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
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
|
// Copyright (c) 2009-2010 Satoshi Nakamoto
// Copyright (c) 2009-2022 The Bitcoin Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef BITCOIN_STREAMS_H
#define BITCOIN_STREAMS_H
#include <serialize.h>
#include <span.h>
#include <support/allocators/zeroafterfree.h>
#include <util/overflow.h>
#include <algorithm>
#include <assert.h>
#include <cstddef>
#include <cstdio>
#include <ios>
#include <limits>
#include <optional>
#include <stdint.h>
#include <string.h>
#include <string>
#include <utility>
#include <vector>
namespace util {
inline void Xor(Span<std::byte> write, Span<const std::byte> key, size_t key_offset = 0)
{
if (key.size() == 0) {
return;
}
key_offset %= key.size();
for (size_t i = 0, j = key_offset; i != write.size(); i++) {
write[i] ^= key[j++];
// This potentially acts on very many bytes of data, so it's
// important that we calculate `j`, i.e. the `key` index in this
// way instead of doing a %, which would effectively be a division
// for each byte Xor'd -- much slower than need be.
if (j == key.size())
j = 0;
}
}
} // namespace util
template<typename Stream>
class OverrideStream
{
Stream* stream;
const int nType;
const int nVersion;
public:
OverrideStream(Stream* stream_, int nType_, int nVersion_) : stream(stream_), nType(nType_), nVersion(nVersion_) {}
template<typename T>
OverrideStream<Stream>& operator<<(const T& obj)
{
::Serialize(*this, obj);
return (*this);
}
template<typename T>
OverrideStream<Stream>& operator>>(T&& obj)
{
::Unserialize(*this, obj);
return (*this);
}
void write(Span<const std::byte> src)
{
stream->write(src);
}
void read(Span<std::byte> dst)
{
stream->read(dst);
}
int GetVersion() const { return nVersion; }
int GetType() const { return nType; }
size_t size() const { return stream->size(); }
void ignore(size_t size) { return stream->ignore(size); }
};
/* Minimal stream for overwriting and/or appending to an existing byte vector
*
* The referenced vector will grow as necessary
*/
class CVectorWriter
{
public:
/*
* @param[in] nTypeIn Serialization Type
* @param[in] nVersionIn Serialization Version (including any flags)
* @param[in] vchDataIn Referenced byte vector to overwrite/append
* @param[in] nPosIn Starting position. Vector index where writes should start. The vector will initially
* grow as necessary to max(nPosIn, vec.size()). So to append, use vec.size().
*/
CVectorWriter(int nTypeIn, int nVersionIn, std::vector<unsigned char>& vchDataIn, size_t nPosIn) : nType(nTypeIn), nVersion(nVersionIn), vchData(vchDataIn), nPos(nPosIn)
{
if(nPos > vchData.size())
vchData.resize(nPos);
}
/*
* (other params same as above)
* @param[in] args A list of items to serialize starting at nPosIn.
*/
template <typename... Args>
CVectorWriter(int nTypeIn, int nVersionIn, std::vector<unsigned char>& vchDataIn, size_t nPosIn, Args&&... args) : CVectorWriter(nTypeIn, nVersionIn, vchDataIn, nPosIn)
{
::SerializeMany(*this, std::forward<Args>(args)...);
}
void write(Span<const std::byte> src)
{
assert(nPos <= vchData.size());
size_t nOverwrite = std::min(src.size(), vchData.size() - nPos);
if (nOverwrite) {
memcpy(vchData.data() + nPos, src.data(), nOverwrite);
}
if (nOverwrite < src.size()) {
vchData.insert(vchData.end(), UCharCast(src.data()) + nOverwrite, UCharCast(src.end()));
}
nPos += src.size();
}
template<typename T>
CVectorWriter& operator<<(const T& obj)
{
::Serialize(*this, obj);
return (*this);
}
int GetVersion() const
{
return nVersion;
}
int GetType() const
{
return nType;
}
private:
const int nType;
const int nVersion;
std::vector<unsigned char>& vchData;
size_t nPos;
};
/** Minimal stream for reading from an existing byte array by Span.
