summaryrefslogtreecommitdiff
path: root/package/busybox/busybox-1.11.0-bunzip2.patch
blob: 97f9fec0b80331f59e0bac9f9877e9ecd24a7296 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
--- busybox-1.11.0/archival/libunarchive/decompress_bunzip2.c	Wed Jun 25 14:51:26 2008
+++ busybox-1.11.0-bunzip2/archival/libunarchive/decompress_bunzip2.c	Sat Jun 28 20:04:31 2008
@@ -66,7 +66,6 @@
  *  | grep 'bd->' | sed 's/^.*bd->/bd->/' | sort | $PAGER
  * and moved it (inbufBitCount) to offset 0.
  */
-
 struct bunzip_data {
 	/* I/O tracking data (file handles, buffers, positions, etc.) */
 	unsigned inbufBitCount, inbufBits;
@@ -102,11 +101,9 @@
 
 	/* If we need to get more data from the byte buffer, do so.  (Loop getting
 	   one byte at a time to enforce endianness and avoid unaligned access.) */
-
 	while ((int)(bd->inbufBitCount) < bits_wanted) {
 
 		/* If we need to read more data from file into byte buffer, do so */
-
 		if (bd->inbufPos == bd->inbufCount) {
 			/* if "no input fd" case: in_fd == -1, read fails, we jump */
 			bd->inbufCount = read(bd->in_fd, bd->inbuf, IOBUF_SIZE);
@@ -116,7 +113,6 @@
 		}
 
 		/* Avoid 32-bit overflow (dump bit buffer to top of output) */
-
 		if (bd->inbufBitCount >= 24) {
 			bits = bd->inbufBits & ((1 << bd->inbufBitCount) - 1);
 			bits_wanted -= bd->inbufBitCount;
@@ -125,13 +121,11 @@
 		}
 
 		/* Grab next 8 bits of input from buffer. */
-
 		bd->inbufBits = (bd->inbufBits << 8) | bd->inbuf[bd->inbufPos++];
 		bd->inbufBitCount += 8;
 	}
 
 	/* Calculate result */
-
 	bd->inbufBitCount -= bits_wanted;
 	bits |= (bd->inbufBits >> bd->inbufBitCount) & ((1 << bits_wanted) - 1);
 
@@ -139,7 +133,6 @@
 }
 
 /* Unpacks the next block and sets up for the inverse burrows-wheeler step. */
-
 static int get_next_block(bunzip_data *bd)
 {
 	struct group_data *hufGroup;
@@ -153,13 +146,11 @@
 	selectors = bd->selectors;
 
 	/* Reset longjmp I/O error handling */
-
 	i = setjmp(bd->jmpbuf);
 	if (i) return i;
 
 	/* Read in header signature and CRC, then validate signature.
 	   (last block signature means CRC is for whole file, return now) */
-
 	i = get_bits(bd, 24);
 	j = get_bits(bd, 24);
 	bd->headerCRC = get_bits(bd, 32);
@@ -169,7 +160,6 @@
 	/* We can add support for blockRandomised if anybody complains.  There was
 	   some code for this in busybox 1.0.0-pre3, but nobody ever noticed that
 	   it didn't actually work. */
-
 	if (get_bits(bd, 1)) return RETVAL_OBSOLETE_INPUT;
 	origPtr = get_bits(bd, 24);
 	if ((int)origPtr > dbufSize) return RETVAL_DATA_ERROR;
@@ -179,7 +169,6 @@
 	   symbols to deal with, and writes a sparse bitfield indicating which
 	   values were present.  We make a translation table to convert the symbols
 	   back to the corresponding bytes. */
-
 	t = get_bits(bd, 16);
 	symTotal = 0;
 	for (i = 0; i < 16; i++) {
@@ -192,7 +181,6 @@
 	}
 
