BIP155 (addrv2)

Tor v3 + i2p
This commit is contained in:
zanzibar
2023-01-06 15:21:08 +00:00
parent fe9f1ef9e4
commit 512da314a5
108 changed files with 8214 additions and 2173 deletions

View File

@@ -0,0 +1,47 @@
// Copyright (c) 2015-2020 The Bitcoin Core developers
// Copyright (c) 2017 The Zcash developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
#include <fs.h>
#include <limits>
#include <stdio.h>
#include <string>
#include <utility>
std::pair<bool,std::string> ReadBinaryFile(const fs::path &filename, size_t maxsize=std::numeric_limits<size_t>::max())
{
FILE *f = fsbridge::fopen(filename, "rb");
if (f == nullptr)
return std::make_pair(false,"");
std::string retval;
char buffer[128];
do {
const size_t n = fread(buffer, 1, sizeof(buffer), f);
// Check for reading errors so we don't return any data if we couldn't
// read the entire file (or up to maxsize)
if (ferror(f)) {
fclose(f);
return std::make_pair(false,"");
}
retval.append(buffer, buffer+n);
} while (!feof(f) && retval.size() <= maxsize);
fclose(f);
return std::make_pair(true,retval);
}
bool WriteBinaryFile(const fs::path &filename, const std::string &data)
{
FILE *f = fsbridge::fopen(filename, "wb");
if (f == nullptr)
return false;
if (fwrite(data.data(), 1, data.size(), f) != data.size()) {
fclose(f);
return false;
}
if (fclose(f) != 0) {
return false;
}
return true;
}

29
src/util/readwritefile.h Normal file
View File

@@ -0,0 +1,29 @@
// Copyright (c) 2015-2020 The Bitcoin Core developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
#ifndef BITCOIN_UTIL_READWRITEFILE_H
#define BITCOIN_UTIL_READWRITEFILE_H
#include <fs.h>
#include <limits>
#include <string>
#include <utility>
/** Read full contents of a file and return them in a std::string.
* Returns a pair <status, string>.
* If an error occurred, status will be false, otherwise status will be true and the data will be returned in string.
*
* @param maxsize Puts a maximum size limit on the file that is read. If the file is larger than this, truncated data
* (with len > maxsize) will be returned.
*/
std::pair<bool,std::string> ReadBinaryFile(const fs::path &filename, size_t maxsize=std::numeric_limits<size_t>::max());
/** Write contents of std::string to a file.
* @return true on success.
*/
bool WriteBinaryFile(const fs::path &filename, const std::string &data);
#endif /* BITCOIN_UTIL_READWRITEFILE_H */

