-59*/
 398 |   unsigned second;  /*0-60 (to allow for leap seconds)*/
 399 | } LodePNGTime;
 400 | #endif /*LODEPNG_COMPILE_ANCILLARY_CHUNKS*/
 401 | 
 402 | /*Information about the PNG image, except pixels, width and height.*/
 403 | typedef struct LodePNGInfo
 404 | {
 405 |   /*header (IHDR), palette (PLTE) and transparency (tRNS) chunks*/
 406 |   unsigned compression_method;/*compression method of the original file. Always 0.*/
 407 |   unsigned filter_method;     /*filter method of the original file*/
 408 |   unsigned interlace_method;  /*interlace method of the original file*/
 409 |   LodePNGColorMode color;     /*color type and bits, palette and transparency of the PNG file*/
 410 | 
 411 | #ifdef LODEPNG_COMPILE_ANCILLARY_CHUNKS
 412 |   /*
 413 |   suggested background color chunk (bKGD)
 414 |   This color uses the same color mode as the PNG (except alpha channel), which can be 1-bit to 16-bit.
 415 | 
 416 |   For greyscale PNGs, r, g and b will all 3 be set to the same. When encoding
 417 |   the encoder writes the red one. For palette PNGs: When decoding, the RGB value
 418 |   will be stored, not a palette index. But when encoding, specify the index of
 419 |   the palette in background_r, the other two are then ignored.
 420 | 
 421 |   The decoder does not use this background color to edit the color of pixels.
 422 |   */
 423 |   unsigned background_defined; /*is a suggested background color given?*/
 424 |   unsigned background_r;       /*red component of suggested background color*/
 425 |   unsigned background_g;       /*green component of suggested background color*/
 426 |   unsigned background_b;       /*blue component of suggested background color*/
 427 | 
 428 |   /*
 429 |   non-international text chunks (tEXt and zTXt)
 430 | 
 431 |   The char** arrays each contain num strings. The actual messages are in
 432 |   text_strings, while text_keys are keywords that give a short description what
 433 |   the actual text represents, e.g. Title, Author, Description, or anything else.
 434 | 
 435 |   A keyword is minimum 1 character and maximum 79 characters long. It's
 436 |   discouraged to use a single line length longer than 79 characters for texts.
 437 | 
 438 |   Don't allocate these text buffers yourself. Use the init/cleanup functions
 439 |   correctly and use lodepng_add_text and lodepng_clear_text.
 440 |   */
 441 |   size_t text_num; /*the amount of texts in these char** buffers (there may be more texts in itext)*/
 442 |   char** text_keys; /*the keyword of a text chunk (e.g. "Comment")*/
 443 |   char** text_strings; /*the actual text*/
 444 | 
 445 |   /*
 446 |   international text chunks (iTXt)
 447 |   Similar to the non-international text chunks, but with additional strings
 448 |   "langtags" and "transkeys".
 449 |   */
 450 |   size_t itext_num; /*the amount of international texts in this PNG*/
 451 |   char** itext_keys; /*the English keyword of the text chunk (e.g. "Comment")*/
 452 |   char** itext_langtags; /*language tag for this text's language, ISO/IEC 646 string, e.g. ISO 639 language tag*/
 453 |   char** itext_transkeys; /*keyword translated to the international language - UTF-8 string*/
 454 |   char** itext_strings; /*the actual international text - UTF-8 string*/
 455 | 
 456 |   /*time chunk (tIME)*/
 457 |   unsigned time_defined; /*set to 1 to make the encoder generate a tIME chunk*/
 458 |   LodePNGTime time;
 459 | 
 460 |   /*phys chunk (pHYs)*/
 461 |   unsigned phys_defined; /*if 0, there is no pHYs chunk and the values below are undefined, if 1 else there is one*/
 462 |   unsigned phys_x; /*pixels per unit in x direction*/
 463 |   unsigned phys_y; /*pixels per unit in y direction*/
 464 |   unsigned phys_unit; /*may be 0 (unknown unit) or 1 (metre)*/
 465 | 
 466 |   /*
 467 |   unknown chunks
 468 |   There are 3 buffers, one for each position in the PNG where unknown chunks can appear
 469 |   each buffer contains all unknown chunks for that position consecutively
 470 |   The 3 buffers are the unknown chunks between certain critical chunks:
 471 |   0: IHDR-PLTE, 1: PLTE-IDAT, 2: IDAT-IEND
 472 |   Do not allocate or traverse this data yourself. Use the chunk traversing functions declared
 473 |   later, such as lodepng_chunk_next and lodepng_chunk_append, to read/write this struct.
 474 |   */
 475 |   unsigned char* unknown_chunks_data[3];
 476 |   size_t unknown_chunks_size[3]; /*size in bytes of the unknown chunks, given for protection*/
 477 | #endif /*LODEPNG_COMPILE_ANCILLARY_CHUNKS*/
 478 | } LodePNGInfo;
 479 | 
 480 | /*init, cleanup and copy functions to use with this struct*/
 481 | void lodepng_info_init(LodePNGInfo* info);
 482 | void lodepng_info_cleanup(LodePNGInfo* info);
 483 | /*return value is error code (0 means no error)*/
 484 | unsigned lodepng_info_copy(LodePNGInfo* dest, const LodePNGInfo* source);
 485 | 
 486 | #ifdef LODEPNG_COMPILE_ANCILLARY_CHUNKS
 487 | void lodepng_clear_text(LodePNGInfo* info); /*use this to clear the texts again after you filled them in*/
 488 | unsigned lodepng_add_text(LodePNGInfo* info, const char* key, const char* str); /*push back both texts at once*/
 489 | 
 490 | void lodepng_clear_itext(LodePNGInfo* info); /*use this to clear the itexts again after you filled them in*/
 491 | unsigned lodepng_add_itext(LodePNGInfo* info, const char* key, const char* langtag,
 492 |                            const char* transkey, const char* str); /*push back the 4 texts of 1 chunk at once*/
 493 | #endif /*LODEPNG_COMPILE_ANCILLARY_CHUNKS*/
 494 | 
 495 | /*
 496 | Converts raw buffer from one color type to another color type, based on
 497 | LodePNGColorMode structs to describe the input and output color type.
 498 | See the reference manual at the end of this header file to see which color conversions are supported.
 499 | return value = LodePNG error code (0 if all went ok, an error if the conversion isn't supported)
 500 | The out buffer must have size (w * h * bpp + 7) / 8, where bpp is the bits per pixel
 501 | of the output color type (lodepng_get_bpp).
 502 | For < 8 bpp images, there should not be padding bits at the end of scanlines.
 503 | For 16-bit per channel colors, uses big endian format like PNG does.
 504 | Return value is LodePNG error code
 505 | */
 506 | unsigned lodepng_convert(unsigned char* out, const unsigned char* in,
 507 |                          LodePNGColorMode* mode_out, const LodePNGColorMode* mode_in,
 508 |                          unsigned w, unsigned h);
 509 | 
 510 | #ifdef LODEPNG_COMPILE_DECODER
 511 | /*
 512 | Settings for the decoder. This contains settings for the PNG and the Zlib
 513 | decoder, but not the Info settings from the Info structs.
 514 | */
 515 | typedef struct LodePNGDecoderSettings
 516 | {
 517 |   LodePNGDecompressSettings zlibsettings; /*in here is the setting to ignore Adler32 checksums*/
 518 | 
 519 |   unsigned ignore_crc; /*ignore CRC checksums*/
 520 | 
 521 |   unsigned color_convert; /*whether to convert the PNG to the color type you want. Default: yes*/
 522 | 
 523 | #ifdef LODEPNG_COMPILE_ANCILLARY_CHUNKS
 524 |   unsigned read_text_chunks; /*if false but remember_unknown_chunks is true, they're stored in the unknown chunks*/
 525 |   /*store all bytes from unknown chunks in the LodePNGInfo (off by default, useful for a png editor)*/
 526 |   unsigned remember_unknown_chunks;
 527 | #endif /*LODEPNG_COMPILE_ANCILLARY_CHUNKS*/
 528 | } LodePNGDecoderSettings;
 529 | 
 530 | void lodepng_decoder_settings_init(LodePNGDecoderSettings* settings);
 531 | #endif /*LODEPNG_COMPILE_DECODER*/
 532 | 
 533 | #ifdef LODEPNG_COMPILE_ENCODER
 534 | /*automatically use color type with less bits per pixel if losslessly possible. Default: AUTO*/
 535 | typedef enum LodePNGFilterStrategy
 536 | {
 537 |   /*every filter at zero*/
 538 |   LFS_ZERO,
 539 |   /*Use filter that gives minumum sum, as described in the official PNG filter heuristic.*/
 540 |   LFS_MINSUM,
 541 |   /*Use the filter type that gives smallest Shannon entropy for this scanline. Depending
 542 |   on the image, this is better or worse than minsum.*/
 543 |   LFS_ENTROPY,
 544 |   /*
 545 |   Brute-force-search PNG filters by compressing each filter for each scanline.
 546 |   Experimental, very slow, and only rarely gives better compression than MINSUM.
 547 |   */
 548 |   LFS_BRUTE_FORCE,
 549 |   /*use predefined_filters buffer: you specify the filter type for each scanline*/
 550 |   LFS_PREDEFINED
 551 | } LodePNGFilterStrategy;
 552 | 
 553 | /*Gives characteristics about the colors of the image, which helps decide which color model to use for encoding.
 554 | Used internally by default if "auto_convert" is enabled. Public because it's useful for custom algorithms.*/
 555 | typedef struct LodePNGColorProfile
 556 | {
 557 |   unsigned colored; /*not greyscale*/
 558 |   unsigned key; /*if true, image is not opaque. Only if true and alpha is false, color key is possible.*/
 559 |   unsigned short key_r; /*these values are always in 16-bit bitdepth in the profile*/
 560 |   unsigned short key_g;
 561 |   unsigned short key_b;
 562 |   unsigned alpha; /*alpha channel or alpha palette required*/
 563 |   unsigned numcolors; /*amount of colors, up to 257. Not valid if bits == 16.*/
 564 |   unsigned char palette[1024]; /*Remembers up to the first 256 RGBA colors, in no particular order*/
 565 |   unsigned bits; /*bits per channel (not for palette). 1,2 or 4 for greyscale only. 16 if 16-bit per channel required.*/
 566 | } LodePNGColorProfile;
 567 | 
 568 | void lodepng_color_profile_init(LodePNGColorProfile* profile);
 569 | 
 570 | /*Get a LodePNGColorProfile of the image.*/
 571 | unsigned get_color_profile(LodePNGColorProfile* profile,
 572 |                            const unsigned char* image, unsigned w, unsigned h,
 573 |                            const LodePNGColorMode* mode_in);
 574 | /*The function LodePNG uses internally to decide the PNG color with auto_convert.
 575 | Chooses an optimal color model, e.g. grey if only grey pixels, palette if < 256 colors, ...*/
 576 | unsigned lodepng_auto_choose_color(LodePNGColorMode* mode_out,
 577 |                                    const unsigned char* image, unsigned w, unsigned h,
 578 |                                    const LodePNGColorMode* mode_in);
 579 | 
 580 | /*Settings for the encoder.*/
 581 | typedef struct LodePNGEncoderSettings
 582 | {
 583 |   LodePNGCompressSettings zlibsettings; /*settings for the zlib encoder, such as window size, ...*/
 584 | 
 585 |   unsigned auto_convert; /*automatically choose output PNG color type. Default: true*/
 586 | 
 587 |   /*If true, follows the official PNG heuristic: if the PNG uses a palette or lower than
 588 |   8 bit depth, set all filters to zero. Otherwise use the filter_strategy. Note that to
 589 |   completely follow the official PNG heuristic, filter_palette_zero must be true and
 590 |   filter_strategy must be LFS_MINSUM*/
 591 |   unsigned filter_palette_zero;
 592 |   /*Which filter strategy to use when not using zeroes due to filter_palette_zero.
 593 |   Set filter_palette_zero to 0 to ensure always using your chosen strategy. Default: LFS_MINSUM*/
 594 |   LodePNGFilterStrategy filter_strategy;
 595 |   /*used if filter_strategy is LFS_PREDEFINED. In that case, this must point to a buffer with
 596 |   the same length as the amount of scanlines in the image, and each value must <= 5. You
 597 |   have to cleanup this buffer, LodePNG will never free it. Don't forget that filter_palette_zero
 598 |   must be set to 0 to ensure this is also used on palette or low bitdepth images.*/
 599 |   const unsigned char* predefined_filters;
 600 | 
 601 |   /*force creating a PLTE chunk if colortype is 2 or 6 (= a suggested palette).
 602 |   If colortype is 3, PLTE is _always_ created.*/
 603 |   unsigned force_palette;
 604 | #ifdef LODEPNG_COMPILE_ANCILLARY_CHUNKS
 605 |   /*add LodePNG identifier and version as a text chunk, for debugging*/
 606 |   unsigned add_id;
 607 |   /*encode text chunks as zTXt chunks instead of tEXt chunks, and use compression in iTXt chunks*/
 608 |   unsigned text_compression;
 609 | #endif /*LODEPNG_COMPILE_ANCILLARY_CHUNKS*/
 610 | } LodePNGEncoderSettings;
 611 | 
 612 | void lodepng_encoder_settings_init(LodePNGEncoderSettings* settings);
 613 | #endif /*LODEPNG_COMPILE_ENCODER*/
 614 | 
 615 | 
 616 | #if defined(LODEPNG_COMPILE_DECODER) || defined(LODEPNG_COMPILE_ENCODER)
 617 | /*The settings, state and information for extended encoding and decoding.*/
 618 | typedef struct LodePNGState
 619 | {
 620 | #ifdef LODEPNG_COMPILE_DECODER
 621 |   LodePNGDecoderSettings decoder; /*the decoding settings*/
 622 | #endif /*LODEPNG_COMPILE_DECODER*/
 623 | #ifdef LODEPNG_COMPILE_ENCODER
 624 |   LodePNGEncoderSettings encoder; /*the encoding settings*/
 625 | #endif /*LODEPNG_COMPILE_ENCODER*/
 626 |   LodePNGColorMode info_raw; /*specifies the format in which you would like to get the raw pixel buffer*/
 627 |   LodePNGInfo info_png; /*info of the PNG image obtained after decoding*/
 628 |   unsigned error;
 629 | #ifdef LODEPNG_COMPILE_CPP
 630 |   //For the lodepng::State subclass.
 631 |   virtual ~LodePNGState(){}
 632 | #endif
 633 | } LodePNGState;
 634 | 
 635 | /*init, cleanup and copy functions to use with this struct*/
 636 | void lodepng_state_init(LodePNGState* state);
 637 | void lodepng_state_cleanup(LodePNGState* state);
 638 | void lodepng_state_copy(LodePNGState* dest, const LodePNGState* source);
 639 | #endif /* defined(LODEPNG_COMPILE_DECODER) || defined(LODEPNG_COMPILE_ENCODER) */
 640 | 
 641 | #ifdef LODEPNG_COMPILE_DECODER
 642 | /*
 643 | Same as lodepng_decode_memory, but uses a LodePNGState to allow custom settings and
 644 | getting much more information about the PNG image and color mode.
 645 | */
 646 | unsigned lodepng_decode(unsigned char** out, unsigned* w, unsigned* h,
 647 |                         LodePNGState* state,
 648 |                         const unsigned char* in, size_t insize);
 649 | 
 650 | /*
 651 | Read the PNG header, but not the actual data. This returns only the information
 652 | that is in the header chunk of the PNG, such as width, height and color type. The
 653 | information is placed in the info_png field of the LodePNGState.
 654 | */
 655 | unsigned lodepng_inspect(unsigned* w, unsigned* h,
 656 |                          LodePNGState* state,
 657 |                          const unsigned char* in, size_t insize);
 658 | #endif /*LODEPNG_COMPILE_DECODER*/
 659 | 
 660 | 
 661 | #ifdef LODEPNG_COMPILE_ENCODER
 662 | /*This function allocates the out buffer with standard malloc and stores the size in *outsize.*/
 663 | unsigned lodepng_encode(unsigned char** out, size_t* outsize,
 664 |                         const unsigned char* image, unsigned w, unsigned h,
 665 |                         LodePNGState* state);
 666 | #endif /*LODEPNG_COMPILE_ENCODER*/
 667 | 
 668 | /*
 669 | The lodepng_chunk functions are normally not needed, except to traverse the
 670 | unknown chunks stored in the LodePNGInfo struct, or add new ones to it.
 671 | It also allows traversing the chunks of an encoded PNG file yourself.
 672 | 
 673 | PNG standard chunk naming conventions:
 674 | First byte: uppercase = critical, lowercase = ancillary
 675 | Second byte: uppercase = public, lowercase = private
 676 | Third byte: must be uppercase
 677 | Fourth byte: uppercase = unsafe to copy, lowercase = safe to copy
 678 | */
 679 | 
 680 | /*get the length of the data of the chunk. Total chunk length has 12 bytes more.*/
 681 | unsigned lodepng_chunk_length(const unsigned char* chunk);
 682 | 
 683 | /*puts the 4-byte type in null terminated string*/
 684 | void lodepng_chunk_type(char type[5], const unsigned char* chunk);
 685 | 
 686 | /*check if the type is the given type*/
 687 | unsigned char lodepng_chunk_type_equals(const unsigned char* chunk, const char* type);
 688 | 
 689 | /*0: it's one of the critical chunk types, 1: it's an ancillary chunk (see PNG standard)*/
 690 | unsigned char lodepng_chunk_ancillary(const unsigned char* chunk);
 691 | 
 692 | /*0: public, 1: private (see PNG standard)*/
 693 | unsigned char lodepng_chunk_private(const unsigned char* chunk);
 694 | 
 695 | /*0: the chunk is unsafe to copy, 1: the chunk is safe to copy (see PNG standard)*/
 696 | unsigned char lodepng_chunk_safetocopy(const unsigned char* chunk);
 697 | 
 698 | /*get pointer to the data of the chunk, where the input points to the header of the chunk*/
 699 | unsigned char* lodepng_chunk_data(unsigned char* chunk);
 700 | const unsigned char* lodepng_chunk_data_const(const unsigned char* chunk);
 701 | 
 702 | /*returns 0 if the crc is correct, 1 if it's incorrect (0 for OK as usual!)*/
 703 | unsigned lodepng_chunk_check_crc(const unsigned char* chunk);
 704 | 
 705 | /*generates the correct CRC from the data and puts it in the last 4 bytes of the chunk*/
 706 | void lodepng_chunk_generate_crc(unsigned char* chunk);
 707 | 
 708 | /*iterate to next chunks. don't use on IEND chunk, as there is no next chunk then*/
 709 | unsigned char* lodepng_chunk_next(unsigned char* chunk);
 710 | const unsigned char* lodepng_chunk_next_const(const unsigned char* chunk);
 711 | 
 712 | /*
 713 | Appends chunk to the data in out. The given chunk should already have its chunk header.
 714 | The out variable and outlength are updated to reflect the new reallocated buffer.
 715 | Returns error code (0 if it went ok)
 716 | */
 717 | unsigned lodepng_chunk_append(unsigned char** out, size_t* outlength, const unsigned char* chunk);
 718 | 
 719 | /*
 720 | Appends new chunk to out. The chunk to append is given by giving its length, type
 721 | and data separately. The type is a 4-letter string.
 722 | The out variable and outlength are updated to reflect the new reallocated buffer.
 723 | Returne error code (0 if it went ok)
 724 | */
 725 | unsigned lodepng_chunk_create(unsigned char** out, size_t* outlength, unsigned length,
 726 |                               const char* type, const unsigned char* data);
 727 | 
 728 | 
 729 | /*Calculate CRC32 of buffer*/
 730 | unsigned lodepng_crc32(const unsigned char* buf, size_t len);
 731 | #endif /*LODEPNG_COMPILE_PNG*/
 732 | 
 733 | 
 734 | #ifdef LODEPNG_COMPILE_ZLIB
 735 | /*
 736 | This zlib part can be used independently to zlib compress and decompress a
 737 | buffer. It cannot be used to create gzip files however, and it only supports the
 738 | part of zlib that is required for PNG, it does not support dictionaries.
 739 | */
 740 | 
 741 | #ifdef LODEPNG_COMPILE_DECODER
 742 | /*Inflate a buffer. Inflate is the decompression step of deflate. Out buffer must be freed after use.*/
 743 | unsigned lodepng_inflate(unsigned char** out, size_t* outsize,
 744 |                          const unsigned char* in, size_t insize,
 745 |                          const LodePNGDecompressSettings* settings);
 746 | 
 747 | /*
 748 | Decompresses Zlib data. Reallocates the out buffer and appends the data. The
 749 | data must be according to the zlib specification.
 750 | Either, *out must be NULL and *outsize must be 0, or, *out must be a valid
 751 | buffer and *outsize its size in bytes. out must be freed by user after usage.
 752 | */
 753 | unsigned lodepng_zlib_decompress(unsigned char** out, size_t* outsize,
 754 |                                  const unsigned char* in, size_t insize,
 755 |                                  const LodePNGDecompressSettings* settings);
 756 | #endif /*LODEPNG_COMPILE_DECODER*/
 757 | 
 758 | #ifdef LODEPNG_COMPILE_ENCODER
 759 | /*
 760 | Compresses data with Zlib. Reallocates the out buffer and appends the data.
 761 | Zlib adds a small header and trailer around the deflate data.
 762 | The data is output in the format of the zlib specification.
 763 | Either, *out must be NULL and *outsize must be 0, or, *out must be a valid
 764 | buffer and *outsize its size in bytes. out must be freed by user after usage.
 765 | */
 766 | unsigned lodepng_zlib_compress(unsigned char** out, size_t* outsize,
 767 |                                const unsigned char* in, size_t insize,
 768 |                                const LodePNGCompressSettings* settings);
 769 | 
 770 | /*
 771 | Find length-limited Huffman code for given frequencies. This function is in the
 772 | public interface only for tests, it's used internally by lodepng_deflate.
 773 | */
 774 | unsigned lodepng_huffman_code_lengths(unsigned* lengths, const unsigned* frequencies,
 775 |                                       size_t numcodes, unsigned maxbitlen);
 776 | 
 777 | /*Compress a buffer with deflate. See RFC 1951. Out buffer must be freed after use.*/
 778 | unsigned lodepng_deflate(unsigned char** out, size_t* outsize,
 779 |                          const unsigned char* in, size_t insize,
 780 |                          const LodePNGCompressSettings* settings);
 781 | 
 782 | #endif /*LODEPNG_COMPILE_ENCODER*/
 783 | #endif /*LODEPNG_COMPILE_ZLIB*/
 784 | 
 785 | #ifdef LODEPNG_COMPILE_DISK
 786 | /*
 787 | Load a file from disk into buffer. The function allocates the out buffer, and
 788 | after usage you should free it.
 789 | out: output parameter, contains pointer to loaded buffer.
 790 | outsize: output parameter, size of the allocated out buffer
 791 | filename: the path to the file to load
 792 | return value: error code (0 means ok)
 793 | */
 794 | unsigned lodepng_load_file(unsigned char** out, size_t* outsize, const char* filename);
 795 | 
 796 | /*
 797 | Save a file from buffer to disk. Warning, if it exists, this function overwrites
 798 | the file without warning!
 799 | buffer: the buffer to write
 800 | buffersize: size of the buffer to write
 801 | filename: the path to the file to save to
 802 | return value: error code (0 means ok)
 803 | */
 804 | unsigned lodepng_save_file(const unsigned char* buffer, size_t buffersize, const char* filename);
 805 | #endif /*LODEPNG_COMPILE_DISK*/
 806 | 
 807 | #ifdef LODEPNG_COMPILE_CPP
 808 | //The LodePNG C++ wrapper uses std::vectors instead of manually allocated memory buffers.
 809 | namespace lodepng
 810 | {
 811 | #ifdef LODEPNG_COMPILE_PNG
 812 | class State : public LodePNGState
 813 | {
 814 |   public:
 815 |     State();
 816 |     State(const State& other);
 817 |     virtual ~State();
 818 |     State& operator=(const State& other);
 819 | };
 820 | 
 821 | #ifdef LODEPNG_COMPILE_DECODER
 822 | //Same as other lodepng::decode, but using a State for more settings and information.
 823 | unsigned decode(std::vector<unsigned char>& out, unsigned& w, unsigned& h,
 824 |                 State& state,
 825 |                 const unsigned char* in, size_t insize);
 826 | unsigned decode(std::vector<unsigned char>& out, unsigned& w, unsigned& h,
 827 |                 State& state,
 828 |                 const std::vector<unsigned char>& in);
 829 | #endif /*LODEPNG_COMPILE_DECODER*/
 830 | 
 831 | #ifdef LODEPNG_COMPILE_ENCODER
 832 | //Same as other lodepng::encode, but using a State for more settings and information.
 833 | unsigned encode(std::vector<unsigned char>& out,
 834 |                 const unsigned char* in, unsigned w, unsigned h,