 	/* How many different Huffman coding groups does this block use? */
-
 	groupCount = get_bits(bd, 3);
 	if (groupCount < 2 || groupCount > MAX_GROUPS)
 		return RETVAL_DATA_ERROR;
@@ -201,19 +189,16 @@
 	   group.  Read in the group selector list, which is stored as MTF encoded
 	   bit runs.  (MTF=Move To Front, as each value is used it's moved to the
 	   start of the list.) */
-
 	nSelectors = get_bits(bd, 15);
 	if (!nSelectors) return RETVAL_DATA_ERROR;
 	for (i = 0; i < groupCount; i++) mtfSymbol[i] = i;
 	for (i = 0; i < nSelectors; i++) {
 
 		/* Get next value */
-
 		for (j = 0; get_bits(bd, 1); j++)
 			if (j >= groupCount) return RETVAL_DATA_ERROR;
 
 		/* Decode MTF to get the next selector */
-
 		uc = mtfSymbol[j];
 		for (;j;j--) mtfSymbol[j] = mtfSymbol[j-1];
 		mtfSymbol[0] = selectors[i] = uc;
@@ -221,10 +206,11 @@
 
 	/* Read the Huffman coding tables for each group, which code for symTotal
 	   literal symbols, plus two run symbols (RUNA, RUNB) */
-
 	symCount = symTotal + 2;
 	for (j = 0; j < groupCount; j++) {
-		unsigned char length[MAX_SYMBOLS], temp[MAX_HUFCODE_BITS+1];
+		unsigned char length[MAX_SYMBOLS];
+		/* 8 bits is ALMOST enough for temp[], see below */
+		unsigned temp[MAX_HUFCODE_BITS+1];
 		int minLen, maxLen, pp;
 
 		/* Read Huffman code lengths for each symbol.  They're stored in
@@ -233,7 +219,6 @@
 		   (Subtracting 1 before the loop and then adding it back at the end is
 		   an optimization that makes the test inside the loop simpler: symbol
 		   length 0 becomes negative, so an unsigned inequality catches it.) */
-
 		t = get_bits(bd, 5) - 1;
 		for (i = 0; i < symCount; i++) {
 			for (;;) {
@@ -243,7 +228,6 @@
 				/* If first bit is 0, stop.  Else second bit indicates whether
 				   to increment or decrement the value.  Optimization: grab 2
 				   bits and unget the second if the first was 0. */
-
 				k = get_bits(bd, 2);
 				if (k < 2) {
 					bd->inbufBitCount++;
@@ -251,17 +235,14 @@
 				}
 
 				/* Add one if second bit 1, else subtract 1.  Avoids if/else */
-
 				t += (((k+1) & 2) - 1);
 			}
 
 			/* Correct for the initial -1, to get the final symbol length */
-
 			length[i] = t + 1;
 		}
 
 		/* Find largest and smallest lengths in this group */
-
 		minLen = maxLen = length[0];
 		for (i = 1; i < symCount; i++) {
 			if (length[i] > maxLen) maxLen = length[i];
@@ -278,7 +259,6 @@
 		 * number of bits can have.  This is how the Huffman codes can vary in
 		 * length: each code with a value>limit[length] needs another bit.
 		 */
-
 		hufGroup = bd->groups + j;
 		hufGroup->minLen = minLen;
 		hufGroup->maxLen = maxLen;
@@ -286,12 +266,10 @@
 		/* Note that minLen can't be smaller than 1, so we adjust the base
 		   and limit array pointers so we're not always wasting the first
 		   entry.  We do this again when using them (during symbol decoding).*/
-
 		base = hufGroup->base - 1;
 		limit = hufGroup->limit - 1;
 
 		/* Calculate permute[].  Concurently, initialize temp[] and limit[]. */
-
 		pp = 0;
 		for (i = minLen; i <= maxLen; i++) {
 			temp[i] = limit[i] = 0;
@@ -301,14 +279,14 @@
 		}
 
 		/* Count symbols coded for at each bit length */
-
+		/* NB: in pathological cases, temp[8] can end ip being 256.
+		 * That's why uint8_t is too small for temp[]. */
 		for (i = 0; i < symCount; i++) temp[length[i]]++;
 