328
src/util/sock.cpp Normal file
View File

@@ -0,0 +1,328 @@
// Copyright (c) 2020-2021 The Bitcoin Core developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
#include <compat.h>
#include <tinyformat.h>
#include <util/sock.h>
#include <utiltime.h>
#include <util.h>
#include <codecvt>
#include <cwchar>
#include <locale>
#include <stdexcept>
#include <string>
#ifdef USE_POLL
#include <poll.h>
#endif
static inline bool IOErrorIsPermanent(int err)
{
return err != WSAEAGAIN && err != WSAEINTR && err != WSAEWOULDBLOCK && err != WSAEINPROGRESS;
}
Sock::Sock() : m_socket(INVALID_SOCKET) {}
Sock::Sock(SOCKET s) : m_socket(s) {}
Sock::Sock(Sock&& other)
{
m_socket = other.m_socket;
other.m_socket = INVALID_SOCKET;
}
Sock::~Sock() { Reset(); }
Sock& Sock::operator=(Sock&& other)
{
Reset();
m_socket = other.m_socket;
other.m_socket = INVALID_SOCKET;
return *this;
}
SOCKET Sock::Get() const { return m_socket; }
SOCKET Sock::Release()
{
const SOCKET s = m_socket;
m_socket = INVALID_SOCKET;
return s;
}
void Sock::Reset() { CloseSocket(m_socket); }
ssize_t Sock::Send(const void* data, size_t len, int flags) const
{
return send(m_socket, static_cast<const char*>(data), len, flags);
}
ssize_t Sock::Recv(void* buf, size_t len, int flags) const
{
return recv(m_socket, static_cast<char*>(buf), len, flags);
}
int Sock::Connect(const sockaddr* addr, socklen_t addr_len) const
{
return connect(m_socket, addr, addr_len);
}
int Sock::GetSockOpt(int level, int opt_name, void* opt_val, socklen_t* opt_len) const
{
return getsockopt(m_socket, level, opt_name, static_cast<char*>(opt_val), opt_len);
}
bool Sock::Wait(std::chrono::milliseconds timeout, Event requested, Event* occurred) const
{
#ifdef USE_POLL
pollfd fd;
fd.fd = m_socket;
fd.events = 0;
if (requested & RECV) {
fd.events |= POLLIN;
}
if (requested & SEND) {
fd.events |= POLLOUT;
}
if (poll(&fd, 1, count_milliseconds(timeout)) == SOCKET_ERROR) {
return false;
}
if (occurred != nullptr) {
*occurred = 0;
if (fd.revents & POLLIN) {
*occurred |= RECV;
}
if (fd.revents & POLLOUT) {
*occurred |= SEND;
}
}
return true;
#else
if (!IsSelectableSocket(m_socket)) {
return false;
}
fd_set fdset_recv;
fd_set fdset_send;
FD_ZERO(&fdset_recv);
FD_ZERO(&fdset_send);
if (requested & RECV) {
FD_SET(m_socket, &fdset_recv);
}
if (requested & SEND) {
FD_SET(m_socket, &fdset_send);
}
timeval timeout_struct = MillisToTimeval(timeout.count());
if (select(m_socket + 1, &fdset_recv, &fdset_send, nullptr, &timeout_struct) == SOCKET_ERROR) {
return false;
}
if (occurred != nullptr) {
*occurred = 0;
if (FD_ISSET(m_socket, &fdset_recv)) {
*occurred |= RECV;
}
if (FD_ISSET(m_socket, &fdset_send)) {
*occurred |= SEND;
}
}
return true;
#endif /* USE_POLL */
}
void Sock::SendComplete(const std::string& data,
std::chrono::milliseconds timeout) const
{
const auto deadline = GetTime<std::chrono::milliseconds>() + timeout;
size_t sent{0};
for (;;) {
const ssize_t ret{Send(data.data() + sent, data.size() - sent, MSG_NOSIGNAL)};
if (ret > 0) {
sent += static_cast<size_t>(ret);
if (sent == data.size()) {
break;
}
} else {
const int err{WSAGetLastError()};
if (IOErrorIsPermanent(err)) {
throw std::runtime_error(strprintf("send(): %s", NetworkErrorString(err)));
}
}
const auto now = GetTime<std::chrono::milliseconds>();
if (now >= deadline) {
throw std::runtime_error(strprintf(
"Send timeout (sent only %u of %u bytes before that)", sent, data.size()));
}
// Wait for a short while (or the socket to become ready for sending) before retrying
// if nothing was sent.
const auto wait_time = std::min(deadline - now, std::chrono::milliseconds{MAX_WAIT_FOR_IO});
(void)Wait(wait_time, SEND);
}
}
std::string Sock::RecvUntilTerminator(uint8_t terminator,
std::chrono::milliseconds timeout,
size_t max_data) const
{
const auto deadline = GetTime<std::chrono::milliseconds>() + timeout;
std::string data;