 835 |                 State& state);
 836 | unsigned encode(std::vector<unsigned char>& out,
 837 |                 const std::vector<unsigned char>& in, unsigned w, unsigned h,
 838 |                 State& state);
 839 | #endif /*LODEPNG_COMPILE_ENCODER*/
 840 | 
 841 | #ifdef LODEPNG_COMPILE_DISK
 842 | /*
 843 | Load a file from disk into an std::vector. If the vector is empty, then either
 844 | the file doesn't exist or is an empty file.
 845 | */
 846 | void load_file(std::vector<unsigned char>& buffer, const std::string& filename);
 847 | 
 848 | /*
 849 | Save the binary data in an std::vector to a file on disk. The file is overwritten
 850 | without warning.
 851 | */
 852 | void save_file(const std::vector<unsigned char>& buffer, const std::string& filename);
 853 | #endif //LODEPNG_COMPILE_DISK
 854 | #endif //LODEPNG_COMPILE_PNG
 855 | 
 856 | #ifdef LODEPNG_COMPILE_ZLIB
 857 | #ifdef LODEPNG_COMPILE_DECODER
 858 | //Zlib-decompress an unsigned char buffer
 859 | unsigned decompress(std::vector<unsigned char>& out, const unsigned char* in, size_t insize,
 860 |                     const LodePNGDecompressSettings& settings = lodepng_default_decompress_settings);
 861 | 
 862 | //Zlib-decompress an std::vector
 863 | unsigned decompress(std::vector<unsigned char>& out, const std::vector<unsigned char>& in,
 864 |                     const LodePNGDecompressSettings& settings = lodepng_default_decompress_settings);
 865 | #endif //LODEPNG_COMPILE_DECODER
 866 | 
 867 | #ifdef LODEPNG_COMPILE_ENCODER
 868 | //Zlib-compress an unsigned char buffer
 869 | unsigned compress(std::vector<unsigned char>& out, const unsigned char* in, size_t insize,
 870 |                   const LodePNGCompressSettings& settings = lodepng_default_compress_settings);
 871 | 
 872 | //Zlib-compress an std::vector
 873 | unsigned compress(std::vector<unsigned char>& out, const std::vector<unsigned char>& in,
 874 |                   const LodePNGCompressSettings& settings = lodepng_default_compress_settings);
 875 | #endif //LODEPNG_COMPILE_ENCODER
 876 | #endif //LODEPNG_COMPILE_ZLIB
 877 | } //namespace lodepng
 878 | #endif /*LODEPNG_COMPILE_CPP*/
 879 | 
 880 | /*
 881 | TODO:
 882 | [.] test if there are no memory leaks or security exploits - done a lot but needs to be checked often
 883 | [.] check compatibility with vareous compilers  - done but needs to be redone for every newer version
 884 | [X] converting color to 16-bit per channel types
 885 | [ ] read all public PNG chunk types (but never let the color profile and gamma ones touch RGB values)
 886 | [ ] make sure encoder generates no chunks with size > (2^31)-1
 887 | [ ] partial decoding (stream processing)
 888 | [X] let the "isFullyOpaque" function check color keys and transparent palettes too
 889 | [X] better name for the variables "codes", "codesD", "codelengthcodes", "clcl" and "lldl"
 890 | [ ] don't stop decoding on errors like 69, 57, 58 (make warnings)
 891 | [ ] make option to choose if the raw image with non multiple of 8 bits per scanline should have padding bits or not
 892 | [ ] let the C++ wrapper catch exceptions coming from the standard library and return LodePNG error codes
 893 | */
 894 | 
 895 | #endif /*LODEPNG_H inclusion guard*/
 896 | 
 897 | /*
 898 | LodePNG Documentation
 899 | ---------------------
 900 | 
 901 | 0. table of contents
 902 | --------------------
 903 | 
 904 |   1. about
 905 |    1.1. supported features
 906 |    1.2. features not supported
 907 |   2. C and C++ version
 908 |   3. security
 909 |   4. decoding
 910 |   5. encoding
 911 |   6. color conversions
 912 |     6.1. PNG color types
 913 |     6.2. color conversions
 914 |     6.3. padding bits
 915 |     6.4. A note about 16-bits per channel and endianness
 916 |   7. error values
 917 |   8. chunks and PNG editing
 918 |   9. compiler support
 919 |   10. examples
 920 |    10.1. decoder C++ example
 921 |    10.2. decoder C example
 922 |   11. changes
 923 |   12. contact information
 924 | 
 925 | 
 926 | 1. about
 927 | --------
 928 | 
 929 | PNG is a file format to store raster images losslessly with good compression,
 930 | supporting different color types and alpha channel.
 931 | 
 932 | LodePNG is a PNG codec according to the Portable Network Graphics (PNG)
 933 | Specification (Second Edition) - W3C Recommendation 10 November 2003.
 934 | 
 935 | The specifications used are:
 936 | 
 937 | *) Portable Network Graphics (PNG) Specification (Second Edition):
 938 |      http://www.w3.org/TR/2003/REC-PNG-20031110
 939 | *) RFC 1950 ZLIB Compressed Data Format version 3.3:
 940 |      http://www.gzip.org/zlib/rfc-zlib.html
 941 | *) RFC 1951 DEFLATE Compressed Data Format Specification ver 1.3:
 942 |      http://www.gzip.org/zlib/rfc-deflate.html
 943 | 
 944 | The most recent version of LodePNG can currently be found at
 945 | http://lodev.org/lodepng/
 946 | 
 947 | LodePNG works both in C (ISO C90) and C++, with a C++ wrapper that adds
 948 | extra functionality.
 949 | 
 950 | LodePNG exists out of two files:
 951 | -lodepng.h: the header file for both C and C++
 952 | -lodepng.c(pp): give it the name lodepng.c or lodepng.cpp (or .cc) depending on your usage
 953 | 
 954 | If you want to start using LodePNG right away without reading this doc, get the
 955 | examples from the LodePNG website to see how to use it in code, or check the
 956 | smaller examples in chapter 13 here.
 957 | 
 958 | LodePNG is simple but only supports the basic requirements. To achieve
 959 | simplicity, the following design choices were made: There are no dependencies
 960 | on any external library. There are functions to decode and encode a PNG with
 961 | a single function call, and extended versions of these functions taking a
 962 | LodePNGState struct allowing to specify or get more information. By default
 963 | the colors of the raw image are always RGB or RGBA, no matter what color type
 964 | the PNG file uses. To read and write files, there are simple functions to
 965 | convert the files to/from buffers in memory.
 966 | 
 967 | This all makes LodePNG suitable for loading textures in games, demos and small
 968 | programs, ... It's less suitable for full fledged image editors, loading PNGs
 969 | over network (it requires all the image data to be available before decoding can
 970 | begin), life-critical systems, ...
 971 | 
 972 | 1.1. supported features
 973 | -----------------------
 974 | 
 975 | The following features are supported by the decoder:
 976 | 
 977 | *) decoding of PNGs with any color type, bit depth and interlace mode, to a 24- or 32-bit color raw image,
 978 |    or the same color type as the PNG
 979 | *) encoding of PNGs, from any raw image to 24- or 32-bit color, or the same color type as the raw image
 980 | *) Adam7 interlace and deinterlace for any color type
 981 | *) loading the image from harddisk or decoding it from a buffer from other sources than harddisk
 982 | *) support for alpha channels, including RGBA color model, translucent palettes and color keying
 983 | *) zlib decompression (inflate)
 984 | *) zlib compression (deflate)
 985 | *) CRC32 and ADLER32 checksums
 986 | *) handling of unknown chunks, allowing making a PNG editor that stores custom and unknown chunks.
 987 | *) the following chunks are supported (generated/interpreted) by both encoder and decoder:
 988 |     IHDR: header information
 989 |     PLTE: color palette
 990 |     IDAT: pixel data
 991 |     IEND: the final chunk
 992 |     tRNS: transparency for palettized images
 993 |     tEXt: textual information
 994 |     zTXt: compressed textual information
 995 |     iTXt: international textual information
 996 |     bKGD: suggested background color
 997 |     pHYs: physical dimensions
 998 |     tIME: modification time
 999 | 
1000 | 1.2. features not supported
1001 | ---------------------------
1002 | 
1003 | The following features are _not_ supported:
1004 | 
1005 | *) some features needed to make a conformant PNG-Editor might be still missing.
1006 | *) partial loading/stream processing. All data must be available and is processed in one call.
1007 | *) The following public chunks are not supported but treated as unknown chunks by LodePNG
1008 |     cHRM, gAMA, iCCP, sRGB, sBIT, hIST, sPLT
1009 |    Some of these are not supported on purpose: LodePNG wants to provide the RGB values
1010 |    stored in the pixels, not values modified by system dependent gamma or color models.
1011 | 
1012 | 
1013 | 2. C and C++ version
1014 | --------------------
1015 | 
1016 | The C version uses buffers allocated with alloc that you need to free()
1017 | yourself. You need to use init and cleanup functions for each struct whenever
1018 | using a struct from the C version to avoid exploits and memory leaks.
1019 | 
1020 | The C++ version has extra functions with std::vectors in the interface and the
1021 | lodepng::State class which is a LodePNGState with constructor and destructor.
1022 | 
1023 | These files work without modification for both C and C++ compilers because all
1024 | the additional C++ code is in "#ifdef __cplusplus" blocks that make C-compilers
1025 | ignore it, and the C code is made to compile both with strict ISO C90 and C++.
1026 | 
1027 | To use the C++ version, you need to rename the source file to lodepng.cpp
1028 | (instead of lodepng.c), and compile it with a C++ compiler.
1029 | 
1030 | To use the C version, you need to rename the source file to lodepng.c (instead
1031 | of lodepng.cpp), and compile it with a C compiler.
1032 | 
1033 | 
1034 | 3. Security
1035 | -----------
1036 | 
1037 | Even if carefully designed, it's always possible that LodePNG contains possible
1038 | exploits. If you discover one, please let me know, and it will be fixed.
1039 | 
1040 | When using LodePNG, care has to be taken with the C version of LodePNG, as well
1041 | as the C-style structs when working with C++. The following conventions are used
1042 | for all C-style structs:
1043 | 
1044 | -if a struct has a corresponding init function, always call the init function when making a new one
1045 | -if a struct has a corresponding cleanup function, call it before the struct disappears to avoid memory leaks
1046 | -if a struct has a corresponding copy function, use the copy function instead of "=".
1047 |  The destination must also be inited already.
1048 | 
1049 | 
1050 | 4. Decoding
1051 | -----------
1052 | 
1053 | Decoding converts a PNG compressed image to a raw pixel buffer.
1054 | 
1055 | Most documentation on using the decoder is at its declarations in the header
1056 | above. For C, simple decoding can be done with functions such as
1057 | lodepng_decode32, and more advanced decoding can be done with the struct
1058 | LodePNGState and lodepng_decode. For C++, all decoding can be done with the
1059 | various lodepng::decode functions, and lodepng::State can be used for advanced
1060 | features.
1061 | 
1062 | When using the LodePNGState, it uses the following fields for decoding:
1063 | *) LodePNGInfo info_png: it stores extra information about the PNG (the input) in here
1064 | *) LodePNGColorMode info_raw: here you can say what color mode of the raw image (the output) you want to get
1065 | *) LodePNGDecoderSettings decoder: you can specify a few extra settings for the decoder to use
1066 | 
1067 | LodePNGInfo info_png
1068 | --------------------
1069 | 
1070 | After decoding, this contains extra information of the PNG image, except the actual
1071 | pixels, width and height because these are already gotten directly from the decoder
1072 | functions.
1073 | 
1074 | It contains for example the original color type of the PNG image, text comments,
1075 | suggested background color, etc... More details about the LodePNGInfo struct are
1076 | at its declaration documentation.
1077 | 
1078 | LodePNGColorMode info_raw
1079 | -------------------------
1080 | 
1081 | When decoding, here you can specify which color type you want
1082 | the resulting raw image to be. If this is different from the colortype of the
1083 | PNG, then the decoder will automatically convert the result. This conversion
1084 | always works, except if you want it to convert a color PNG to greyscale or to
1085 | a palette with missing colors.
1086 | 
1087 | By default, 32-bit color is used for the result.
1088 | 
1089 | LodePNGDecoderSettings decoder
1090 | ------------------------------
1091 | 
1092 | The settings can be used to ignore the errors created by invalid CRC and Adler32
1093 | chunks, and to disable the decoding of tEXt chunks.
1094 | 
1095 | There's also a setting color_convert, true by default. If false, no conversion
1096 | is done, the resulting data will be as it was in the PNG (after decompression)
1097 | and you'll have to puzzle the colors of the pixels together yourself using the
1098 | color type information in the LodePNGInfo.
1099 | 
1100 | 
1101 | 5. Encoding
1102 | -----------
1103 | 
1104 | Encoding converts a raw pixel buffer to a PNG compressed image.
1105 | 
1106 | Most documentation on using the encoder is at its declarations in the header
1107 | above. For C, simple encoding can be done with functions such as
1108 | lodepng_encode32, and more advanced decoding can be done with the struct
1109 | LodePNGState and lodepng_encode. For C++, all encoding can be done with the
1110 | various lodepng::encode functions, and lodepng::State can be used for advanced
1111 | features.
1112 | 
1113 | Like the decoder, the encoder can also give errors. However it gives less errors
1114 | since the encoder input is trusted, the decoder input (a PNG image that could
1115 | be forged by anyone) is not trusted.
1116 | 
1117 | When using the LodePNGState, it uses the following fields for encoding:
1118 | *) LodePNGInfo info_png: here you specify how you want the PNG (the output) to be.
1119 | *) LodePNGColorMode info_raw: here you say what color type of the raw image (the input) has
1120 | *) LodePNGEncoderSettings encoder: you can specify a few settings for the encoder to use
1121 | 
1122 | LodePNGInfo info_png
1123 | --------------------
1124 | 
1125 | When encoding, you use this the opposite way as when decoding: for encoding,
1126 | you fill in the values you want the PNG to have before encoding. By default it's
1127 | not needed to specify a color type for the PNG since it's automatically chosen,
1128 | but it's possible to choose it yourself given the right settings.
1129 | 
1130 | The encoder will not always exactly match the LodePNGInfo struct you give,
1131 | it tries as close as possible. Some things are ignored by the encoder. The
1132 | encoder uses, for example, the following settings from it when applicable:
1133 | colortype and bitdepth, text chunks, time chunk, the color key, the palette, the
1134 | background color, the interlace method, unknown chunks, ...
1135 | 
1136 | When encoding to a PNG with colortype 3, the encoder will generate a PLTE chunk.
1137 | If the palette contains any colors for which the alpha channel is not 255 (so
1138 | there are translucent colors in the palette), it'll add a tRNS chunk.
1139 | 
1140 | LodePNGColorMode info_raw
1141 | -------------------------
1142 | 
1143 | You specify the color type of the raw image that you give to the input here,
1144 | including a possible transparent color key and palette you happen to be using in
1145 | your raw image data.
1146 | 
1147 | By default, 32-bit color is assumed, meaning your input has to be in RGBA
1148 | format with 4 bytes (unsigned chars) per pixel.
1149 | 
1150 | LodePNGEncoderSettings encoder
1151 | ------------------------------
1152 | 
1153 | The following settings are supported (some are in sub-structs):
1154 | *) auto_convert: when this option is enabled, the encoder will
1155 | automatically choose the smallest possible color mode (including color key) that
1156 | can encode the colors of all pixels without information loss.
1157 | *) btype: the block type for LZ77. 0 = uncompressed, 1 = fixed huffman tree,
1158 |    2 = dynamic huffman tree (best compression). Should be 2 for proper
1159 |    compression.
1160 | *) use_lz77: whether or not to use LZ77 for compressed block types. Should be
1161 |    true for proper compression.
1162 | *) windowsize: the window size used by the LZ77 encoder (1 - 32768). Has value
1163 |    2048 by default, but can be set to 32768 for better, but slow, compression.
1164 | *) force_palette: if colortype is 2 or 6, you can make the encoder write a PLTE
1165 |    chunk if force_palette is true. This can used as suggested palette to convert
1166 |    to by viewers that don't support more than 256 colors (if those still exist)
1167 | *) add_id: add text chunk "Encoder: LodePNG <version>" to the image.
1168 | *) text_compression: default 1. If 1, it'll store texts as zTXt instead of tEXt chunks.
1169 |   zTXt chunks use zlib compression on the text. This gives a smaller result on
1170 |   large texts but a larger result on small texts (such as a single program name).
1171 |   It's all tEXt or all zTXt though, there's no separate setting per text yet.
1172 | 
1173 | 
1174 | 6. color conversions
1175 | --------------------
1176 | 
1177 | An important thing to note about LodePNG, is that the color type of the PNG, and
1178 | the color type of the raw image, are completely independent. By default, when
1179 | you decode a PNG, you get the result as a raw image in the color type you want,
1180 | no matter whether the PNG was encoded with a palette, greyscale or RGBA color.
1181 | And if you encode an image, by default LodePNG will automatically choose the PNG
1182 | color type that gives good compression based on the values of colors and amount
1183 | of colors in the image. It can be configured to let you control it instead as
1184 | well, though.
1185 | 
1186 | To be able to do this, LodePNG does conversions from one color mode to another.
1187 | It can convert from almost any color type to any other color type, except the
1188 | following conversions: RGB to greyscale is not supported, and converting to a
1189 | palette when the palette doesn't have a required color is not supported. This is
1190 | not supported on purpose: this is information loss which requires a color
1191 | reduction algorithm that is beyong the scope of a PNG encoder (yes, RGB to grey
1192 | is easy, but there are multiple ways if you want to give some channels more
1193 | weight).
1194 | 
1195 | By default, when decoding, you get the raw image in 32-bit RGBA or 24-bit RGB
1196 | color, no matter what color type the PNG has. And by default when encoding,
1197 | LodePNG automatically picks the best color model for the output PNG, and expects
1198 | the input image to be 32-bit RGBA or 24-bit RGB. So, unless you want to control
1199 | the color format of the images yourself, you can skip this chapter.
1200 | 
1201 | 6.1. PNG color types
1202 | --------------------
1203 | 
1204 | A PNG image can have many color types, ranging from 1-bit color to 64-bit color,
1205 | as well as palettized color modes. After the zlib decompression and unfiltering
1206 | in the PNG image is done, the raw pixel data will have that color type and thus
1207 | a certain amount of bits per pixel. If you want the output raw image after
1208 | decoding to have another color type, a conversion is done by LodePNG.
1209 | 
1210 | The PNG specification gives the following color types:
1211 | 
1212 | 0: greyscale, bit depths 1, 2, 4, 8, 16
1213 | 2: RGB, bit depths 8 and 16
1214 | 3: palette, bit depths 1, 2, 4 and 8
1215 | 4: greyscale with alpha, bit depths 8 and 16
1216 | 6: RGBA, bit depths 8 and 16
1217 | 
1218 | Bit depth is the amount of bits per pixel per color channel. So the total amount
1219 | of bits per pixel is: amount of channels * bitdepth.
1220 | 
1221 | 6.2. color conversions
1222 | ----------------------
1223 | 
1224 | As explained in the sections about the encoder and decoder, you can specify
1225 | color types and bit depths in info_png and info_raw to change the default
1226 | behaviour.
1227 | 
1228 | If, when decoding, you want the raw image to be something else than the default,
1229 | you need to set the color type and bit depth you want in the LodePNGColorMode,
1230 | or the parameters colortype and bitdepth of the simple decoding function.
1231 | 
1232 | If, when encoding, you use another color type than the default in the raw input
1233 | image, you need to specify its color type and bit depth in the LodePNGColorMode
1234 | of the raw image, or use the parameters colortype and bitdepth of the simple
1235 | encoding function.
1236 | 
1237 | If, when encoding, you don't want LodePNG to choose the output PNG color type
1238 | but control it yourself, you need to set auto_convert in the encoder settings
1239 | to false, and specify the color type you want in the LodePNGInfo of the
1240 | encoder (including palette: it can generate a palette if auto_convert is true,
1241 | otherwise not).
1242 | 
1243 | If the input and output color type differ (whether user chosen or auto chosen),
1244 | LodePNG will do a color conversion, which follows the rules below, and may
1245 | sometimes result in an error.
1246 | 
1247 | To avoid some confusion:
1248 | -the decoder converts from PNG to raw image
1249 | -the encoder converts from raw image to PNG
1250 | -the colortype and bitdepth in LodePNGColorMode info_raw, are those of the raw image
1251 | -the colortype and bitdepth in the color field of LodePNGInfo info_png, are those of the PNG
1252 | -when encoding, the color type in LodePNGInfo is ignored if auto_convert
1253 |  is enabled, it is automatically generated instead
1254 | -when decoding, the color type in LodePNGInfo is set by the decoder to that of the original
1255 |  PNG image, but it can be ignored since the raw image has the color type you requested instead
1256 | -if the color type of the LodePNGColorMode and PNG image aren't the same, a conversion
1257 |  between the color types is done if the color types are supported. If it is not
1258 |  supported, an error is returned. If the types are the same, no conversion is done.
1259 | -even though some conversions aren't supported, LodePNG supports loading PNGs from any
1260 |  colortype and saving PNGs to any colortype, sometimes it just requires preparing
1261 |  the raw image correctly before encoding.
1262 | -both encoder and decoder use the same color converter.
1263 | 
1264 | Non supported color conversions:
1265 | -color to greyscale: no error is thrown, but the result will look ugly because
1266 | only the red channel is taken
1267 | -anything to palette when that palette does not have that color in it: in this
1268 | case an error is thrown
1269 | 
1270 | Supported color conversions:
1271 | -anything to 8-bit RGB, 8-bit RGBA, 16-bit RGB, 16-bit RGBA
1272 | -any grey or grey+alpha, to grey or grey+alpha
1273 | -anything to a palette, as long as the palette has the requested colors in it
1274 | -removing alpha channel
1275 | -higher to smaller bitdepth, and vice versa
1276 | 
1277 | If you want no color conversion to be done (e.g. for speed or control):
1278 | -In the encoder, you can make it save a PNG with any color type by giving the