 		/* Calculate limit[] (the largest symbol-coding value at each bit
 		 * length, which is (previous limit<<1)+symbols at this level), and
 		 * base[] (number of symbols to ignore at each bit length, which is
 		 * limit minus the cumulative count of symbols coded for already). */
-
 		pp = t = 0;
 		for (i = minLen; i < maxLen; i++) {
 			pp += temp[i];
@@ -319,7 +297,6 @@
 			   each level we're really only interested in the first few bits,
 			   so here we set all the trailing to-be-ignored bits to 1 so they
 			   don't affect the value>limit[length] comparison. */
-
 			limit[i] = (pp << (maxLen - i)) - 1;
 			pp <<= 1;
 			t += temp[i];
@@ -335,7 +312,6 @@
 	   and run length encoding, saving the result into dbuf[dbufCount++] = uc */
 
 	/* Initialize symbol occurrence counters and symbol Move To Front table */
-
 	memset(byteCount, 0, sizeof(byteCount)); /* smaller, maybe slower? */
 	for (i = 0; i < 256; i++) {
 		//byteCount[i] = 0;
@@ -347,8 +323,7 @@
 	runPos = dbufCount = selector = 0;
 	for (;;) {
 
-		/* fetch next Huffman coding group from list. */
-
+		/* Fetch next Huffman coding group from list. */
 		symCount = GROUP_SIZE - 1;
 		if (selector >= nSelectors) return RETVAL_DATA_ERROR;
 		hufGroup = bd->groups + selectors[selector++];
@@ -367,7 +342,6 @@
 		   dry).  The following (up to got_huff_bits:) is equivalent to
 		   j = get_bits(bd, hufGroup->maxLen);
 		 */
-
 		while ((int)(bd->inbufBitCount) < hufGroup->maxLen) {
 			if (bd->inbufPos == bd->inbufCount) {
 				j = get_bits(bd, hufGroup->maxLen);
@@ -382,13 +356,11 @@
  got_huff_bits:
 
 		/* Figure how how many bits are in next symbol and unget extras */
-
 		i = hufGroup->minLen;
 		while (j > limit[i]) ++i;
 		bd->inbufBitCount += (hufGroup->maxLen - i);
 
 		/* Huffman decode value to get nextSym (with bounds checking) */
-
 		if (i > hufGroup->maxLen)
 			return RETVAL_DATA_ERROR;
 		j = (j >> (hufGroup->maxLen - i)) - base[i];
@@ -400,11 +372,9 @@
 		   byte, or a repeated run of the most recent literal byte.  First,
 		   check if nextSym indicates a repeated run, and if so loop collecting
 		   how many times to repeat the last literal. */
-
 		if ((unsigned)nextSym <= SYMBOL_RUNB) { /* RUNA or RUNB */
 
 			/* If this is the start of a new run, zero out counter */
-
 			if (!runPos) {
 				runPos = 1;
 				t = 0;
@@ -417,7 +387,6 @@
 			   the basic or 0/1 method (except all bits 0, which would use no
 			   symbols, but a run of length 0 doesn't mean anything in this
 			   context).  Thus space is saved. */
-
 			t += (runPos << nextSym); /* +runPos if RUNA; +2*runPos if RUNB */
 			if (runPos < dbufSize) runPos <<= 1;
 			goto end_of_huffman_loop;
@@ -427,7 +396,6 @@
 		   how many times to repeat the last literal, so append that many
 		   copies to our buffer of decoded symbols (dbuf) now.  (The last
 		   literal used is the one at the head of the mtfSymbol array.) */
-
 		if (runPos) {
 			runPos = 0;
 			if (dbufCount + t >= dbufSize) return RETVAL_DATA_ERROR;
@@ -438,7 +406,6 @@
 		}
 
 		/* Is this the terminating symbol? */
-
 		if (nextSym > symTotal) break;
 