bool terminator_found{false};
// We must not consume any bytes past the terminator from the socket.
// One option is to read one byte at a time and check if we have read a terminator.
// However that is very slow. Instead, we peek at what is in the socket and only read
// as many bytes as possible without crossing the terminator.
// Reading 64 MiB of random data with 262526 terminator chars takes 37 seconds to read
// one byte at a time VS 0.71 seconds with the "peek" solution below. Reading one byte
// at a time is about 50 times slower.
for (;;) {
if (data.size() >= max_data) {
throw std::runtime_error(
strprintf("Received too many bytes without a terminator (%u)", data.size()));
}
char buf[512];
const ssize_t peek_ret{Recv(buf, std::min(sizeof(buf), max_data - data.size()), MSG_PEEK)};
switch (peek_ret) {
case -1: {
const int err{WSAGetLastError()};
if (IOErrorIsPermanent(err)) {
throw std::runtime_error(strprintf("recv(): %s", NetworkErrorString(err)));
}
break;
}
case 0:
throw std::runtime_error("Connection unexpectedly closed by peer");
default:
auto end = buf + peek_ret;
auto terminator_pos = std::find(buf, end, terminator);
terminator_found = terminator_pos != end;
const size_t try_len{terminator_found ? terminator_pos - buf + 1 :
static_cast<size_t>(peek_ret)};
const ssize_t read_ret{Recv(buf, try_len, 0)};
if (read_ret < 0 || static_cast<size_t>(read_ret) != try_len) {
throw std::runtime_error(
strprintf("recv() returned %u bytes on attempt to read %u bytes but previous "
"peek claimed %u bytes are available",
read_ret, try_len, peek_ret));
}
// Don't include the terminator in the output.
const size_t append_len{terminator_found ? try_len - 1 : try_len};
data.append(buf, buf + append_len);
if (terminator_found) {
return data;
}
}
const auto now = GetTime<std::chrono::milliseconds>();
if (now >= deadline) {
throw std::runtime_error(strprintf(
"Receive timeout (received %u bytes without terminator before that)", data.size()));
}
// Wait for a short while (or the socket to become ready for reading) before retrying.
const auto wait_time = std::min(deadline - now, std::chrono::milliseconds{MAX_WAIT_FOR_IO});
(void)Wait(wait_time, RECV);
}
}
bool Sock::IsConnected(std::string& errmsg) const
{
if (m_socket == INVALID_SOCKET) {
errmsg = "not connected";
return false;
}
char c;
switch (Recv(&c, sizeof(c), MSG_PEEK)) {
case -1: {
const int err = WSAGetLastError();
if (IOErrorIsPermanent(err)) {
errmsg = NetworkErrorString(err);
return false;
}
return true;
}
case 0:
errmsg = "closed";
return false;
default:
return true;
}
}
#ifdef WIN32
std::string NetworkErrorString(int err)
{
wchar_t buf[256];
buf[0] = 0;
if(FormatMessageW(FORMAT_MESSAGE_FROM_SYSTEM | FORMAT_MESSAGE_IGNORE_INSERTS | FORMAT_MESSAGE_MAX_WIDTH_MASK,
nullptr, err, MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT),
buf, ARRAYSIZE(buf), nullptr))
{
return strprintf("%s (%d)", std::wstring_convert<std::codecvt_utf8_utf16<wchar_t>,wchar_t>().to_bytes(buf), err);
}
else
{
return strprintf("Unknown error (%d)", err);
}
}
#else
std::string NetworkErrorString(int err)
{
char buf[256];
buf[0] = 0;
/* Too bad there are two incompatible implementations of the
* thread-safe strerror. */
const char *s;
#ifdef STRERROR_R_CHAR_P /* GNU variant can return a pointer outside the passed buffer */
s = strerror_r(err, buf, sizeof(buf));
#else /* POSIX variant always returns message in buffer */
s = buf;
if (strerror_r(err, buf, sizeof(buf)))
buf[0] = 0;
#endif
return strprintf("%s (%d)", s, err);
}
#endif
bool CloseSocket(SOCKET& hSocket)
{
if (hSocket == INVALID_SOCKET)
return false;
#ifdef WIN32
int ret = closesocket(hSocket);
#else
int ret = close(hSocket);
#endif
if (ret) {
LogPrintf("Socket close failed: %d. Error: %s\n", hSocket, NetworkErrorString(WSAGetLastError()));
}
hSocket = INVALID_SOCKET;
return ret != SOCKET_ERROR;
}