*/
class SpanReader
{
private:
const int m_type;
const int m_version;
Span<const unsigned char> m_data;
public:
/**
* @param[in] type Serialization Type
* @param[in] version Serialization Version (including any flags)
* @param[in] data Referenced byte vector to overwrite/append
*/
SpanReader(int type, int version, Span<const unsigned char> data)
: m_type(type), m_version(version), m_data(data) {}
template<typename T>
SpanReader& operator>>(T&& obj)
{
::Unserialize(*this, obj);
return (*this);
}
int GetVersion() const { return m_version; }
int GetType() const { return m_type; }
size_t size() const { return m_data.size(); }
bool empty() const { return m_data.empty(); }
void read(Span<std::byte> dst)
{
if (dst.size() == 0) {
return;
}
// Read from the beginning of the buffer
if (dst.size() > m_data.size()) {
throw std::ios_base::failure("SpanReader::read(): end of data");
}
memcpy(dst.data(), m_data.data(), dst.size());
m_data = m_data.subspan(dst.size());
}
};
/** Double ended buffer combining vector and stream-like interfaces.
*
* >> and << read and write unformatted data using the above serialization templates.
* Fills with data in linear time; some stringstream implementations take N^2 time.
*/
class DataStream
{
protected:
using vector_type = SerializeData;
vector_type vch;
vector_type::size_type m_read_pos{0};
public:
typedef vector_type::allocator_type allocator_type;
typedef vector_type::size_type size_type;
typedef vector_type::difference_type difference_type;
typedef vector_type::reference reference;
typedef vector_type::const_reference const_reference;
typedef vector_type::value_type value_type;
typedef vector_type::iterator iterator;
typedef vector_type::const_iterator const_iterator;
typedef vector_type::reverse_iterator reverse_iterator;
explicit DataStream() {}
explicit DataStream(Span<const uint8_t> sp) : DataStream{AsBytes(sp)} {}
explicit DataStream(Span<const value_type> sp) : vch(sp.data(), sp.data() + sp.size()) {}
std::string str() const
{
return std::string{UCharCast(data()), UCharCast(data() + size())};
}
//
// Vector subset
//
const_iterator begin() const { return vch.begin() + m_read_pos; }
iterator begin() { return vch.begin() + m_read_pos; }
const_iterator end() const { return vch.end(); }
iterator end() { return vch.end(); }
size_type size() const { return vch.size() - m_read_pos; }
bool empty() const { return vch.size() == m_read_pos; }
void resize(size_type n, value_type c = value_type{}) { vch.resize(n + m_read_pos, c); }
void reserve(size_type n) { vch.reserve(n + m_read_pos); }
const_reference operator[](size_type pos) const { return vch[pos + m_read_pos]; }
reference operator[](size_type pos) { return vch[pos + m_read_pos]; }
void clear() { vch.clear(); m_read_pos = 0; }
value_type* data() { return vch.data() + m_read_pos; }
const value_type* data() const { return vch.data() + m_read_pos; }
inline void Compact()
{
vch.erase(vch.begin(), vch.begin() + m_read_pos);
m_read_pos = 0;
}
bool Rewind(std::optional<size_type> n = std::nullopt)
{
// Total rewind if no size is passed
if (!n) {
m_read_pos = 0;
return true;
}
// Rewind by n characters if the buffer hasn't been compacted yet
if (*n > m_read_pos)
return false;
m_read_pos -= *n;
return true;
}
//
// Stream subset
//
bool eof() const { return size() == 0; }
int in_avail() const { return size(); }
void read(Span<value_type> dst)
{
if (dst.size() == 0) return;
// Read from the beginning of the buffer
auto next_read_pos{CheckedAdd(m_read_pos, dst.size())};
if (!next_read_pos.has_value() || next_read_pos.value() > vch.size()) {
throw std::ios_base::failure("DataStream::read(): end of data");
}
memcpy(dst.data(), &vch[m_read_pos], dst.size());
if (next_read_pos.value() == vch.size()) {
m_read_pos = 0;
vch.clear();
return;
}
m_read_pos = next_read_pos.value();
}
void ignore(size_t num_ignore)
{
// Ignore from the beginning of the buffer
auto next_read_pos{CheckedAdd(m_read_pos, num_ignore)};
if (!next_read_pos.has_value() || next_read_pos.value() > vch.size()) {
throw std::ios_base::failure("DataStream::ignore(): end of data");
}
if (next_read_pos.value() == vch.size()) {
m_read_pos = 0;
vch.clear();
return;
}
m_read_pos = next_read_pos.value();
}
void write(Span<const value_type> src)
{
// Write to the end of the buffer
vch.insert(vch.end(), src.begin(), src.end());
}
template<typename T>
DataStream& operator<<(const T& obj)
{
::Serialize(*this, obj);
return (*this);
}
template<typename T>
DataStream& operator>>(T&& obj)
{
::Unserialize(*this, obj);
return (*this);
}
/**
* XOR the contents of this stream with a certain key.