 		/* At this point, nextSym indicates a new literal character.  Subtract
@@ -448,7 +415,6 @@
 		   first symbol in the mtf array, position 0, would have been handled
 		   as part of a run above.  Therefore 1 unused mtf position minus
 		   2 non-literal nextSym values equals -1.) */
-
 		if (dbufCount >= dbufSize) return RETVAL_DATA_ERROR;
 		i = nextSym - 1;
 		uc = mtfSymbol[i];
@@ -457,7 +423,6 @@
 		 * small number of symbols, and are bound by 256 in any case, using
 		 * memmove here would typically be bigger and slower due to function
 		 * call overhead and other assorted setup costs. */
-
 		do {
 			mtfSymbol[i] = mtfSymbol[i-1];
 		} while (--i);
@@ -465,13 +430,11 @@
 		uc = symToByte[uc];
 
 		/* We have our literal byte.  Save it into dbuf. */
-
 		byteCount[uc]++;
 		dbuf[dbufCount++] = (unsigned)uc;
 
 		/* Skip group initialization if we're not done with this group.  Done
 		 * this way to avoid compiler warning. */
-
  end_of_huffman_loop:
 		if (symCount--) goto continue_this_group;
 	}
@@ -484,7 +447,6 @@
 	 */
 
 	/* Turn byteCount into cumulative occurrence counts of 0 to n-1. */
-
 	j = 0;
 	for (i = 0; i < 256; i++) {
 		k = j + byteCount[i];
@@ -493,7 +455,6 @@
 	}
 
 	/* Figure out what order dbuf would be in if we sorted it. */
-
 	for (i = 0; i < dbufCount; i++) {
 		uc = (unsigned char)(dbuf[i] & 0xff);
 		dbuf[byteCount[uc]] |= (i << 8);
@@ -503,11 +464,10 @@
 	/* Decode first byte by hand to initialize "previous" byte.  Note that it
 	   doesn't get output, and if the first three characters are identical
 	   it doesn't qualify as a run (hence writeRunCountdown=5). */
-
 	if (dbufCount) {
 		if ((int)origPtr >= dbufCount) return RETVAL_DATA_ERROR;
 		bd->writePos = dbuf[origPtr];
-	    bd->writeCurrent = (unsigned char)(bd->writePos & 0xff);
+		bd->writeCurrent = (unsigned char)(bd->writePos & 0xff);
 		bd->writePos >>= 8;
 		bd->writeRunCountdown = 5;
 	}
@@ -522,7 +482,6 @@
    error (all errors are negative numbers).  If out_fd!=-1, outbuf and len
    are ignored, data is written to out_fd and return is RETVAL_OK or error.
 */
-
 int read_bunzip(bunzip_data *bd, char *outbuf, int len)
 {
 	const unsigned *dbuf;
@@ -539,19 +498,15 @@
 	/* We will always have pending decoded data to write into the output
 	   buffer unless this is the very first call (in which case we haven't
 	   Huffman-decoded a block into the intermediate buffer yet). */
-
 	if (bd->writeCopies) {
 
 		/* Inside the loop, writeCopies means extra copies (beyond 1) */
-
 		--bd->writeCopies;
 
 		/* Loop outputting bytes */
-
 		for (;;) {
 
 			/* If the output buffer is full, snapshot state and return */
-
 			if (gotcount >= len) {
 				bd->writePos = pos;
 				bd->writeCurrent = current;
@@ -560,13 +515,11 @@
 			}
 
 			/* Write next byte into output buffer, updating CRC */
-
 			outbuf[gotcount++] = current;
 			bd->writeCRC = (bd->writeCRC << 8)
-						  ^ bd->crc32Table[(bd->writeCRC >> 24) ^ current];
+				^ bd->crc32Table[(bd->writeCRC >> 24) ^ current];
 
 			/* Loop now if we're outputting multiple copies of this byte */
-
 			if (bd->writeCopies) {
 				--bd->writeCopies;
 				continue;
@@ -582,35 +535,29 @@
 			/* After 3 consecutive copies of the same byte, the 4th
 			 * is a repeat count.  We count down from 4 instead
 			 * of counting up because testing for non-zero is faster */
-
 			if (--bd->writeRunCountdown) {
 				if (current != previous)
 					bd->writeRunCountdown = 4;
 			} else {
 