1279 | raw color mode and LodePNGInfo the same color mode, and setting auto_convert to
1280 | false.
1281 | -In the decoder, you can make it store the pixel data in the same color type
1282 | as the PNG has, by setting the color_convert setting to false. Settings in
1283 | info_raw are then ignored.
1284 | 
1285 | The function lodepng_convert does the color conversion. It is available in the
1286 | interface but normally isn't needed since the encoder and decoder already call
1287 | it.
1288 | 
1289 | 6.3. padding bits
1290 | -----------------
1291 | 
1292 | In the PNG file format, if a less than 8-bit per pixel color type is used and the scanlines
1293 | have a bit amount that isn't a multiple of 8, then padding bits are used so that each
1294 | scanline starts at a fresh byte. But that is NOT true for the LodePNG raw input and output.
1295 | The raw input image you give to the encoder, and the raw output image you get from the decoder
1296 | will NOT have these padding bits, e.g. in the case of a 1-bit image with a width
1297 | of 7 pixels, the first pixel of the second scanline will the the 8th bit of the first byte,
1298 | not the first bit of a new byte.
1299 | 
1300 | 6.4. A note about 16-bits per channel and endianness
1301 | ----------------------------------------------------
1302 | 
1303 | LodePNG uses unsigned char arrays for 16-bit per channel colors too, just like
1304 | for any other color format. The 16-bit values are stored in big endian (most
1305 | significant byte first) in these arrays. This is the opposite order of the
1306 | little endian used by x86 CPU's.
1307 | 
1308 | LodePNG always uses big endian because the PNG file format does so internally.
1309 | Conversions to other formats than PNG uses internally are not supported by
1310 | LodePNG on purpose, there are myriads of formats, including endianness of 16-bit
1311 | colors, the order in which you store R, G, B and A, and so on. Supporting and
1312 | converting to/from all that is outside the scope of LodePNG.
1313 | 
1314 | This may mean that, depending on your use case, you may want to convert the big
1315 | endian output of LodePNG to little endian with a for loop. This is certainly not
1316 | always needed, many applications and libraries support big endian 16-bit colors
1317 | anyway, but it means you cannot simply cast the unsigned char* buffer to an
1318 | unsigned short* buffer on x86 CPUs.
1319 | 
1320 | 
1321 | 7. error values
1322 | ---------------
1323 | 
1324 | All functions in LodePNG that return an error code, return 0 if everything went
1325 | OK, or a non-zero code if there was an error.
1326 | 
1327 | The meaning of the LodePNG error values can be retrieved with the function
1328 | lodepng_error_text: given the numerical error code, it returns a description
1329 | of the error in English as a string.
1330 | 
1331 | Check the implementation of lodepng_error_text to see the meaning of each code.
1332 | 
1333 | 
1334 | 8. chunks and PNG editing
1335 | -------------------------
1336 | 
1337 | If you want to add extra chunks to a PNG you encode, or use LodePNG for a PNG
1338 | editor that should follow the rules about handling of unknown chunks, or if your
1339 | program is able to read other types of chunks than the ones handled by LodePNG,
1340 | then that's possible with the chunk functions of LodePNG.
1341 | 
1342 | A PNG chunk has the following layout:
1343 | 
1344 | 4 bytes length
1345 | 4 bytes type name
1346 | length bytes data
1347 | 4 bytes CRC
1348 | 
1349 | 8.1. iterating through chunks
1350 | -----------------------------
1351 | 
1352 | If you have a buffer containing the PNG image data, then the first chunk (the
1353 | IHDR chunk) starts at byte number 8 of that buffer. The first 8 bytes are the
1354 | signature of the PNG and are not part of a chunk. But if you start at byte 8
1355 | then you have a chunk, and can check the following things of it.
1356 | 
1357 | NOTE: none of these functions check for memory buffer boundaries. To avoid
1358 | exploits, always make sure the buffer contains all the data of the chunks.
1359 | When using lodepng_chunk_next, make sure the returned value is within the
1360 | allocated memory.
1361 | 
1362 | unsigned lodepng_chunk_length(const unsigned char* chunk):
1363 | 
1364 | Get the length of the chunk's data. The total chunk length is this length + 12.
1365 | 
1366 | void lodepng_chunk_type(char type[5], const unsigned char* chunk):
1367 | unsigned char lodepng_chunk_type_equals(const unsigned char* chunk, const char* type):
1368 | 
1369 | Get the type of the chunk or compare if it's a certain type
1370 | 
1371 | unsigned char lodepng_chunk_critical(const unsigned char* chunk):
1372 | unsigned char lodepng_chunk_private(const unsigned char* chunk):
1373 | unsigned char lodepng_chunk_safetocopy(const unsigned char* chunk):
1374 | 
1375 | Check if the chunk is critical in the PNG standard (only IHDR, PLTE, IDAT and IEND are).
1376 | Check if the chunk is private (public chunks are part of the standard, private ones not).
1377 | Check if the chunk is safe to copy. If it's not, then, when modifying data in a critical
1378 | chunk, unsafe to copy chunks of the old image may NOT be saved in the new one if your
1379 | program doesn't handle that type of unknown chunk.
1380 | 
1381 | unsigned char* lodepng_chunk_data(unsigned char* chunk):
1382 | const unsigned char* lodepng_chunk_data_const(const unsigned char* chunk):
1383 | 
1384 | Get a pointer to the start of the data of the chunk.
1385 | 
1386 | unsigned lodepng_chunk_check_crc(const unsigned char* chunk):
1387 | void lodepng_chunk_generate_crc(unsigned char* chunk):
1388 | 
1389 | Check if the crc is correct or generate a correct one.
1390 | 
1391 | unsigned char* lodepng_chunk_next(unsigned char* chunk):
1392 | const unsigned char* lodepng_chunk_next_const(const unsigned char* chunk):
1393 | 
1394 | Iterate to the next chunk. This works if you have a buffer with consecutive chunks. Note that these
1395 | functions do no boundary checking of the allocated data whatsoever, so make sure there is enough
1396 | data available in the buffer to be able to go to the next chunk.
1397 | 
1398 | unsigned lodepng_chunk_append(unsigned char** out, size_t* outlength, const unsigned char* chunk):
1399 | unsigned lodepng_chunk_create(unsigned char** out, size_t* outlength, unsigned length,
1400 |                               const char* type, const unsigned char* data):
1401 | 
1402 | These functions are used to create new chunks that are appended to the data in *out that has
1403 | length *outlength. The append function appends an existing chunk to the new data. The create
1404 | function creates a new chunk with the given parameters and appends it. Type is the 4-letter
1405 | name of the chunk.
1406 | 
1407 | 8.2. chunks in info_png
1408 | -----------------------
1409 | 
1410 | The LodePNGInfo struct contains fields with the unknown chunk in it. It has 3
1411 | buffers (each with size) to contain 3 types of unknown chunks:
1412 | the ones that come before the PLTE chunk, the ones that come between the PLTE
1413 | and the IDAT chunks, and the ones that come after the IDAT chunks.
1414 | It's necessary to make the distionction between these 3 cases because the PNG
1415 | standard forces to keep the ordering of unknown chunks compared to the critical
1416 | chunks, but does not force any other ordering rules.
1417 | 
1418 | info_png.unknown_chunks_data[0] is the chunks before PLTE
1419 | info_png.unknown_chunks_data[1] is the chunks after PLTE, before IDAT
1420 | info_png.unknown_chunks_data[2] is the chunks after IDAT
1421 | 
1422 | The chunks in these 3 buffers can be iterated through and read by using the same
1423 | way described in the previous subchapter.
1424 | 
1425 | When using the decoder to decode a PNG, you can make it store all unknown chunks
1426 | if you set the option settings.remember_unknown_chunks to 1. By default, this
1427 | option is off (0).
1428 | 
1429 | The encoder will always encode unknown chunks that are stored in the info_png.
1430 | If you need it to add a particular chunk that isn't known by LodePNG, you can
1431 | use lodepng_chunk_append or lodepng_chunk_create to the chunk data in
1432 | info_png.unknown_chunks_data[x].
1433 | 
1434 | Chunks that are known by LodePNG should not be added in that way. E.g. to make
1435 | LodePNG add a bKGD chunk, set background_defined to true and add the correct
1436 | parameters there instead.
1437 | 
1438 | 
1439 | 9. compiler support
1440 | -------------------
1441 | 
1442 | No libraries other than the current standard C library are needed to compile
1443 | LodePNG. For the C++ version, only the standard C++ library is needed on top.
1444 | Add the files lodepng.c(pp) and lodepng.h to your project, include
1445 | lodepng.h where needed, and your program can read/write PNG files.
1446 | 
1447 | It is compatible with C90 and up, and C++03 and up.
1448 | 
1449 | If performance is important, use optimization when compiling! For both the
1450 | encoder and decoder, this makes a large difference.
1451 | 
1452 | Make sure that LodePNG is compiled with the same compiler of the same version
1453 | and with the same settings as the rest of the program, or the interfaces with
1454 | std::vectors and std::strings in C++ can be incompatible.
1455 | 
1456 | CHAR_BITS must be 8 or higher, because LodePNG uses unsigned chars for octets.
1457 | 
1458 | *) gcc and g++
1459 | 
1460 | LodePNG is developed in gcc so this compiler is natively supported. It gives no
1461 | warnings with compiler options "-Wall -Wextra -pedantic -ansi", with gcc and g++
1462 | version 4.7.1 on Linux, 32-bit and 64-bit.
1463 | 
1464 | *) Clang
1465 | 
1466 | Fully supported and warning-free.
1467 | 
1468 | *) Mingw
1469 | 
1470 | The Mingw compiler (a port of gcc for Windows) should be fully supported by
1471 | LodePNG.
1472 | 
1473 | *) Visual Studio and Visual C++ Express Edition
1474 | 
1475 | LodePNG should be warning-free with warning level W4. Two warnings were disabled
1476 | with pragmas though: warning 4244 about implicit conversions, and warning 4996
1477 | where it wants to use a non-standard function fopen_s instead of the standard C
1478 | fopen.
1479 | 
1480 | Visual Studio may want "stdafx.h" files to be included in each source file and
1481 | give an error "unexpected end of file while looking for precompiled header".
1482 | This is not standard C++ and will not be added to the stock LodePNG. You can
1483 | disable it for lodepng.cpp only by right clicking it, Properties, C/C++,
1484 | Precompiled Headers, and set it to Not Using Precompiled Headers there.
1485 | 
1486 | NOTE: Modern versions of VS should be fully supported, but old versions, e.g.
1487 | VS6, are not guaranteed to work.
1488 | 
1489 | *) Compilers on Macintosh
1490 | 
1491 | LodePNG has been reported to work both with gcc and LLVM for Macintosh, both for
1492 | C and C++.
1493 | 
1494 | *) Other Compilers
1495 | 
1496 | If you encounter problems on any compilers, feel free to let me know and I may
1497 | try to fix it if the compiler is modern and standards complient.
1498 | 
1499 | 
1500 | 10. examples
1501 | ------------
1502 | 
1503 | This decoder example shows the most basic usage of LodePNG. More complex
1504 | examples can be found on the LodePNG website.
1505 | 
1506 | 10.1. decoder C++ example
1507 | -------------------------
1508 | 
1509 | #include "lodepng.h"
1510 | #include <iostream>
1511 | 
1512 | int main(int argc, char *argv[])
1513 | {
1514 |   const char* filename = argc > 1 ? argv[1] : "test.png";
1515 | 
1516 |   //load and decode
1517 |   std::vector<unsigned char> image;
1518 |   unsigned width, height;
1519 |   unsigned error = lodepng::decode(image, width, height, filename);
1520 | 
1521 |   //if there's an error, display it
1522 |   if(error) std::cout << "decoder error " << error << ": " << lodepng_error_text(error) << std::endl;
1523 | 
1524 |   //the pixels are now in the vector "image", 4 bytes per pixel, ordered RGBARGBA..., use it as texture, draw it, ...
1525 | }
1526 | 
1527 | 10.2. decoder C example
1528 | -----------------------
1529 | 
1530 | #include "lodepng.h"
1531 | 
1532 | int main(int argc, char *argv[])
1533 | {
1534 |   unsigned error;
1535 |   unsigned char* image;
1536 |   size_t width, height;
1537 |   const char* filename = argc > 1 ? argv[1] : "test.png";
1538 | 
1539 |   error = lodepng_decode32_file(&image, &width, &height, filename);
1540 | 
1541 |   if(error) printf("decoder error %u: %s\n", error, lodepng_error_text(error));
1542 | 
1543 |   / * use image here * /
1544 | 
1545 |   free(image);
1546 |   return 0;
1547 | }
1548 | 
1549 | 
1550 | 11. changes
1551 | -----------
1552 | 
1553 | The version number of LodePNG is the date of the change given in the format
1554 | yyyymmdd.
1555 | 
1556 | Some changes aren't backwards compatible. Those are indicated with a (!)
1557 | symbol.
1558 | 
1559 | *) 28 jun 2014: Removed fix_png setting, always support palette OOB for
1560 |     simplicity. Made ColorProfile public.
1561 | *) 09 jun 2014: Faster encoder by fixing hash bug and more zeros optimization.
1562 | *) 22 dec 2013: Power of two windowsize required for optimization.
1563 | *) 15 apr 2013: Fixed bug with LAC_ALPHA and color key.
1564 | *) 25 mar 2013: Added an optional feature to ignore some PNG errors (fix_png).
1565 | *) 11 mar 2013 (!): Bugfix with custom free. Changed from "my" to "lodepng_"
1566 |     prefix for the custom allocators and made it possible with a new #define to
1567 |     use custom ones in your project without needing to change lodepng's code.
1568 | *) 28 jan 2013: Bugfix with color key.
1569 | *) 27 okt 2012: Tweaks in text chunk keyword length error handling.
1570 | *) 8 okt 2012 (!): Added new filter strategy (entropy) and new auto color mode.
1571 |     (no palette). Better deflate tree encoding. New compression tweak settings.
1572 |     Faster color conversions while decoding. Some internal cleanups.
1573 | *) 23 sep 2012: Reduced warnings in Visual Studio a little bit.
1574 | *) 1 sep 2012 (!): Removed #define's for giving custom (de)compression functions
1575 |     and made it work with function pointers instead.
1576 | *) 23 jun 2012: Added more filter strategies. Made it easier to use custom alloc
1577 |     and free functions and toggle #defines from compiler flags. Small fixes.
1578 | *) 6 may 2012 (!): Made plugging in custom zlib/deflate functions more flexible.
1579 | *) 22 apr 2012 (!): Made interface more consistent, renaming a lot. Removed
1580 |     redundant C++ codec classes. Reduced amount of structs. Everything changed,
1581 |     but it is cleaner now imho and functionality remains the same. Also fixed
1582 |     several bugs and shrinked the implementation code. Made new samples.
1583 | *) 6 nov 2011 (!): By default, the encoder now automatically chooses the best
1584 |     PNG color model and bit depth, based on the amount and type of colors of the
1585 |     raw image. For this, autoLeaveOutAlphaChannel replaced by auto_choose_color.
1586 | *) 9 okt 2011: simpler hash chain implementation for the encoder.
1587 | *) 8 sep 2011: lz77 encoder lazy matching instead of greedy matching.
1588 | *) 23 aug 2011: tweaked the zlib compression parameters after benchmarking.
1589 |     A bug with the PNG filtertype heuristic was fixed, so that it chooses much
1590 |     better ones (it's quite significant). A setting to do an experimental, slow,
1591 |     brute force search for PNG filter types is added.
1592 | *) 17 aug 2011 (!): changed some C zlib related function names.
1593 | *) 16 aug 2011: made the code less wide (max 120 characters per line).
1594 | *) 17 apr 2011: code cleanup. Bugfixes. Convert low to 16-bit per sample colors.
1595 | *) 21 feb 2011: fixed compiling for C90. Fixed compiling with sections disabled.
1596 | *) 11 dec 2010: encoding is made faster, based on suggestion by Peter Eastman
1597 |     to optimize long sequences of zeros.
1598 | *) 13 nov 2010: added LodePNG_InfoColor_hasPaletteAlpha and
1599 |     LodePNG_InfoColor_canHaveAlpha functions for convenience.
1600 | *) 7 nov 2010: added LodePNG_error_text function to get error code description.
1601 | *) 30 okt 2010: made decoding slightly faster
1602 | *) 26 okt 2010: (!) changed some C function and struct names (more consistent).
1603 |      Reorganized the documentation and the declaration order in the header.
1604 | *) 08 aug 2010: only changed some comments and external samples.
1605 | *) 05 jul 2010: fixed bug thanks to warnings in the new gcc version.
1606 | *) 14 mar 2010: fixed bug where too much memory was allocated for char buffers.
1607 | *) 02 sep 2008: fixed bug where it could create empty tree that linux apps could
1608 |     read by ignoring the problem but windows apps couldn't.
1609 | *) 06 jun 2008: added more error checks for out of memory cases.
1610 | *) 26 apr 2008: added a few more checks here and there to ensure more safety.
1611 | *) 06 mar 2008: crash with encoding of strings fixed
1612 | *) 02 feb 2008: support for international text chunks added (iTXt)
1613 | *) 23 jan 2008: small cleanups, and #defines to divide code in sections
1614 | *) 20 jan 2008: support for unknown chunks allowing using LodePNG for an editor.
1615 | *) 18 jan 2008: support for tIME and pHYs chunks added to encoder and decoder.
1616 | *) 17 jan 2008: ability to encode and decode compressed zTXt chunks added
1617 |     Also vareous fixes, such as in the deflate and the padding bits code.
1618 | *) 13 jan 2008: Added ability to encode Adam7-interlaced images. Improved
1619 |     filtering code of encoder.
1620 | *) 07 jan 2008: (!) changed LodePNG to use ISO C90 instead of C++. A
1621 |     C++ wrapper around this provides an interface almost identical to before.
1622 |     Having LodePNG be pure ISO C90 makes it more portable. The C and C++ code
1623 |     are together in these files but it works both for C and C++ compilers.
1624 | *) 29 dec 2007: (!) changed most integer types to unsigned int + other tweaks
1625 | *) 30 aug 2007: bug fixed which makes this Borland C++ compatible
1626 | *) 09 aug 2007: some VS2005 warnings removed again
1627 | *) 21 jul 2007: deflate code placed in new namespace separate from zlib code
1628 | *) 08 jun 2007: fixed bug with 2- and 4-bit color, and small interlaced images
1629 | *) 04 jun 2007: improved support for Visual Studio 2005: crash with accessing
1630 |     invalid std::vector element [0] fixed, and level 3 and 4 warnings removed
1631 | *) 02 jun 2007: made the encoder add a tag with version by default
1632 | *) 27 may 2007: zlib and png code separated (but still in the same file),
1633 |     simple encoder/decoder functions added for more simple usage cases
1634 | *) 19 may 2007: minor fixes, some code cleaning, new error added (error 69),
1635 |     moved some examples from here to lodepng_examples.cpp
1636 | *) 12 may 2007: palette decoding bug fixed
1637 | *) 24 apr 2007: changed the license from BSD to the zlib license
1638 | *) 11 mar 2007: very simple addition: ability to encode bKGD chunks.
1639 | *) 04 mar 2007: (!) tEXt chunk related fixes, and support for encoding
1640 |     palettized PNG images. Plus little interface change with palette and texts.
1641 | *) 03 mar 2007: Made it encode dynamic Huffman shorter with repeat codes.
1642 |     Fixed a bug where the end code of a block had length 0 in the Huffman tree.
1643 | *) 26 feb 2007: Huffman compression with dynamic trees (BTYPE 2) now implemented
1644 |     and supported by the encoder, resulting in smaller PNGs at the output.
1645 | *) 27 jan 2007: Made the Adler-32 test faster so that a timewaste is gone.
1646 | *) 24 jan 2007: gave encoder an error interface. Added color conversion from any
1647 |     greyscale type to 8-bit greyscale with or without alpha.
1648 | *) 21 jan 2007: (!) Totally changed the interface. It allows more color types
1649 |     to convert to and is more uniform. See the manual for how it works now.
1650 | *) 07 jan 2007: Some cleanup & fixes, and a few changes over the last days:
1651 |     encode/decode custom tEXt chunks, separate classes for zlib & deflate, and
1652 |     at last made the decoder give errors for incorrect Adler32 or Crc.
1653 | *) 01 jan 2007: Fixed bug with encoding PNGs with less than 8 bits per channel.
1654 | *) 29 dec 2006: Added support for encoding images without alpha channel, and
1655 |     cleaned out code as well as making certain parts faster.
1656 | *) 28 dec 2006: Added "Settings" to the encoder.
1657 | *) 26 dec 2006: The encoder now does LZ77 encoding and produces much smaller files now.
1658 |     Removed some code duplication in the decoder. Fixed little bug in an example.
1659 | *) 09 dec 2006: (!) Placed output parameters of public functions as first parameter.
1660 |     Fixed a bug of the decoder with 16-bit per color.
1661 | *) 15 okt 2006: Changed documentation structure
1662 | *) 09 okt 2006: Encoder class added. It encodes a valid PNG image from the
1663 |     given image buffer, however for now it's not compressed.
1664 | *) 08 sep 2006: (!) Changed to interface with a Decoder class
1665 | *) 30 jul 2006: (!) LodePNG_InfoPng , width and height are now retrieved in different
1666 |     way. Renamed decodePNG to decodePNGGeneric.
1667 | *) 29 jul 2006: (!) Changed the interface: image info is now returned as a
1668 |     struct of type LodePNG::LodePNG_Info, instead of a vector, which was a bit clumsy.
1669 | *) 28 jul 2006: Cleaned the code and added new error checks.
1670 |     Corrected terminology "deflate" into "inflate".
1671 | *) 23 jun 2006: Added SDL example in the documentation in the header, this
1672 |     example allows easy debugging by displaying the PNG and its transparency.
1673 | *) 22 jun 2006: (!) Changed way to obtain error value. Added
1674 |     loadFile function for convenience. Made decodePNG32 faster.
1675 | *) 21 jun 2006: (!) Changed type of info vector to unsigned.
1676 |     Changed position of palette in info vector. Fixed an important bug that
1677 |     happened on PNGs with an uncompressed block.
1678 | *) 16 jun 2006: Internally changed unsigned into unsigned where
1679 |     needed, and performed some optimizations.
1680 | *) 07 jun 2006: (!) Renamed functions to decodePNG and placed them
1681 |     in LodePNG namespace. Changed the order of the parameters. Rewrote the
1682 |     documentation in the header. Renamed files to lodepng.cpp and lodepng.h
1683 | *) 22 apr 2006: Optimized and improved some code
1684 | *) 07 sep 2005: (!) Changed to std::vector interface
1685 | *) 12 aug 2005: Initial release (C++, decoder only)
1686 | 
1687 | 
1688 | 12. contact information
1689 | -----------------------
1690 | 
1691 | Feel free to contact me with suggestions, problems, comments, ... concerning
1692 | LodePNG. If you encounter a PNG image that doesn't work properly with this
1693 | decoder, feel free to send it and I'll use it to find and fix the problem.
1694 | 
1695 | My email address is (puzzle the account and domain together with an @ symbol):
1696 | Domain: gmail dot com.
1697 | Account: lode dot vandevenne.
1698 | 
1699 | 
1700 | Copyright (c) 2005-2014 Lode Vandevenne
1701 | */
1702 | 