*
* @param[in] key The key used to XOR the data in this stream.
*/
void Xor(const std::vector<unsigned char>& key)
{
util::Xor(MakeWritableByteSpan(*this), MakeByteSpan(key));
}
};
class CDataStream : public DataStream
{
private:
int nType;
int nVersion;
public:
explicit CDataStream(int nTypeIn, int nVersionIn)
: nType{nTypeIn},
nVersion{nVersionIn} {}
explicit CDataStream(Span<const uint8_t> sp, int type, int version) : CDataStream{AsBytes(sp), type, version} {}
explicit CDataStream(Span<const value_type> sp, int nTypeIn, int nVersionIn)
: DataStream{sp},
nType{nTypeIn},
nVersion{nVersionIn} {}
int GetType() const { return nType; }
void SetVersion(int n) { nVersion = n; }
int GetVersion() const { return nVersion; }
template <typename T>
CDataStream& operator<<(const T& obj)
{
::Serialize(*this, obj);
return *this;
}
template <typename T>
CDataStream& operator>>(T&& obj)
{
::Unserialize(*this, obj);
return *this;
}
};
template <typename IStream>
class BitStreamReader
{
private:
IStream& m_istream;
/// Buffered byte read in from the input stream. A new byte is read into the
/// buffer when m_offset reaches 8.
uint8_t m_buffer{0};
/// Number of high order bits in m_buffer already returned by previous
/// Read() calls. The next bit to be returned is at this offset from the
/// most significant bit position.
int m_offset{8};
public:
explicit BitStreamReader(IStream& istream) : m_istream(istream) {}
/** Read the specified number of bits from the stream. The data is returned
* in the nbits least significant bits of a 64-bit uint.
*/
uint64_t Read(int nbits) {
if (nbits < 0 || nbits > 64) {
throw std::out_of_range("nbits must be between 0 and 64");
}
uint64_t data = 0;
while (nbits > 0) {
if (m_offset == 8) {
m_istream >> m_buffer;
m_offset = 0;
}
int bits = std::min(8 - m_offset, nbits);
data <<= bits;
data |= static_cast<uint8_t>(m_buffer << m_offset) >> (8 - bits);
m_offset += bits;
nbits -= bits;
}
return data;
}
};
template <typename OStream>
class BitStreamWriter
{
private:
OStream& m_ostream;
/// Buffered byte waiting to be written to the output stream. The byte is
/// written buffer when m_offset reaches 8 or Flush() is called.
uint8_t m_buffer{0};
/// Number of high order bits in m_buffer already written by previous
/// Write() calls and not yet flushed to the stream. The next bit to be
/// written to is at this offset from the most significant bit position.
int m_offset{0};
public:
explicit BitStreamWriter(OStream& ostream) : m_ostream(ostream) {}
~BitStreamWriter()
{
Flush();
}
/** Write the nbits least significant bits of a 64-bit int to the output
* stream. Data is buffered until it completes an octet.
*/
void Write(uint64_t data, int nbits) {
if (nbits < 0 || nbits > 64) {
throw std::out_of_range("nbits must be between 0 and 64");
}
while (nbits > 0) {
int bits = std::min(8 - m_offset, nbits);
m_buffer |= (data << (64 - nbits)) >> (64 - 8 + m_offset);
m_offset += bits;
nbits -= bits;
if (m_offset == 8) {
Flush();
}
}
}
/** Flush any unwritten bits to the output stream, padding with 0's to the
* next byte boundary.