 				/* We have a repeated run, this byte indicates the count */
-
 				bd->writeCopies = current;
 				current = previous;
 				bd->writeRunCountdown = 5;
 
 				/* Sometimes there are just 3 bytes (run length 0) */
-
 				if (!bd->writeCopies) goto decode_next_byte;
 
 				/* Subtract the 1 copy we'd output anyway to get extras */
-
 				--bd->writeCopies;
 			}
 		}
 
 		/* Decompression of this block completed successfully */
-
 		bd->writeCRC = ~bd->writeCRC;
 		bd->totalCRC = ((bd->totalCRC << 1) | (bd->totalCRC >> 31)) ^ bd->writeCRC;
 
 		/* If this block had a CRC error, force file level CRC error. */
-
 		if (bd->writeCRC != bd->headerCRC) {
 			bd->totalCRC = bd->headerCRC + 1;
 			return RETVAL_LAST_BLOCK;
@@ -619,7 +566,6 @@
 
 	/* Refill the intermediate buffer by Huffman-decoding next block of input */
 	/* (previous is just a convenient unused temp variable here) */
-
 	previous = get_next_block(bd);
 	if (previous) {
 		bd->writeCount = previous;
@@ -631,7 +577,6 @@
 	goto decode_next_byte;
 }
 
-
 /* Allocate the structure, read file header.  If in_fd==-1, inbuf must contain
    a complete bunzip file (len bytes long).  If in_fd!=-1, inbuf and len are
    ignored, and data is read from file handle into temporary buffer. */
@@ -639,7 +584,6 @@
 /* Because bunzip2 is used for help text unpacking, and because bb_show_usage()
    should work for NOFORK applets too, we must be extremely careful to not leak
    any allocations! */
-
 int start_bunzip(bunzip_data **bdp, int in_fd, const unsigned char *inbuf,
 						int len)
 {
@@ -650,16 +594,13 @@
 	};
 
 	/* Figure out how much data to allocate */
-
 	i = sizeof(bunzip_data);
 	if (in_fd != -1) i += IOBUF_SIZE;
 
 	/* Allocate bunzip_data.  Most fields initialize to zero. */
-
 	bd = *bdp = xzalloc(i);
 
 	/* Setup input buffer */
-
 	bd->in_fd = in_fd;
 	if (-1 == in_fd) {
 		/* in this case, bd->inbuf is read-only */
@@ -669,22 +610,18 @@
 		bd->inbuf = (unsigned char *)(bd + 1);
 
 	/* Init the CRC32 table (big endian) */
-
 	crc32_filltable(bd->crc32Table, 1);
 
 	/* Setup for I/O error handling via longjmp */
-
 	i = setjmp(bd->jmpbuf);
 	if (i) return i;
 
 	/* Ensure that file starts with "BZh['1'-'9']." */
-
 	i = get_bits(bd, 32);
 	if ((unsigned)(i - BZh0 - 1) >= 9) return RETVAL_NOT_BZIP_DATA;
 
-	/* Fourth byte (ascii '1'-'9'), indicates block size in units of 100k of
+	/* Fourth byte (ascii '1'-'9') indicates block size in units of 100k of
 	   uncompressed data.  Allocate intermediate buffer for block. */
-
 	bd->dbufSize = 100000 * (i - BZh0);
 
 	/* Cannot use xmalloc - may leak bd in NOFORK case! */
@@ -704,7 +641,6 @@
 
 
 /* Decompress src_fd to dst_fd.  Stops at end of bzip data, not end of file. */
-
 USE_DESKTOP(long long) int
 unpack_bz2_stream(int src_fd, int dst_fd)
 {
@@ -761,9 +697,9 @@
 	char c;
 
 	if (i < 0)
-		fprintf(stderr,"%s\n", bunzip_errors[-i]);
+		fprintf(stderr, "%s\n", bunzip_errors[-i]);
 	else if (read(STDIN_FILENO, &c, 1))
-		fprintf(stderr,"Trailing garbage ignored\n");
+		fprintf(stderr, "Trailing garbage ignored\n");
 	return -i;
 }
 #endif