--------------------------------------------------------------------------------
/misc_utils/libahff/pixel.c:
--------------------------------------------------------------------------------
  1 | /*
  2 |  * pixel.c - Routines for converting common GL pixel formats to RGBA32.
  3 |  * Copyright (c) 2015 The Holy Constituency of the Summer Triangle.
  4 |  * All rights reserved.
  5 |  */
  6 | #include "rg_etc1wrap.h"
  7 | 
  8 | void copy_1bpp_luma(byte *raw, int len, byte *output) {
  9 |     memset(output, 255, len * 4);
 10 |     for (int i = 0, ctr = 0; i < len; ctr = (++i) * 4) {
 11 |         output[ctr] = raw[i];
 12 |         output[ctr + 1] = raw[i];
 13 |         output[ctr + 2] = raw[i];
 14 |     }
 15 | }
 16 | 
 17 | void copy_1bpp_alpha(byte *raw, int len, byte *output) {
 18 |     memset(output, 0, len * 4);
 19 |     for (int i = 0, ctr = 0; i < len; ctr = (++i) * 4) {
 20 |         output[ctr + 3] = raw[i];
 21 |     }
 22 | }
 23 | 
 24 | void copy_2bpp_lumalpha(byte *raw, int len, byte *output) {
 25 |     memset(output, 0, len * 2);
 26 |     for (int i = 0, ctr = 0; i < len; ctr = (i += 2) * 4) {
 27 |         output[ctr] = raw[i];
 28 |         output[ctr + 1] = raw[i];
 29 |         output[ctr + 2] = raw[i];
 30 |         output[ctr + 3] = raw[i + 1];
 31 |     }
 32 | }
 33 | 
 34 | void copy_2bpp_rgb565(byte *raw, int len, byte *output) {
 35 |     memset(output, 255, len * 2);
 36 |     unsigned short *pixels = (unsigned short *) raw;
 37 |     unsigned int len2 = len / 2;
 38 |     for (int i = 0, ctr = 0; i < len2; ctr = (++i) * 4) {
 39 |         unsigned short pixel = pixels[i];
 40 |         byte shift = (pixel & 0xF800) >> 8;
 41 |         output[ctr] = shift | (shift >> 5);
 42 |         shift = (pixel & 0x07E0) >> 3;
 43 |         output[ctr + 1] = shift | (shift >> 6);
 44 |         shift = (pixel & 0x001F) << 3;
 45 |         output[ctr + 2] = shift | (shift >> 5);
 46 |     }
 47 | }
 48 | 
 49 | void copy_2bpp_rgba5551(byte *raw, int len, byte *output) {
 50 |     unsigned short *pixels = (unsigned short *) raw;
 51 |     unsigned int len2 = len / 2;
 52 |     for (int i = 0, ctr = 0; i < len2; ctr = (++i) * 4) {
 53 |         unsigned short pixel = pixels[i];
 54 |         byte shift = (pixel & 0xF800) >> 8;
 55 |         output[ctr] = shift | (shift >> 5);
 56 |         shift = (pixel & 0x07C0) >> 3;
 57 |         output[ctr + 1] = shift | (shift >> 5);
 58 |         shift = (pixel & 0x003E) << 3;
 59 |         output[ctr + 2] = shift | (shift >> 5);
 60 |         output[ctr + 3] = (pixel % 2)? 255 : 0;
 61 |     }
 62 | }
 63 | 
 64 | void copy_2bpp_rgba4444(byte *raw, int len, byte *output) {
 65 |     unsigned short *pixels = (unsigned short *) raw;
 66 |     unsigned int len2 = len / 2;
 67 |     for (int i = 0, ctr = 0; i < len2; ctr = (++i) * 4) {
 68 |         unsigned short pixel = pixels[i];
 69 |         byte shift = (pixel & 0xF000) >> 8;
 70 |         output[ctr] = shift | (shift >> 4);
 71 |         shift = (pixel & 0x0F00) >> 4;
 72 |         output[ctr + 1] = shift | (shift >> 4);
 73 |         shift = pixel & 0x00F0;
 74 |         output[ctr + 2] = shift | (shift >> 4);
 75 |         shift = pixel & 0x000F;
 76 |         output[ctr + 3] = shift | (shift << 4);
 77 |     }
 78 | }
 79 | 
 80 | void copy_3bpp_rgb(byte *raw, int len, byte *output) {
 81 |     byte *in_pixel_ptr = raw;
 82 |     byte *out_pixel_ptr = output;
 83 | 
 84 |     while (in_pixel_ptr < raw + len) {
 85 |         *out_pixel_ptr++ = *in_pixel_ptr++;
 86 |         *out_pixel_ptr++ = *in_pixel_ptr++;
 87 |         *out_pixel_ptr++ = *in_pixel_ptr++;
 88 |         *out_pixel_ptr++ = 255;
 89 |     }
 90 | }
 91 | 
 92 | void copy_etc1_rgb(byte *raw, int len, byte *output, int width) {
 93 |     byte dst[4 * 4 * 4];
 94 |     byte *row1 = output,
 95 |          *row2 = row1 + (width * 4),
 96 |          *row3 = row2 + (width * 4),
 97 |          *row4 = row3 + (width * 4);
 98 |     int offset = 0;
 99 |     for (int i = 0; i < len; i += 8) {
100 |         assert(unpack_etc1_block_c(raw + i, dst, 0) == 1);
101 | 
102 |         memcpy(row1 + offset, dst     , 4 * 4);
103 |         memcpy(row2 + offset, dst + 16, 4 * 4);
104 |         memcpy(row3 + offset, dst + 32, 4 * 4);
105 |         memcpy(row4 + offset, dst + 48, 4 * 4);
106 | 
107 |         offset += 16;
108 | 
109 |         if (offset >= width * 4) {
110 |             row1 = row4 + (width * 4);
111 |             row2 = row1 + (width * 4);
112 |             row3 = row2 + (width * 4);
113 |             row4 = row3 + (width * 4);
114 |             offset = 0;
115 |         }
116 |     }
117 | }
118 | 