184
src/util/sock.h Normal file
View File

@@ -0,0 +1,184 @@
// Copyright (c) 2020-2021 The Bitcoin Core developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
#ifndef BITCOIN_UTIL_SOCK_H
#define BITCOIN_UTIL_SOCK_H
#include <compat.h>
// #include <threadinterrupt.h>
#include <utiltime.h>
#include <chrono>
#include <string>
/**
* Maximum time to wait for I/O readiness.
* It will take up until this time to break off in case of an interruption.
*/
static constexpr auto MAX_WAIT_FOR_IO = 1000;
/**
* RAII helper class that manages a socket. Mimics `std::unique_ptr`, but instead of a pointer it
* contains a socket and closes it automatically when it goes out of scope.
*/
class Sock
{
public:
/**
* Default constructor, creates an empty object that does nothing when destroyed.
*/
Sock();
/**
* Take ownership of an existent socket.
*/
explicit Sock(SOCKET s);
/**
* Copy constructor, disabled because closing the same socket twice is undesirable.
*/
Sock(const Sock&) = delete;
/**
* Move constructor, grab the socket from another object and close ours (if set).
*/
Sock(Sock&& other);
/**
* Destructor, close the socket or do nothing if empty.
*/
virtual ~Sock();
/**
* Copy assignment operator, disabled because closing the same socket twice is undesirable.
*/
Sock& operator=(const Sock&) = delete;
/**
* Move assignment operator, grab the socket from another object and close ours (if set).
*/
virtual Sock& operator=(Sock&& other);
/**
* Get the value of the contained socket.
* @return socket or INVALID_SOCKET if empty
*/
virtual SOCKET Get() const;
/**
* Get the value of the contained socket and drop ownership. It will not be closed by the
* destructor after this call.
* @return socket or INVALID_SOCKET if empty
*/
virtual SOCKET Release();
/**
* Close if non-empty.
*/
virtual void Reset();
/**
* send(2) wrapper. Equivalent to `send(this->Get(), data, len, flags);`. Code that uses this
* wrapper can be unit tested if this method is overridden by a mock Sock implementation.
*/
virtual ssize_t Send(const void* data, size_t len, int flags) const;
/**
* recv(2) wrapper. Equivalent to `recv(this->Get(), buf, len, flags);`. Code that uses this
* wrapper can be unit tested if this method is overridden by a mock Sock implementation.
*/
virtual ssize_t Recv(void* buf, size_t len, int flags) const;
/**
* connect(2) wrapper. Equivalent to `connect(this->Get(), addr, addrlen)`. Code that uses this
* wrapper can be unit tested if this method is overridden by a mock Sock implementation.
*/
virtual int Connect(const sockaddr* addr, socklen_t addr_len) const;
/**
* getsockopt(2) wrapper. Equivalent to
* `getsockopt(this->Get(), level, opt_name, opt_val, opt_len)`. Code that uses this
* wrapper can be unit tested if this method is overridden by a mock Sock implementation.
*/
virtual int GetSockOpt(int level,
int opt_name,
void* opt_val,
socklen_t* opt_len) const;
using Event = uint8_t;
/**
* If passed to `Wait()`, then it will wait for readiness to read from the socket.
*/
static constexpr Event RECV = 0b01;
/**
* If passed to `Wait()`, then it will wait for readiness to send to the socket.
*/
static constexpr Event SEND = 0b10;
/**
* Wait for readiness for input (recv) or output (send).
* @param[in] timeout Wait this much for at least one of the requested events to occur.
* @param[in] requested Wait for those events, bitwise-or of `RECV` and `SEND`.
* @param[out] occurred If not nullptr and `true` is returned, then upon return this
* indicates which of the requested events occurred. A timeout is indicated by return
* value of `true` and `occurred` being set to 0.
* @return true on success and false otherwise
*/
virtual bool Wait(std::chrono::milliseconds timeout,
Event requested,
Event* occurred = nullptr) const;
/* Higher level, convenience, methods. These may throw. */
/**
* Send the given data, retrying on transient errors.
* @param[in] data Data to send.
* @param[in] timeout Timeout for the entire operation.
* @param[in] interrupt If this is signaled then the operation is canceled.
* @throws std::runtime_error if the operation cannot be completed. In this case only some of
* the data will be written to the socket.
*/
virtual void SendComplete(const std::string& data,
std::chrono::milliseconds timeout) const;
/**
* Read from socket until a terminator character is encountered. Will never consume bytes past
* the terminator from the socket.
* @param[in] terminator Character up to which to read from the socket.
* @param[in] timeout Timeout for the entire operation.
* @param[in] interrupt If this is signaled then the operation is canceled.
* @param[in] max_data The maximum amount of data (in bytes) to receive. If this many bytes
* are received and there is still no terminator, then this method will throw an exception.
* @return The data that has been read, without the terminating character.
* @throws std::runtime_error if the operation cannot be completed. In this case some bytes may
* have been consumed from the socket.
*/
virtual std::string RecvUntilTerminator(uint8_t terminator,
std::chrono::milliseconds timeout,
size_t max_data) const;
/**
* Check if still connected.
* @param[out] errmsg The error string, if the socket has been disconnected.
* @return true if connected
*/
virtual bool IsConnected(std::string& errmsg) const;
protected:
/**
* Contained socket. `INVALID_SOCKET` designates the object is empty.
*/
SOCKET m_socket;
};
/** Return readable error string for a network error code */
std::string NetworkErrorString(int err);
/** Close socket and set hSocket to INVALID_SOCKET */
bool CloseSocket(SOCKET& hSocket);
#endif // BITCOIN_UTIL_SOCK_H