*/
void Flush() {
if (m_offset == 0) {
return;
}
m_ostream << m_buffer;
m_buffer = 0;
m_offset = 0;
}
};
/** Non-refcounted RAII wrapper for FILE*
*
* Will automatically close the file when it goes out of scope if not null.
* If you're returning the file pointer, return file.release().
* If you need to close the file early, use file.fclose() instead of fclose(file).
*/
class AutoFile
{
protected:
std::FILE* m_file;
const std::vector<std::byte> m_xor;
public:
explicit AutoFile(std::FILE* file, std::vector<std::byte> data_xor={}) : m_file{file}, m_xor{std::move(data_xor)} {}
~AutoFile() { fclose(); }
// Disallow copies
AutoFile(const AutoFile&) = delete;
AutoFile& operator=(const AutoFile&) = delete;
bool feof() const { return std::feof(m_file); }
int fclose()
{
if (auto rel{release()}) return std::fclose(rel);
return 0;
}
/** Get wrapped FILE* with transfer of ownership.
* @note This will invalidate the AutoFile object, and makes it the responsibility of the caller
* of this function to clean up the returned FILE*.
*/
std::FILE* release()
{
std::FILE* ret{m_file};
m_file = nullptr;
return ret;
}
/** Get wrapped FILE* without transfer of ownership.
* @note Ownership of the FILE* will remain with this class. Use this only if the scope of the
* AutoFile outlives use of the passed pointer.
*/
std::FILE* Get() const { return m_file; }
/** Return true if the wrapped FILE* is nullptr, false otherwise.
*/
bool IsNull() const { return m_file == nullptr; }
/** Implementation detail, only used internally. */
std::size_t detail_fread(Span<std::byte> dst);
//
// Stream subset
//
void read(Span<std::byte> dst);
void ignore(size_t nSize);
void write(Span<const std::byte> src);
template <typename T>
AutoFile& operator<<(const T& obj)
{
::Serialize(*this, obj);
return *this;
}
template <typename T>
AutoFile& operator>>(T&& obj)
{
::Unserialize(*this, obj);
return *this;
}
};
class CAutoFile : public AutoFile
{
private:
const int nVersion;
public:
explicit CAutoFile(std::FILE* file, int version, std::vector<std::byte> data_xor = {}) : AutoFile{file, std::move(data_xor)}, nVersion{version} {}
int GetVersion() const { return nVersion; }
template<typename T>
CAutoFile& operator<<(const T& obj)
{
::Serialize(*this, obj);
return (*this);
}
template<typename T>
CAutoFile& operator>>(T&& obj)
{
::Unserialize(*this, obj);
return (*this);
}
};
/** Non-refcounted RAII wrapper around a FILE* that implements a ring buffer to
* deserialize from. It guarantees the ability to rewind a given number of bytes.
*
* Will automatically close the file when it goes out of scope if not null.
* If you need to close the file early, use file.fclose() instead of fclose(file).
*/
class BufferedFile
{
private:
const int nVersion;
FILE *src; //!< source file
uint64_t nSrcPos{0}; //!< how many bytes have been read from source
uint64_t m_read_pos{0}; //!< how many bytes have been read from this
uint64_t nReadLimit; //!< up to which position we're allowed to read
uint64_t nRewind; //!< how many bytes we guarantee to rewind
std::vector<std::byte> vchBuf; //!< the buffer
//! read data from the source to fill the buffer
bool Fill() {
unsigned int pos = nSrcPos % vchBuf.size();
unsigned int readNow = vchBuf.size() - pos;
unsigned int nAvail = vchBuf.size() - (nSrcPos - m_read_pos) - nRewind;
if (nAvail < readNow)
readNow = nAvail;
if (readNow == 0)
return false;
size_t nBytes = fread((void*)&vchBuf[pos], 1, readNow, src);
if (nBytes == 0) {
throw std::ios_base::failure(feof(src) ? "BufferedFile::Fill: end of file" : "BufferedFile::Fill: fread failed");
}
nSrcPos += nBytes;
return true;
}
//! Advance the stream's read pointer (m_read_pos) by up to 'length' bytes,
//! filling the buffer from the file so that at least one byte is available.