--------------------------------------------------------------------------------
/misc_utils/libahff/rg_etc1.h:
--------------------------------------------------------------------------------
 1 | // File: rg_etc1.h - Fast, high quality ETC1 block packer/unpacker - Rich Geldreich <richgel99@gmail.com>
 2 | // Please see ZLIB license at the end of this file.
 3 | #pragma once
 4 | 
 5 | namespace rg_etc1
 6 | {
 7 |    // Unpacks an 8-byte ETC1 compressed block to a block of 4x4 32bpp RGBA pixels.
 8 |    // Returns false if the block is invalid. Invalid blocks will still be unpacked with clamping.
 9 |    // This function is thread safe, and does not dynamically allocate any memory.
10 |    // If preserve_alpha is true, the alpha channel of the destination pixels will not be overwritten. Otherwise, alpha will be set to 255.
11 |    bool unpack_etc1_block(const void *pETC1_block, unsigned int* pDst_pixels_rgba, bool preserve_alpha = false);
12 | 
13 |    // Quality setting = the higher the quality, the slower. 
14 |    // To pack large textures, it is highly recommended to call pack_etc1_block() in parallel, on different blocks, from multiple threads (particularly when using cHighQuality).
15 |    enum etc1_quality
16 |    { 
17 |       cLowQuality,
18 |       cMediumQuality,
19 |       cHighQuality,
20 |    };
21 |       
22 |    struct etc1_pack_params
23 |    {
24 |       etc1_quality m_quality;
25 |       bool m_dithering;
26 |                               
27 |       inline etc1_pack_params() 
28 |       {
29 |          clear();
30 |       }
31 | 
32 |       void clear()
33 |       {
34 |          m_quality = cHighQuality;
35 |          m_dithering = false;
36 |       }
37 |    };
38 | 
39 |    // Important: pack_etc1_block_init() must be called before calling pack_etc1_block().
40 |    void pack_etc1_block_init();
41 | 
42 |    // Packs a 4x4 block of 32bpp RGBA pixels to an 8-byte ETC1 block.
43 |    // 32-bit RGBA pixels must always be arranged as (R,G,B,A) (R first, A last) in memory, independent of platform endianness. A should always be 255.
44 |    // Returns squared error of result.
45 |    // This function is thread safe, and does not dynamically allocate any memory.
46 |    // pack_etc1_block() does not currently support "perceptual" colorspace metrics - it primarily optimizes for RGB RMSE.
47 |    unsigned int pack_etc1_block(void* pETC1_block, const unsigned int* pSrc_pixels_rgba, etc1_pack_params& pack_params);
48 |             
49 | } // namespace rg_etc1
50 | 
51 | //------------------------------------------------------------------------------
52 | //
53 | // rg_etc1 uses the ZLIB license:
54 | // http://opensource.org/licenses/Zlib
55 | //
56 | // Copyright (c) 2012 Rich Geldreich
57 | //
58 | // This software is provided 'as-is', without any express or implied
59 | // warranty.  In no event will the authors be held liable for any damages
60 | // arising from the use of this software.
61 | //
62 | // Permission is granted to anyone to use this software for any purpose,
63 | // including commercial applications, and to alter it and redistribute it
64 | // freely, subject to the following restrictions:
65 | //
66 | // 1. The origin of this software must not be misrepresented; you must not
67 | // claim that you wrote the original software. If you use this software
68 | // in a product, an acknowledgment in the product documentation would be
69 | // appreciated but is not required.
70 | //
71 | // 2. Altered source versions must be plainly marked as such, and must not be
72 | // misrepresented as being the original software.
73 | //
74 | // 3. This notice may not be removed or altered from any source distribution.
75 | //
76 | //------------------------------------------------------------------------------
77 | 