68
src/util/spanparsing.cpp Normal file
View File

@@ -0,0 +1,68 @@
// Copyright (c) 2018-2019 The Bitcoin Core developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
#include <util/spanparsing.h>
#include <span.h>
#include <string>
#include <vector>
namespace spanparsing {
bool Const(const std::string& str, Span<const char>& sp)
{
if ((size_t)sp.size() >= str.size() && std::equal(str.begin(), str.end(), sp.begin())) {
sp = sp.subspan(str.size());
return true;
}
return false;
}
bool Func(const std::string& str, Span<const char>& sp)
{
if ((size_t)sp.size() >= str.size() + 2 && sp[str.size()] == '(' && sp[sp.size() - 1] == ')' && std::equal(str.begin(), str.end(), sp.begin())) {
sp = sp.subspan(str.size() + 1, sp.size() - str.size() - 2);
return true;
}
return false;
}
Span<const char> Expr(Span<const char>& sp)
{
int level = 0;
auto it = sp.begin();
while (it != sp.end()) {
if (*it == '(' || *it == '{') {
++level;
} else if (level && (*it == ')' || *it == '}')) {
--level;
} else if (level == 0 && (*it == ')' || *it == '}' || *it == ',')) {
break;
}
++it;
}
Span<const char> ret = sp.first(it - sp.begin());
sp = sp.subspan(it - sp.begin());
return ret;
}
std::vector<Span<const char>> Split(const Span<const char>& sp, char sep)
{
std::vector<Span<const char>> ret;
auto it = sp.begin();
auto start = it;
while (it != sp.end()) {
if (*it == sep) {
ret.emplace_back(start, it);
start = it + 1;
}
++it;
}
ret.emplace_back(start, it);
return ret;
}
} // namespace spanparsing

50
src/util/spanparsing.h Normal file
View File

@@ -0,0 +1,50 @@
// Copyright (c) 2018-2019 The Bitcoin Core developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
#ifndef BITCOIN_UTIL_SPANPARSING_H
#define BITCOIN_UTIL_SPANPARSING_H
#include <span.h>
#include <string>
#include <vector>
namespace spanparsing {
/** Parse a constant.
*
* If sp's initial part matches str, sp is updated to skip that part, and true is returned.
* Otherwise sp is unmodified and false is returned.
*/
bool Const(const std::string& str, Span<const char>& sp);
/** Parse a function call.
*
* If sp's initial part matches str + "(", and sp ends with ")", sp is updated to be the
* section between the braces, and true is returned. Otherwise sp is unmodified and false
* is returned.
*/
bool Func(const std::string& str, Span<const char>& sp);
/** Extract the expression that sp begins with.
*
* This function will return the initial part of sp, up to (but not including) the first
* comma or closing brace, skipping ones that are surrounded by braces. So for example,
* for "foo(bar(1),2),3" the initial part "foo(bar(1),2)" will be returned. sp will be
* updated to skip the initial part that is returned.
*/
Span<const char> Expr(Span<const char>& sp);
/** Split a string on every instance of sep, returning a vector.
*
* If sep does not occur in sp, a singleton with the entirety of sp is returned.
*
* Note that this function does not care about braces, so splitting
* "foo(bar(1),2),3) on ',' will return {"foo(bar(1)", "2)", "3)"}.
*/
std::vector<Span<const char>> Split(const Span<const char>& sp, char sep);
} // namespace spanparsing
#endif // BITCOIN_UTIL_SPANPARSING_H