//! Return a pointer to the available buffer data and the number of bytes
//! (which may be less than the requested length) that may be accessed
//! beginning at that pointer.
std::pair<std::byte*, size_t> AdvanceStream(size_t length)
{
assert(m_read_pos <= nSrcPos);
if (m_read_pos + length > nReadLimit) {
throw std::ios_base::failure("Attempt to position past buffer limit");
}
// If there are no bytes available, read from the file.
if (m_read_pos == nSrcPos && length > 0) Fill();
size_t buffer_offset{static_cast<size_t>(m_read_pos % vchBuf.size())};
size_t buffer_available{static_cast<size_t>(vchBuf.size() - buffer_offset)};
size_t bytes_until_source_pos{static_cast<size_t>(nSrcPos - m_read_pos)};
size_t advance{std::min({length, buffer_available, bytes_until_source_pos})};
m_read_pos += advance;
return std::make_pair(&vchBuf[buffer_offset], advance);
}
public:
BufferedFile(FILE* fileIn, uint64_t nBufSize, uint64_t nRewindIn, int nVersionIn)
: nVersion{nVersionIn}, nReadLimit{std::numeric_limits<uint64_t>::max()}, nRewind{nRewindIn}, vchBuf(nBufSize, std::byte{0})
{
if (nRewindIn >= nBufSize)
throw std::ios_base::failure("Rewind limit must be less than buffer size");
src = fileIn;
}
~BufferedFile()
{
fclose();
}
// Disallow copies
BufferedFile(const BufferedFile&) = delete;
BufferedFile& operator=(const BufferedFile&) = delete;
int GetVersion() const { return nVersion; }
void fclose()
{
if (src) {
::fclose(src);
src = nullptr;
}
}
//! check whether we're at the end of the source file
bool eof() const {
return m_read_pos == nSrcPos && feof(src);
}
//! read a number of bytes
void read(Span<std::byte> dst)
{
while (dst.size() > 0) {
auto [buffer_pointer, length]{AdvanceStream(dst.size())};
memcpy(dst.data(), buffer_pointer, length);
dst = dst.subspan(length);
}
}
//! Move the read position ahead in the stream to the given position.
//! Use SetPos() to back up in the stream, not SkipTo().
void SkipTo(const uint64_t file_pos)
{
assert(file_pos >= m_read_pos);
while (m_read_pos < file_pos) AdvanceStream(file_pos - m_read_pos);
}
//! return the current reading position
uint64_t GetPos() const {
return m_read_pos;
}
//! rewind to a given reading position
bool SetPos(uint64_t nPos) {
size_t bufsize = vchBuf.size();
if (nPos + bufsize < nSrcPos) {
// rewinding too far, rewind as far as possible
m_read_pos = nSrcPos - bufsize;
return false;
}
if (nPos > nSrcPos) {
// can't go this far forward, go as far as possible
m_read_pos = nSrcPos;
return false;
}
m_read_pos = nPos;
return true;
}
//! prevent reading beyond a certain position
//! no argument removes the limit
bool SetLimit(uint64_t nPos = std::numeric_limits<uint64_t>::max()) {
if (nPos < m_read_pos)
return false;
nReadLimit = nPos;
return true;
}
template<typename T>
BufferedFile& operator>>(T&& obj) {
::Unserialize(*this, obj);
return (*this);
}
//! search for a given byte in the stream, and remain positioned on it
void FindByte(std::byte byte)
{
// For best performance, avoid mod operation within the loop.
size_t buf_offset{size_t(m_read_pos % uint64_t(vchBuf.size()))};
while (true) {
if (m_read_pos == nSrcPos) {
// No more bytes available; read from the file into the buffer,
// setting nSrcPos to one beyond the end of the new data.
// Throws exception if end-of-file reached.
Fill();
}
const size_t len{std::min<size_t>(vchBuf.size() - buf_offset, nSrcPos - m_read_pos)};
const auto it_start{vchBuf.begin() + buf_offset};
const auto it_find{std::find(it_start, it_start + len, byte)};
const size_t inc{size_t(std::distance(it_start, it_find))};
m_read_pos += inc;
if (inc < len) break;
buf_offset += inc;
if (buf_offset >= vchBuf.size()) buf_offset = 0;
}
}
};
#endif // BITCOIN_STREAMS_H
|