--------------------------------------------------------------------------------
/misc_utils/libahff/rg_etc1wrap.cpp:
--------------------------------------------------------------------------------
 1 | #include "rg_etc1.h"
 2 | 
 3 | extern "C" {
 4 | 
 5 | int unpack_etc1_block_c(const unsigned char *pETC1_block, unsigned char *pDst_pixels_rgba, int preserve_alpha) {
 6 |     return rg_etc1::unpack_etc1_block(pETC1_block, (unsigned int *)pDst_pixels_rgba, preserve_alpha) ? 1 : 0;
 7 | }
 8 | 
 9 | }
10 | 


--------------------------------------------------------------------------------
/misc_utils/libahff/rg_etc1wrap.h:
--------------------------------------------------------------------------------
1 | int unpack_etc1_block_c(const unsigned char *pETC1_block, unsigned char *pDst_pixels_rgba, int preserve_alpha);
2 | 


--------------------------------------------------------------------------------
/misc_utils/linux/libahff.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/summertriangle-dev/starlight_sync/3efd64b6eb84afadea13e33177f8299d96d9a72e/misc_utils/linux/libahff.dylib


--------------------------------------------------------------------------------
/misc_utils/linux/make_icon_sheet:
--------------------------------------------------------------------------------
1 | #!/bin/bash
2 | OUTPUT=$1
3 | IMAGES=${@:2}
4 | 
5 | montage -background transparent \
6 |         -geometry 48x48+0+0 -tile 8x8 -depth 8 $IMAGES $OUTPUT
7 | 


--------------------------------------------------------------------------------
/misc_utils/linux/make_icon_sheet2:
--------------------------------------------------------------------------------
1 | #!/bin/bash
2 | OUTPUT=$1
3 | IMAGES=${@:2}
4 | 
5 | montage -background transparent \
6 |         -geometry 96x96+0+0 -tile 8x8 -quality 90 $IMAGES $OUTPUT
7 | 


--------------------------------------------------------------------------------
/misc_utils/linux/scan_dec_hca:
--------------------------------------------------------------------------------
 1 | #!/bin/bash
 2 | 
 3 | SCRIPT_PATH=`dirname "$0"`; SCRIPT_PATH=`eval "cd \"$SCRIPT_PATH\" && pwd"`
 4 | HCA_DECODER="$SCRIPT_PATH/hca_decoder"
 5 | LAME="$SCRIPT_PATH/lame"
 6 | 
 7 | cd "$1"
 8 | for f in *.hca; do
 9 |     ( $HCA_DECODER -o /dev/stdout $f | $LAME -V2 - "${f%.hca}".mp3 ) && rm $f
10 | done
11 | 


--------------------------------------------------------------------------------
/misc_utils/macosx/libahff.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/summertriangle-dev/starlight_sync/3efd64b6eb84afadea13e33177f8299d96d9a72e/misc_utils/macosx/libahff.dylib


--------------------------------------------------------------------------------
/misc_utils/macosx/make_icon_sheet:
--------------------------------------------------------------------------------
1 | #!/bin/bash
2 | OUTPUT=$1
3 | IMAGES=${@:2}
4 | 
5 | montage -background transparent \
6 |         -geometry 48x48+0+0 -tile 8x8 -depth 8 $IMAGES $OUTPUT
7 | 


--------------------------------------------------------------------------------
/misc_utils/macosx/make_icon_sheet2:
--------------------------------------------------------------------------------
1 | #!/bin/bash
2 | OUTPUT=$1
3 | IMAGES=${@:2}
4 | 
5 | montage -background transparent \
6 |         -geometry 96x96+0+0 -tile 8x8 -quality 90 $IMAGES $OUTPUT
7 | 


--------------------------------------------------------------------------------
/misc_utils/macosx/scan_dec_hca:
--------------------------------------------------------------------------------
 1 | #!/bin/bash
 2 | 
 3 | SCRIPT_PATH=`dirname "$0"`; SCRIPT_PATH=`eval "cd \"$SCRIPT_PATH\" && pwd"`
 4 | HCA_DECODER="$SCRIPT_PATH/hca_decoder"
 5 | LAME="$SCRIPT_PATH/lame"
 6 | 
 7 | cd "$1"
 8 | for f in *.hca; do
 9 |     ( $HCA_DECODER -o /dev/stdout $f | $LAME -V2 - "${f%.hca}".mp3 ) && rm $f
10 | done
11 | 


--------------------------------------------------------------------------------
/requirements.txt:
--------------------------------------------------------------------------------
1 | lz4==0.7.0
2 | msgpack-python==0.4.7
3 | plac==0.9.1
4 | requests==2.9.1
5 | tornado==4.3
6 | 