655
src/util/strencodings.cpp Normal file
View File

@@ -0,0 +1,655 @@
// Copyright (c) 2009-2010 Satoshi Nakamoto
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
#include "util/strencodings.h"
#include "util/string.h"
#include <tinyformat.h>
#include <algorithm>
#include <cstdlib>
#include <cstring>
#include <iomanip>
#include <errno.h>
#include <limits>
using namespace std;
static const std::string CHARS_ALPHA_NUM = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
static const std::string SAFE_CHARS[] =
{
CHARS_ALPHA_NUM + " .,;-_/:?@()", // SAFE_CHARS_DEFAULT
CHARS_ALPHA_NUM + " .,;-_?@", // SAFE_CHARS_UA_COMMENT
CHARS_ALPHA_NUM + ".-_", // SAFE_CHARS_FILENAME
CHARS_ALPHA_NUM + "!*'();:@&=+$,/?#[]-_.~%", // SAFE_CHARS_URI
};
std::string SanitizeString(const std::string& str, int rule)
{
std::string strResult;
for (std::string::size_type i = 0; i < str.size(); i++)
{
if (SAFE_CHARS[rule].find(str[i]) != std::string::npos)
strResult.push_back(str[i]);
}
return strResult;
}
string SanitizeFilename(const string& str)
{
/**
* safeChars chosen to restrict filename, keeping it simple to avoid cross-platform issues.
* http://stackoverflow.com/a/2306003
*/
static string safeChars("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890");
string strResult;
for (std::string::size_type i = 0; i < str.size(); i++)
{
if (safeChars.find(str[i]) != std::string::npos)
strResult.push_back(str[i]);
}
return strResult;
}
std::string HexInt(uint32_t val)
{
std::stringstream ss;
ss << std::setfill('0') << std::setw(sizeof(uint32_t) * 2) << std::hex << val;
return ss.str();
}
uint32_t ParseHexToUInt32(const std::string& str) {
std::istringstream converter(str);
uint32_t value;
converter >> std::hex >> value;
return value;
}
const signed char p_util_hexdigit[256] =
{ -1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
0,1,2,3,4,5,6,7,8,9,-1,-1,-1,-1,-1,-1,
-1,0xa,0xb,0xc,0xd,0xe,0xf,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,0xa,0xb,0xc,0xd,0xe,0xf,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,
-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1, };
signed char HexDigit(char c)
{
return p_util_hexdigit[(unsigned char)c];
}
bool IsHex(const std::string& str)
{
for(std::string::const_iterator it(str.begin()); it != str.end(); ++it)
{
if (HexDigit(*it) < 0)
return false;
}
return (str.size() > 0) && (str.size()%2 == 0);
}
bool IsHexNumber(const std::string& str)
{
size_t starting_location = 0;
if (str.size() > 2 && *str.begin() == '0' && *(str.begin()+1) == 'x') {
starting_location = 2;
}
for (const char c : str.substr(starting_location)) {
if (HexDigit(c) < 0) return false;
}
// Return false for empty string or "0x".
return (str.size() > starting_location);
}
std::vector<unsigned char> ParseHex(const char* psz)
{
// convert hex dump to vector
std::vector<unsigned char> vch;
while (true)
{
while (IsSpace(*psz))
psz++;
signed char c = HexDigit(*psz++);
if (c == (signed char)-1)
break;
unsigned char n = (c << 4);
c = HexDigit(*psz++);
if (c == (signed char)-1)
break;
n |= c;
vch.push_back(n);
}
return vch;
}
std::vector<unsigned char> ParseHex(const std::string& str)
{
return ParseHex(str.c_str());
}
void SplitHostPort(std::string in, int &portOut, std::string &hostOut) {
size_t colon = in.find_last_of(':');
// if a : is found, and it either follows a [...], or no other : is in the string, treat it as port separator
bool fHaveColon = colon != in.npos;
bool fBracketed = fHaveColon && (in[0]=='[' && in[colon-1]==']'); // if there is a colon, and in[0]=='[', colon is not 0, so in[colon-1] is safe
bool fMultiColon = fHaveColon && (in.find_last_of(':',colon-1) != in.npos);
if (fHaveColon && (colon==0 || fBracketed || !fMultiColon)) {
int32_t n;
if (ParseInt32(in.substr(colon + 1), &n) && n > 0 && n < 0x10000) {
in = in.substr(0, colon);
portOut = n;
}
}
if (in.size()>0 && in[0] == '[' && in[in.size()-1] == ']')
hostOut = in.substr(1, in.size()-2);
else
hostOut = in;
}
std::string EncodeBase64(Span<const unsigned char> input)
{
static const char *pbase64 = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
std::string str;
str.reserve(((input.size() + 2) / 3) * 4);
ConvertBits<8, 6, true>([&](int v) { str += pbase64[v]; }, input.begin(), input.end());
while (str.size() % 4) str += '=';
return str;
}
std::string EncodeBase64(const unsigned char* pch, size_t len)
{