--------------------------------------------------------------------------------
/rijndael.py:
--------------------------------------------------------------------------------
  1 | # Authors:
  2 | #   Bram Cohen - main author
  3 | #   Trevor Perrin - minor changes for Python compatibility
  4 | #
  5 | # See the LICENSE file for legal information regarding use of this file.
  6 | # Also see Bram Cohen's statement below
  7 | 
  8 | ################ RIJNDAEL ###
  9 | # Trevor edited below code for python 3, and ease of packaging
 10 | 
 11 | """
 12 | A pure python (slow) implementation of rijndael with a decent interface
 13 | 
 14 | To include -
 15 | 
 16 | from rijndael import rijndael
 17 | 
 18 | To do a key setup -
 19 | 
 20 | r = rijndael(key, block_size = 16)
 21 | 
 22 | key must be a string of length 16, 24, or 32
 23 | blocksize must be 16, 24, or 32. Default is 16
 24 | 
 25 | To use -
 26 | 
 27 | ciphertext = r.encrypt(plaintext)
 28 | plaintext = r.decrypt(ciphertext)
 29 | 
 30 | If any strings are of the wrong length a ValueError is thrown
 31 | """
 32 | 
 33 | # ported from the Java reference code by Bram Cohen, bram@gawth.com, April 2001
 34 | # this code is public domain, unless someone makes
 35 | # an intellectual property claim against the reference
 36 | # code, in which case it can be made public domain by
 37 | # deleting all the comments and renaming all the variables
 38 | 
 39 | import copy
 40 | 
 41 | 
 42 | 
 43 | #TREV 2011 - is this still needed? seems not
 44 | #-----------------------
 45 | #TREV - ADDED BECAUSE THERE'S WARNINGS ABOUT INT OVERFLOW BEHAVIOR CHANGING IN
 46 | #2.4.....
 47 | #import os
 48 | #if os.name != "java":
 49 | #    import exceptions
 50 | #    if hasattr(exceptions, "FutureWarning"):
 51 | #        import warnings
 52 | #        warnings.filterwarnings("ignore", category=FutureWarning, append=1)
 53 | #-----------------------
 54 | 
 55 | 
 56 | 
 57 | shifts = [[[0, 0], [1, 3], [2, 2], [3, 1]],
 58 |           [[0, 0], [1, 5], [2, 4], [3, 3]],
 59 |           [[0, 0], [1, 7], [3, 5], [4, 4]]]
 60 | 
 61 | # [keysize][block_size]
 62 | num_rounds = {16: {16: 10, 24: 12, 32: 14}, 
 63 | 24: {16: 12, 24: 12, 32: 14}, 32: {16: 14, 24: 14, 32: 14}}
 64 | 
 65 | A = [[1, 1, 1, 1, 1, 0, 0, 0],
 66 |      [0, 1, 1, 1, 1, 1, 0, 0],
 67 |      [0, 0, 1, 1, 1, 1, 1, 0],
 68 |      [0, 0, 0, 1, 1, 1, 1, 1],
 69 |      [1, 0, 0, 0, 1, 1, 1, 1],
 70 |      [1, 1, 0, 0, 0, 1, 1, 1],
 71 |      [1, 1, 1, 0, 0, 0, 1, 1],
 72 |      [1, 1, 1, 1, 0, 0, 0, 1]]
 73 | 
 74 | # produce log and alog tables, needed for multiplying in the
 75 | # field GF(2^m) (generator = 3)
 76 | alog = [1]
 77 | for i in range(255):
 78 |     j = (alog[-1] << 1) ^ alog[-1]
 79 |     if j & 0x100 != 0:
 80 |         j ^= 0x11B
 81 |     alog.append(j)
 82 | 
 83 | log = [0] * 256
 84 | for i in range(1, 255):
 85 |     log[alog[i]] = i
 86 | 
 87 | # multiply two elements of GF(2^m)
 88 | def mul(a, b):
 89 |     if a == 0 or b == 0:
 90 |         return 0
 91 |     return alog[(log[a & 0xFF] + log[b & 0xFF]) % 255]
 92 | 
 93 | # substitution box based on F^{-1}(x)
 94 | box = [[0] * 8 for i in range(256)]
 95 | box[1][7] = 1
 96 | for i in range(2, 256):
 97 |     j = alog[255 - log[i]]
 98 |     for t in range(8):
 99 |         box[i][t] = (j >> (7 - t)) & 0x01
100 | 
101 | B = [0, 1, 1, 0, 0, 0, 1, 1]
102 | 
103 | # affine transform:  box[i] <- B + A*box[i]
104 | cox = [[0] * 8 for i in range(256)]
105 | for i in range(256):
106 |     for t in range(8):
107 |         cox[i][t] = B[t]
108 |         for j in range(8):
109 |             cox[i][t] ^= A[t][j] * box[i][j]
110 | 
111 | # S-boxes and inverse S-boxes
112 | S =  [0] * 256
113 | Si = [0] * 256
114 | for i in range(256):
115 |     S[i] = cox[i][0] << 7
116 |     for t in range(1, 8):
117 |         S[i] ^= cox[i][t] << (7-t)
118 |     Si[S[i] & 0xFF] = i
119 | 
120 | # T-boxes
121 | G = [[2, 1, 1, 3],
122 |     [3, 2, 1, 1],
123 |     [1, 3, 2, 1],
124 |     [1, 1, 3, 2]]
125 | 
126 | AA = [[0] * 8 for i in range(4)]
127 | 
128 | for i in range(4):
129 |     for j in range(4):
130 |         AA[i][j] = G[i][j]
131 |         AA[i][i+4] = 1
132 | 
133 | for i in range(4):
134 |     pivot = AA[i][i]
135 |     if pivot == 0:
136 |         t = i + 1
137 |         while AA[t][i] == 0 and t < 4:
138 |             t += 1
139 |             assert t != 4, 'G matrix must be invertible'
140 |             for j in range(8):
141 |                 AA[i][j], AA[t][j] = AA[t][j], AA[i][j]
142 |             pivot = AA[i][i]
143 |     for j in range(8):
144 |         if AA[i][j] != 0:
145 |             AA[i][j] = alog[(255 + log[AA[i][j] & 0xFF] - log[pivot & 0xFF]) % 255]
146 |     for t in range(4):
147 |         if i != t:
148 |             for j in range(i+1, 8):
149 |                 AA[t][j] ^= mul(AA[i][j], AA[t][i])
150 |             AA[t][i] = 0
151 | 
152 | iG = [[0] * 4 for i in range(4)]
153 | 
154 | for i in range(4):
155 |     for j in range(4):
156 |         iG[i][j] = AA[i][j + 4]
157 | 
158 | def mul4(a, bs):
159 |     if a == 0:
160 |         return 0
161 |     r = 0
162 |     for b in bs:
163 |         r <<= 8
164 |         if b != 0:
165 |             r = r | mul(a, b)
166 |     return r
167 | 
168 | T1 = []
169 | T2 = []
170 | T3 = []
171 | T4 = []
172 | T5 = []
173 | T6 = []
174 | T7 = []
175 | T8 = []
176 | U1 = []
177 | U2 = []
178 | U3 = []
179 | U4 = []
180 | 
181 | for t in range(256):
182 |     s = S[t]
183 |     T1.append(mul4(s, G[0]))
184 |     T2.append(mul4(s, G[1]))
185 |     T3.append(mul4(s, G[2]))
186 |     T4.append(mul4(s, G[3]))
187 | 
188 |     s = Si[t]
189 |     T5.append(mul4(s, iG[0]))
190 |     T6.append(mul4(s, iG[1]))
191 |     T7.append(mul4(s, iG[2]))
192 |     T8.append(mul4(s, iG[3]))
193 | 
194 |     U1.append(mul4(t, iG[0]))
195 |     U2.append(mul4(t, iG[1]))
196 |     U3.append(mul4(t, iG[2]))
197 |     U4.append(mul4(t, iG[3]))
198 | 
199 | # round constants
200 | rcon = [1]
201 | r = 1
202 | for t in range(1, 30):
203 |     r = mul(2, r)
204 |     rcon.append(r)
205 | 
206 | del A
207 | del AA
208 | del pivot
209 | del B
210 | del G
211 | del box
212 | del log
213 | del alog
214 | del i
215 | del j
216 | del r
217 | del s
218 | del t
219 | del mul
220 | del mul4
221 | del cox
222 | del iG
223 | 
224 | class rijndael:
225 |     def __init__(self, key, block_size = 16):
226 |         if block_size != 16 and block_size != 24 and block_size != 32:
227 |             raise ValueError('Invalid block size: ' + str(block_size))
228 |         if len(key) != 16 and len(key) != 24 and len(key) != 32:
229 |             raise ValueError('Invalid key size: ' + str(len(key)))
230 |         self.block_size = block_size
231 | 
232 |         ROUNDS = num_rounds[len(key)][block_size]
233 |         BC = block_size // 4
234 |         # encryption round keys
235 |         Ke = [[0] * BC for i in range(ROUNDS + 1)]
236 |         # decryption round keys
237 |         Kd = [[0] * BC for i in range(ROUNDS + 1)]
238 |         ROUND_KEY_COUNT = (ROUNDS + 1) * BC
239 |         KC = len(key) // 4
240 | 
241 |         # copy user material bytes into temporary ints
242 |         tk = []
243 |         for i in range(0, KC):
244 |             tk.append((key[i * 4] << 24) | (key[i * 4 + 1] << 16) |
245 |                 (key[i * 4 + 2]) << 8 | key[i * 4 + 3])
246 | 
247 |         # copy values into round key arrays
248 |         t = 0
249 |         j = 0
250 |         while j < KC and t < ROUND_KEY_COUNT:
251 |             Ke[t // BC][t % BC] = tk[j]
252 |             Kd[ROUNDS - (t // BC)][t % BC] = tk[j]
253 |             j += 1
254 |             t += 1
255 |         tt = 0
256 |         rconpointer = 0
257 |         while t < ROUND_KEY_COUNT:
258 |             # extrapolate using phi (the round key evolution function)
259 |             tt = tk[KC - 1]
260 |             tk[0] ^= (S[(tt >> 16) & 0xFF] & 0xFF) << 24 ^  \
261 |                      (S[(tt >>  8) & 0xFF] & 0xFF) << 16 ^  \
262 |                      (S[ tt        & 0xFF] & 0xFF) <<  8 ^  \
263 |                      (S[(tt >> 24) & 0xFF] & 0xFF)       ^  \
264 |                      (rcon[rconpointer]    & 0xFF) << 24
265 |             rconpointer += 1
266 |             if KC != 8:
267 |                 for i in range(1, KC):
268 |                     tk[i] ^= tk[i-1]
269 |             else:
270 |                 for i in range(1, KC // 2):
271 |                     tk[i] ^= tk[i-1]
272 |                 tt = tk[KC // 2 - 1]
273 |                 tk[KC // 2] ^= (S[ tt        & 0xFF] & 0xFF)       ^ \
274 |                               (S[(tt >>  8) & 0xFF] & 0xFF) <<  8 ^ \
275 |                               (S[(tt >> 16) & 0xFF] & 0xFF) << 16 ^ \
276 |                               (S[(tt >> 24) & 0xFF] & 0xFF) << 24
277 |                 for i in range(KC // 2 + 1, KC):
278 |                     tk[i] ^= tk[i-1]
279 |             # copy values into round key arrays
280 |             j = 0
281 |             while j < KC and t < ROUND_KEY_COUNT:
282 |                 Ke[t // BC][t % BC] = tk[j]
283 |                 Kd[ROUNDS - (t // BC)][t % BC] = tk[j]
284 |                 j += 1
285 |                 t += 1
286 |         # inverse MixColumn where needed
287 |         for r in range(1, ROUNDS):
288 |             for j in range(BC):
289 |                 tt = Kd[r][j]
290 |                 Kd[r][j] = U1[(tt >> 24) & 0xFF] ^ \
291 |                            U2[(tt >> 16) & 0xFF] ^ \
292 |                            U3[(tt >>  8) & 0xFF] ^ \
293 |                            U4[ tt        & 0xFF]
294 |         self.Ke = Ke
295 |         self.Kd = Kd
296 | 
297 |     def encrypt(self, plaintext):
298 |         if len(plaintext) != self.block_size:
299 |             raise ValueError('wrong block length, expected ' + 
300 |                 str(self.block_size) + ' got ' + str(len(plaintext)))
301 |         Ke = self.Ke
302 | 
303 |         BC = self.block_size // 4
304 |         ROUNDS = len(Ke) - 1
305 |         if BC == 4:
306 |             SC = 0
307 |         elif BC == 6:
308 |             SC = 1
309 |         else:
310 |             SC = 2
311 |         s1 = shifts[SC][1][0]
312 |         s2 = shifts[SC][2][0]
313 |         s3 = shifts[SC][3][0]
314 |         a = [0] * BC
315 |         # temporary work array
316 |         t = []
317 |         # plaintext to ints + key
318 |         for i in range(BC):
319 |             t.append((plaintext[i * 4    ] << 24 |
320 |                       plaintext[i * 4 + 1] << 16 |
321 |                       plaintext[i * 4 + 2] <<  8 |
322 |                       plaintext[i * 4 + 3]        ) ^ Ke[0][i])
323 |         # apply round transforms
324 |         for r in range(1, ROUNDS):
325 |             for i in range(BC):
326 |                 a[i] = (T1[(t[ i           ] >> 24) & 0xFF] ^
327 |                         T2[(t[(i + s1) % BC] >> 16) & 0xFF] ^
328 |                         T3[(t[(i + s2) % BC] >>  8) & 0xFF] ^
329 |                         T4[ t[(i + s3) % BC]        & 0xFF]  ) ^ Ke[r][i]
330 |             t = copy.copy(a)
331 |         # last round is special
332 |         result = []
333 |         for i in range(BC):
334 |             tt = Ke[ROUNDS][i]
335 |             result.append((S[(t[ i           ] >> 24) & 0xFF] ^ (tt >> 24)) & 0xFF)
336 |             result.append((S[(t[(i + s1) % BC] >> 16) & 0xFF] ^ (tt >> 16)) & 0xFF)
337 |             result.append((S[(t[(i + s2) % BC] >>  8) & 0xFF] ^ (tt >>  8)) & 0xFF)
338 |             result.append((S[ t[(i + s3) % BC]        & 0xFF] ^  tt       ) & 0xFF)
339 |         return bytearray(result)
340 | 
341 |     def decrypt(self, ciphertext):
342 |         if len(ciphertext) != self.block_size:
343 |             raise ValueError('wrong block length, expected ' + 
344 |                 str(self.block_size) + ' got ' + str(len(ciphertext)))
345 |         Kd = self.Kd
346 | 
347 |         BC = self.block_size // 4
348 |         ROUNDS = len(Kd) - 1
349 |         if BC == 4:
350 |             SC = 0
351 |         elif BC == 6:
352 |             SC = 1
353 |         else:
354 |             SC = 2
355 |         s1 = shifts[SC][1][1]
356 |         s2 = shifts[SC][2][1]
357 |         s3 = shifts[SC][3][1]
358 |         a = [0] * BC
359 |         # temporary work array
360 |         t = [0] * BC
361 |         # ciphertext to ints + key
362 |         for i in range(BC):
363 |             t[i] = (ciphertext[i * 4    ] << 24 |
364 |                     ciphertext[i * 4 + 1] << 16 |
365 |                     ciphertext[i * 4 + 2] <<  8 |
366 |                     ciphertext[i * 4 + 3]        ) ^ Kd[0][i]
367 |         # apply round transforms
368 |         for r in range(1, ROUNDS):
369 |             for i in range(BC):
370 |                 a[i] = (T5[(t[ i           ] >> 24) & 0xFF] ^
371 |                         T6[(t[(i + s1) % BC] >> 16) & 0xFF] ^
372 |                         T7[(t[(i + s2) % BC] >>  8) & 0xFF] ^
373 |                         T8[ t[(i + s3) % BC]        & 0xFF]  ) ^ Kd[r][i]
374 |             t = copy.copy(a)
375 |         # last round is special
376 |         result = []
377 |         for i in range(BC):
378 |             tt = Kd[ROUNDS][i]
379 |             result.append((Si[(t[ i           ] >> 24) & 0xFF] ^ (tt >> 24)) & 0xFF)
380 |             result.append((Si[(t[(i + s1) % BC] >> 16) & 0xFF] ^ (tt >> 16)) & 0xFF)
381 |             result.append((Si[(t[(i + s2) % BC] >>  8) & 0xFF] ^ (tt >>  8)) & 0xFF)
382 |             result.append((Si[ t[(i + s3) % BC]        & 0xFF] ^  tt       ) & 0xFF)
383 |         return bytearray(result)
384 | 
385 | def encrypt(key, block):
386 |     return rijndael(key, len(block)).encrypt(block)
387 | 
388 | def decrypt(key, block):
389 |     return rijndael(key, len(block)).decrypt(block)
390 | 