static const char *pbase64 = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
std::string str;
str.reserve(((len + 2) / 3) * 4);
ConvertBits<8, 6, true>([&](int v) { str += pbase64[v]; }, pch, pch + len);
while (str.size() % 4) str += '=';
return str;
}
std::string EncodeBase64(const std::string& str)
{
return EncodeBase64((const unsigned char*)str.data(), str.size());
}
std::vector<unsigned char> DecodeBase64(const char* p, bool* pf_invalid)
{
static const int decode64_table[256] =
{
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, 62, -1, -1, -1, 63, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1,
-1, -1, -1, -1, -1, 0, 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, -1, -1, -1, -1, -1, -1, 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, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1
};
const char* e = p;
std::vector<uint8_t> val;
val.reserve(strlen(p));
while (*p != 0) {
int x = decode64_table[(unsigned char)*p];
if (x == -1) break;
val.push_back(x);
++p;
}
std::vector<unsigned char> ret;
ret.reserve((val.size() * 3) / 4);
bool valid = ConvertBits<6, 8, false>([&](unsigned char c) { ret.push_back(c); }, val.begin(), val.end());
const char* q = p;
while (valid && *p != 0) {
if (*p != '=') {
valid = false;
break;
}
++p;
}
valid = valid && (p - e) % 4 == 0 && p - q < 4;
if (pf_invalid) *pf_invalid = !valid;
return ret;
}
std::string DecodeBase64(const std::string& str, bool* pf_invalid)
{
if (!ValidAsCString(str)) {
if (pf_invalid) {
*pf_invalid = true;
}
return {};
}
std::vector<unsigned char> vchRet = DecodeBase64(str.c_str(), pf_invalid);
return std::string((const char*)vchRet.data(), vchRet.size());
}
std::string EncodeBase32(Span<const unsigned char> input, bool pad)
{
static const char *pbase32 = "abcdefghijklmnopqrstuvwxyz234567";
std::string str;
str.reserve(((input.size() + 4) / 5) * 8);
ConvertBits<8, 5, true>([&](int v) { str += pbase32[v]; }, input.begin(), input.end());
if (pad) {
while (str.size() % 8) {
str += '=';
}
}
return str;
}
std::string EncodeBase32(const std::string& str, bool pad)
{
return EncodeBase32(MakeUCharSpan(str), pad);
}
std::vector<unsigned char> DecodeBase32(const char* p, bool* pf_invalid)
{
static const int decode32_table[256] =
{
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 26, 27, 28, 29, 30, 31, -1, -1, -1, -1,
-1, -1, -1, -1, -1, 0, 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, -1, -1, -1, -1, -1, -1, 0, 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, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1
};
const char* e = p;
std::vector<uint8_t> val;
val.reserve(strlen(p));
while (*p != 0) {
int x = decode32_table[(unsigned char)*p];
if (x == -1) break;
val.push_back(x);
++p;
}
std::vector<unsigned char> ret;
ret.reserve((val.size() * 5) / 8);
bool valid = ConvertBits<5, 8, false>([&](unsigned char c) { ret.push_back(c); }, val.begin(), val.end());
const char* q = p;
while (valid && *p != 0) {
if (*p != '=') {
valid = false;
break;
}
++p;
}
valid = valid && (p - e) % 8 == 0 && p - q < 8;
if (pf_invalid) *pf_invalid = !valid;
return ret;
}
std::string DecodeBase32(const std::string& str, bool* pf_invalid)
{
if (!ValidAsCString(str)) {
if (pf_invalid) {
*pf_invalid = true;
}
return {};
}
std::vector<unsigned char> vchRet = DecodeBase32(str.c_str(), pf_invalid);
return std::string((const char*)vchRet.data(), vchRet.size());
}
static bool ParsePrechecks(const std::string& str)
{
if (str.empty()) // No empty string allowed
return false;
if (str.size() >= 1 && (IsSpace(str[0]) || IsSpace(str[str.size()-1]))) // No padding allowed
return false;
if (!ValidAsCString(str)) // No embedded NUL characters allowed
return false;
return true;
}
bool ParseInt32(const std::string& str, int32_t *out)
{
if (!ParsePrechecks(str))
return false;
char *endp = nullptr;
errno = 0; // strtol will not set errno if valid
long int n = strtol(str.c_str(), &endp, 10);
if(out) *out = (int32_t)n;
// Note that strtol returns a *long int*, so even if strtol doesn't report an over/underflow
// we still have to check that the returned value is within the range of an *int32_t*. On 64-bit
// platforms the size of these types may be different.
return endp && *endp == 0 && !errno &&
n >= std::numeric_limits<int32_t>::min() &&
n <= std::numeric_limits<int32_t>::max();
}
bool ParseInt64(const std::string& str, int64_t *out)