--------------------------------------------------------------------------------
/sbjk.py:
--------------------------------------------------------------------------------
  1 | #!/usr/bin/env python3
  2 | import requests
  3 | import os
  4 | import sqlite3
  5 | import lz4
  6 | import io
  7 | import sys
  8 | import acb
  9 | import re
 10 | import fnmatch
 11 | import functools
 12 | import subprocess
 13 | import assetbundle
 14 | import ctypes
 15 | from collections import namedtuple, OrderedDict
 16 | 
 17 | SESSION = requests.Session()
 18 | 
 19 | # yes, it is still .dylib even on linux
 20 | libahff = ctypes.cdll.LoadLibrary(
 21 |     os.path.join(os.path.dirname(__file__), "misc_utils", os.getenv("PLATFORM"), "libahff.dylib"))
 22 | libahff.ahff_encode_texdata.argtypes = [
 23 |     ctypes.c_int,     # int fmt,
 24 |     ctypes.c_int,     # int width,
 25 |     ctypes.c_int,     # int height,
 26 |     ctypes.c_size_t,  # size_t len,
 27 |     ctypes.c_char_p,  # const uint8_t *data,
 28 |     ctypes.c_char_p]  # const char *out_path
 29 | libahff.ahff_encode_texdata.restype = ctypes.c_int
 30 | 
 31 | def invoke_external_util(util, *args):
 32 |     util = os.path.join(os.path.dirname(__file__), "misc_utils", os.getenv("PLATFORM"), util)
 33 |     try:
 34 |         output = subprocess.check_output((util,) + args, stderr=subprocess.STDOUT)
 35 |     except subprocess.CalledProcessError as inv:
 36 |         print("invoke_external_util: util exited abnormally", util, *args)
 37 |         sys.stdout.buffer.write(inv.output)
 38 | 
 39 | def get_resource(url, asset_name, flags):
 40 |     resp = SESSION.get(url)
 41 |     resp.raise_for_status()
 42 | 
 43 |     buf = resp.content
 44 |     if flags & 1:
 45 |         bio = io.BytesIO()
 46 |         bio.write(buf[4:8])
 47 |         bio.write(buf[16:])
 48 |         buf = lz4.loads(bio.getvalue())
 49 | 
 50 |     return buf
 51 | 
 52 | def noop(url, asset_name, flags):
 53 |     print("noop", asset_name)
 54 | 
 55 | def audio(object_id, use, index):
 56 |     a = (object_id << 40) | ((use & 0xFF) << 24) | ((index & 0xFF) << 16) | 0x11AB
 57 |     # make everything 8 bytes long for reasons
 58 |     a &= 0xFFFFFFFFFFFFFFFF
 59 |     a ^= 0x1042FC1040200700
 60 |     basename = hex(a)[2:]
 61 |     return basename
 62 | 
 63 | def real_extract_image_to(target, file_id_regex, file_format, only_match, url, asset_name, flags):
 64 |     buf = get_resource(url, asset_name, flags)
 65 |     tf = file_format.format(*file_id_regex.findall(asset_name))
 66 |     only_match = only_match or b"*"
 67 | 
 68 |     class fakefd(object):
 69 |         def __init__(self, stream):
 70 |             self.stream = stream
 71 |         def read(self):
 72 |             return self.stream
 73 | 
 74 |     bundle_objects = assetbundle.Asset(fakefd(buf))
 75 |     for obj in bundle_objects.objs:
 76 |         if "image data" not in obj:
 77 |             continue
 78 | 
 79 |         if fnmatch.fnmatch(obj["m_Name"], only_match):
 80 |             tf = os.path.join(target, tf)
 81 |             try:
 82 |                 os.makedirs(os.path.dirname(tf), 0o755)
 83 |             except FileExistsError:
 84 |                 pass
 85 | 
 86 |             data = obj["image data"]
 87 |             width, height, fmt = obj["m_Width"], obj["m_Height"], obj["m_TextureFormat"]
 88 |             libahff.ahff_encode_texdata(fmt, width, height, len(data), data, tf.encode("utf8"))
 89 | 
 90 | def extract_image_to(target, file_id_regex, file_format, only_match=None):
 91 |     target_dir = os.path.join(os.getenv("WORKING_DIR"), "root", target)
 92 |     try:
 93 |         os.makedirs(target_dir, 0o755)
 94 |     except FileExistsError:
 95 |         pass
 96 |     return functools.partial(real_extract_image_to, target_dir, re.compile(file_id_regex), file_format, only_match)
 97 | 
 98 | def process_acb(url, asset_name, flags):
 99 |     target_dir = os.path.join(os.getenv("WORKING_DIR"), "root", "va2")
100 |     try:
101 |         os.makedirs(target_dir, 0o755)
102 |     except FileExistsError:
103 |         pass
104 | 
105 |     buf = get_resource(url, asset_name, flags)
106 | 
107 |     acb_file = io.BytesIO(buf)
108 |     acb_file.seek(0)
109 | 
110 |     utf = acb.UTFTable(acb_file)
111 |     cue = acb.TrackList(utf)
112 |     embedded_awb = io.BytesIO(utf.rows[0]["AwbFile"])
113 |     data_source = acb.AFSArchive(embedded_awb)
114 | 
115 |     for track in cue.tracks:
116 |         print("queueing", track)
117 |         params = [int(x) for x in re.findall("0*[0-9]+", track.name)]
118 |         if len(params) != 3:
119 |             continue
120 | 
121 |         base = audio(*params[:3])
122 |         name = "{0}{1}".format(base, acb.wave_type_ftable.get(track.enc_type, track.enc_type))
123 |         with open(os.path.join(target_dir, name), "wb") as named_out_file:
124 |             named_out_file.write(data_source.file_data_for_cue_id(track.cue_id))
125 | 
126 |     invoke_external_util("scan_dec_hca", target_dir)
127 | 
128 | def do_action_for_file(asset_name, hash, flags):
129 |     if asset_name.endswith(".unity3d"):
130 |         url = ASSETBBASEURL.format(hash)
131 |     if asset_name.endswith(".acb"):
132 |         url = SOUNDBASEURL.format(hash, os.path.dirname(asset_name))
133 |     if asset_name.endswith(".mdb"):
134 |         url = SQLBASEURL.format(hash)
135 | 
136 |     for key in ACTIONS:
137 |         if fnmatch.fnmatch(asset_name, key):
138 |             ACTIONS[key](url, asset_name, flags)
139 |             break
140 | 
141 | 
142 | 
143 | ACTIONS_FAST = OrderedDict([
144 |     ("card_bg_*_*.unity3d",         noop),
145 |     ("card_gacha_*_*_sign.unity3d", extract_image_to("sign",           r"(?:0+)?([0-9]+)", "{0}.png")),
146 |     ("card_petit_*.unity3d",        extract_image_to("puchi",          r"(?:0+)?([0-9]+)", "{0}.png")),
147 |     ("card_bg_*.unity3d",           extract_image_to("spread",         r"(?:0+)?([0-9]+)", "{0}.png")),
148 |     ("card_*_xl.unity3d",           extract_image_to("card",           r"(?:0+)?([0-9]+)", "{0}.png")),
149 |     ("card_*_sm.unity3d",           extract_image_to("../iconcache",   r"(?:0+)?([0-9]+)", "{0}.png", b"*_m")),
150 |     ("chara_icon_*.unity3d",        extract_image_to("icon_char",      r"(?:0+)?([0-9]+)", "{0}.png")),
151 | ])
152 | 
153 | ACTIONS_SLOW = OrderedDict([
154 |     ("chara_*_base.unity3d",        extract_image_to("chara2",         r"(?:0+)?([0-9]+)", "{0}/{1}.png")),
155 |     ("chara_*_face_*.unity3d",      extract_image_to("chara2",         r"(?:0+)?([0-9]+)", "{0}/{1}_{2}.png")),
156 |     ("v/*.acb",                     process_acb),
157 | ])
158 | 
159 | ACTIONS = ACTIONS_FAST
160 | if os.getenv("SBJK_EXEC_MODE") != "fast":
161 |     for k, v in ACTIONS_SLOW.items():
162 |         ACTIONS[k] = v
163 | 
164 | DBMANIFEST = "http://storage.game.starlight-stage.jp/dl/{0}/manifests"
165 | ASSETBBASEURL = "http://storage.game.starlight-stage.jp/dl/resources/High/AssetBundles/Android/{0}"
166 | SOUNDBASEURL = "http://storage.game.starlight-stage.jp/dl/resources/High/Sound/Common/{1}/{0}"
167 | SQLBASEURL = "http://storage.game.starlight-stage.jp/dl/resources/Generic/{0}"
168 | CACHE = os.path.join(os.getenv("WORKING_DIR"), "__manifestloader_cache")
169 | try:
170 |     os.makedirs(CACHE, 0o755)
171 | except FileExistsError:
172 |     pass
173 | 
174 | def batches(n, iterator):
175 |     while 1:
176 |         current_batch = []
177 |         for x in range(n):
178 |             try:
179 |                 current_batch.append(next(iterator))
180 |             except StopIteration:
181 |                 yield current_batch
182 |                 break
183 |         else:
184 |             yield current_batch
185 |             continue
186 |         break
187 | 
188 | def filename(version, platform, asset_qual, sound_qual):
189 |     return "{0}_{1}_{2}_{3}".format(version, platform, asset_qual, sound_qual)
190 | 
191 | def read_manifest(version, platform, asset_qual, sound_qual):
192 |     dest_file = os.path.join(CACHE, filename(version, platform, asset_qual, sound_qual))
193 |     if not os.path.exists(dest_file):
194 |         acquire_manifest(version, platform, asset_qual, sound_qual, dest_file)
195 | 
196 |     conn = sqlite3.connect(dest_file)
197 |     conn.row_factory = sqlite3.Row
198 |     return conn
199 | 
200 | manifest_selector_t = namedtuple("manifest_selector_t", ("filename", "md5", "platform", "asset_qual", "sound_qual"))
201 | def acquire_manifest(version, platform, asset_qual, sound_qual, dest_file):
202 |     meta = "/".join(( DBMANIFEST.format(version), "all_dbmanifest" ))
203 |     m = SESSION.get(meta)
204 |     m.raise_for_status()
205 |     mp = map(lambda x: manifest_selector_t(* x.split(",")), filter(bool, m.text.split("\n")))
206 | 
207 |     get_file = None
208 |     for selector in mp:
209 |         if selector.platform == platform and \
210 |            selector.asset_qual == asset_qual and \
211 |            selector.sound_qual == sound_qual:
212 |             get_file = selector.filename
213 |             break
214 | 
215 |     abso = "/".join(( DBMANIFEST.format(version), get_file ))
216 |     resp = SESSION.get(abso)
217 |     resp.raise_for_status()
218 | 
219 |     buf = resp.content
220 |     bio = io.BytesIO()
221 |     bio.write(buf[4:8])
222 |     bio.write(buf[16:])
223 |     data = lz4.loads(bio.getvalue())
224 |     with open(dest_file, "wb") as write_db:
225 |         write_db.write(data)
226 | 
227 |     return dest_file
228 | 
229 | seen_op_t = namedtuple("seen_op_t", ("is_insert", "name", "hash"))
230 | 
231 | def main(version):
232 |     seen_db = sqlite3.connect(os.path.join(CACHE, "A_SyncCache.db"))
233 |     seen_db.execute("CREATE TABLE IF NOT EXISTS seen (name TEXT, hash TEXT)")
234 |     seen_db.commit()
235 | 
236 |     sql = read_manifest(version, "Android", "High", "High")
237 |     query = "SELECT name, hash, attr FROM manifests WHERE ({0})".format(
238 |         " OR ".join("name GLOB ?" for glob in ACTIONS))
239 |     all_wanted_files = sql.execute(query, tuple(ACTIONS.keys()))
240 | 
241 |     for batch in batches(100, all_wanted_files):
242 |         seen_ops = []
243 | 
244 |         cm = seen_db.execute("SELECT name, hash FROM seen WHERE name IN ({0})".format(
245 |             ",".join("?" for row in batch)
246 |         ), tuple(row["name"] for row in batch))
247 |         cm = {row[0]: row[1] for row in cm}
248 | 
249 |         for row in batch:
250 |             hashid = cm.get(row["name"])
251 |             if hashid != row["hash"]:
252 |                 print(row["name"])
253 | 
254 |                 try:
255 |                     do_action_for_file(row["name"], row["hash"], row["attr"])
256 |                 except Exception as e:
257 |                     print(e)
258 |                     continue
259 | 
260 |                 if hashid is None:
261 |                     seen_ops.append(seen_op_t(1, row["name"], row["hash"]))
262 |                 else:
263 |                     seen_ops.append(seen_op_t(0, row["name"], row["hash"]))
264 | 
265 |         for op in seen_ops:
266 |             if op.is_insert:
267 |                 q = "INSERT INTO seen VALUES (:name, :hash)"
268 |             else:
269 |                 q = "UPDATE seen SET hash = :hash WHERE name = :name"
270 |             seen_db.execute(q, {"name": op.name, "hash": op.hash})
271 | 
272 |         seen_db.commit()
273 |     sql.close()
274 |     seen_db.close()
275 | 
276 | if __name__ == '__main__':
277 |     main(int(sys.argv[1]))
278 | 


--------------------------------------------------------------------------------
/simver:
--------------------------------------------------------------------------------
1 | #!/bin/sh
2 | 
3 | if [[ "x$SIMULATED_VERSION" == "x" ]]; then
4 |     echo $1
5 | else
6 |     echo $SIMULATED_VERSION
7 | fi


--------------------------------------------------------------------------------
/speculate.py:
--------------------------------------------------------------------------------
 1 | #!/usr/bin/env python3
 2 | import sys
 3 | import requests
 4 | 
 5 | URL_BASE = "http://storage.game.starlight-stage.jp/dl/{0}/manifests/all_dbmanifest"
 6 | 
 7 | version = int(sys.argv[1])
 8 | next_major = (version + 100) - (version % 100)
 9 | next_minor50 = (version + 50) - (version % 50)
10 | next_minor = (version + 10)
11 | speculated_versions = [next_major, next_minor, next_minor50]
12 | 
13 | for ver in speculated_versions:
14 |     if requests.head(URL_BASE.format(ver)).status_code != 404:
15 |         print(ver)
16 |         break
17 | else:
18 |     print(version)


--------------------------------------------------------------------------------
/versioncheck.py:
--------------------------------------------------------------------------------
  1 | #!/usr/bin/env python3
  2 | # -!- coding: utf-8 -!-
  3 | #
  4 | # Copyright 2016 Hector Martin <marcan@marcan.st>
  5 | #
  6 | # Licensed under the Apache License, Version 2.0 (the "License");
  7 | # you may not use this file except in compliance with the License.
  8 | # You may obtain a copy of the License at
  9 | #
 10 | #     http://www.apache.org/licenses/LICENSE-2.0
 11 | #
 12 | # Unless required by applicable law or agreed to in writing, software
 13 | # distributed under the License is distributed on an "AS IS" BASIS,
 14 | # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 15 | # See the License for the specific language governing permissions and
 16 | # limitations under the License.
 17 | 
 18 | import base64, msgpack, hashlib, random, time, os, sys
 19 | import requests
 20 | import rijndael
 21 | 
 22 | # laziness
 23 | def VIEWER_ID_KEY():
 24 |     return os.getenv("VC_AES_KEY", "").encode("ascii")
 25 | 
 26 | def SID_KEY():
 27 |     return os.getenv("VC_SID_SALT", "").encode("ascii")
 28 | 
 29 | def decrypt_cbc(s, iv, key):
 30 |     p = b"".join(rijndael.decrypt(key, bytes(s[i:i+len(iv)])) for i in range(0, len(s), len(iv)))
 31 |     return bytes((iv+s)[i] ^ p[i] for i in range(len(p)))
 32 | 
 33 | def encrypt_cbc(s, iv, key):
 34 |     if len(s) % 32:
 35 |         s += b"\x00" * (32 - (len(s) % 32))
 36 |     out = [iv]
 37 |     for i in range(0, len(s), 32):
 38 |         blk = bytes(s[i+j] ^ out[-1][j] for j in range(32))
 39 |         out.append(rijndael.encrypt(key, blk))
 40 |     return b"".join(out[1:])
 41 | 
 42 | class ApiClient(object):
 43 |     BASE = "http://game.starlight-stage.jp"
 44 |     def __init__(self, user, viewer_id, udid, res_ver="10013600"):
 45 |         self.user = user
 46 |         self.viewer_id = viewer_id
 47 |         self.udid = udid
 48 |         self.sid = None
 49 |         self.res_ver = res_ver
 50 | 
 51 |     def lolfuscate(self, s):
 52 |         return "%04x" % len(s) + "".join(
 53 |             "%02d" % random.randrange(100) +
 54 |             chr(ord(c) + 10) + "%d" % random.randrange(10)
 55 |             for c in s) + "%032d" % random.randrange(10**32)
 56 | 
 57 |     def unlolfuscate(self, s):
 58 |         return "".join(chr(ord(c) - 10) for c in s[6::4][:int(s[:4], 16)])
 59 | 
 60 |     def call(self, path, args):
 61 |         vid_iv = ("%032d" % random.randrange(10**32)).encode("ascii")
 62 |         args["viewer_id"] = vid_iv + base64.b64encode(
 63 |             encrypt_cbc(bytes(self.viewer_id, "ascii"),
 64 |                         vid_iv,
 65 |                         VIEWER_ID_KEY()))
 66 |         plain = base64.b64encode(msgpack.packb(args))
 67 |         # I don't even
 68 |         key = base64.b64encode(bytes(random.randrange(255) for i in range(32)))[:32]
 69 |         msg_iv = self.udid.replace("-","").encode("ascii")
 70 |         body = base64.b64encode(encrypt_cbc(plain, msg_iv, key) + key)
 71 |         sid = self.sid if self.sid else str(self.viewer_id) + self.udid
 72 |         headers = {
 73 |             "PARAM": hashlib.sha1(bytes(self.udid + str(self.viewer_id) + path, "ascii") + plain).hexdigest(),
 74 |             "KEYCHAIN": "",
 75 |             "USER_ID": self.lolfuscate(str(self.user)),
 76 |             "CARRIER": "google",
 77 |             "UDID": self.lolfuscate(self.udid),
 78 |             "APP_VER": os.getenv("VC_APP_VER", "2.1.1"),
 79 |             "RES_VER": str(self.res_ver),
 80 |             "IP_ADDRESS": "127.0.0.1",
 81 |             "DEVICE_NAME": "Nexus 42",
 82 |             "X-Unity-Version": "5.1.2f1",
 83 |             "SID": hashlib.md5(bytes(sid, "ascii") + SID_KEY()).hexdigest(),
 84 |             "GRAPHICS_DEVICE_NAME": "3dfx Voodoo2 (TM)",
 85 |             "DEVICE_ID": hashlib.md5(b"Totally a real Android").hexdigest(),
 86 |             "PLATFORM_OS_VERSION": "Android OS 13.3.7 / API-42 (XYZZ1Y/74726f6c6c)",
 87 |             "DEVICE": "2",
 88 |             "Content-Type": "application/x-www-form-urlencoded", # lies
 89 |             "User-Agent": "Dalvik/2.1.0 (Linux; U; Android 13.3.7; Nexus 42 Build/XYZZ1Y)",
 90 |         }
 91 | 
 92 |         response = requests.post(self.BASE + path, headers=headers, data=body)
 93 |         reply = base64.b64decode(response.content)
 94 |         plain = decrypt_cbc(reply[:-32], msg_iv, reply[-32:]).split(b"\0")[0]
 95 |         msg = msgpack.unpackb(base64.b64decode(plain))
 96 |         try:
 97 |             self.sid = msg["data_headers"]["sid"]
 98 |         except:
 99 |             pass
100 | 
101 |         return msg
102 | 
103 | def versioncheck():
104 |     user_id, viewer_id, udid = os.getenv("VC_ACCOUNT", "::").split(":")
105 |     client = ApiClient(user_id, viewer_id, udid)
106 |     args = {
107 |         "campaign_data": "",
108 |         "campaign_user": 1337,
109 |         "campaign_sign": hashlib.md5(b"All your APIs are belong to us").hexdigest(),
110 |         "app_type": 0,
111 |     }
112 |     msg = client.call("/load/check", args)
113 |     return msg[b"data_headers"][b"required_res_ver"]
114 | 
115 | def check(__vcache={"t": 0, "v": 0}):
116 |     if time.time() - __vcache["t"] > 3600:
117 |         nv = versioncheck()
118 |         __vcache["t"] = time.time()
119 |         __vcache["v"] = nv
120 |     return __vcache["v"]
121 | 
122 | if __name__ == '__main__':
123 |     if not all(os.getenv(var) for var in ["VC_ACCOUNT", "VC_SID_SALT", "VC_AES_KEY", "VC_APP_VER"]):
124 |         print("warning: not checking version", file=sys.stderr)
125 |         print(sys.argv[1])
126 |     else:
127 |         print(versioncheck().decode("ascii"))
128 | 


--------------------------------------------------------------------------------