{
if (!ParsePrechecks(str))
return false;
char *endp = nullptr;
errno = 0; // strtoll will not set errno if valid
long long int n = strtoll(str.c_str(), &endp, 10);
if(out) *out = (int64_t)n;
// Note that strtoll returns a *long long int*, so even if strtol doesn't report an over/underflow
// we still have to check that the returned value is within the range of an *int64_t*.
return endp && *endp == 0 && !errno &&
n >= std::numeric_limits<int64_t>::min() &&
n <= std::numeric_limits<int64_t>::max();
}
bool ParseUInt8(const std::string& str, uint8_t *out)
{
uint32_t u32;
if (!ParseUInt32(str, &u32) || u32 > std::numeric_limits<uint8_t>::max()) {
return false;
}
if (out != nullptr) {
*out = static_cast<uint8_t>(u32);
}
return true;
}
bool ParseUInt32(const std::string& str, uint32_t *out)
{
if (!ParsePrechecks(str))
return false;
if (str.size() >= 1 && str[0] == '-') // Reject negative values, unfortunately strtoul accepts these by default if they fit in the range
return false;
char *endp = nullptr;
errno = 0; // strtoul will not set errno if valid
unsigned long int n = strtoul(str.c_str(), &endp, 10);
if(out) *out = (uint32_t)n;
// Note that strtoul returns a *unsigned long int*, so even if it doesn't report an over/underflow
// we still have to check that the returned value is within the range of an *uint32_t*. On 64-bit
// platforms the size of these types may be different.
return endp && *endp == 0 && !errno &&
n <= std::numeric_limits<uint32_t>::max();
}
bool ParseUInt64(const std::string& str, uint64_t *out)
{
if (!ParsePrechecks(str))
return false;
if (str.size() >= 1 && str[0] == '-') // Reject negative values, unfortunately strtoull accepts these by default if they fit in the range
return false;
char *endp = nullptr;
errno = 0; // strtoull will not set errno if valid
unsigned long long int n = strtoull(str.c_str(), &endp, 10);
if(out) *out = (uint64_t)n;
// Note that strtoull returns a *unsigned long long int*, so even if it doesn't report an over/underflow
// we still have to check that the returned value is within the range of an *uint64_t*.
return endp && *endp == 0 && !errno &&
n <= std::numeric_limits<uint64_t>::max();
}
bool ParseDouble(const std::string& str, double *out)
{
if (!ParsePrechecks(str))
return false;
if (str.size() >= 2 && str[0] == '0' && str[1] == 'x') // No hexadecimal floats allowed
return false;
std::istringstream text(str);
text.imbue(std::locale::classic());
double result;
text >> result;
if(out) *out = result;
return text.eof() && !text.fail();
}
std::string FormatParagraph(const std::string& in, size_t width, size_t indent)
{
std::stringstream out;
size_t ptr = 0;
size_t indented = 0;
while (ptr < in.size())
{
size_t lineend = in.find_first_of('\n', ptr);
if (lineend == std::string::npos) {
lineend = in.size();
}
const size_t linelen = lineend - ptr;
const size_t rem_width = width - indented;
if (linelen <= rem_width) {
out << in.substr(ptr, linelen + 1);
ptr = lineend + 1;
indented = 0;
} else {
size_t finalspace = in.find_last_of(" \n", ptr + rem_width);
if (finalspace == std::string::npos || finalspace < ptr) {
// No place to break; just include the entire word and move on
finalspace = in.find_first_of("\n ", ptr);
if (finalspace == std::string::npos) {
// End of the string, just add it and break
out << in.substr(ptr);
break;
}
}
out << in.substr(ptr, finalspace - ptr) << "\n";
if (in[finalspace] == '\n') {
indented = 0;
} else if (indent) {
out << std::string(indent, ' ');
indented = indent;
}
ptr = finalspace + 1;
}
}
return out.str();
}
std::string i64tostr(int64_t n)
{
return strprintf("%d", n);
}
std::string itostr(int n)
{
return strprintf("%d", n);
}
int64_t atoi64(const char* psz)
{
#ifdef _MSC_VER
return _atoi64(psz);
#else
return strtoll(psz, nullptr, 10);
#endif
}
int64_t atoi64(const std::string& str)
{
#ifdef _MSC_VER
return _atoi64(str.c_str());
#else
return strtoll(str.c_str(), nullptr, 10);
#endif
}
int atoi(const std::string& str)
{
return atoi(str.c_str());
}
/** Upper bound for mantissa.
* 10^18-1 is the largest arbitrary decimal that will fit in a signed 64-bit integer.
* Larger integers cannot consist of arbitrary combinations of 0-9:
*
* 999999999999999999 1^18-1
* 9223372036854775807 (1<<63)-1 (max int64_t)
* 9999999999999999999 1^19-1 (would overflow)
*/
static const int64_t UPPER_BOUND = 1000000000000000000LL - 1LL;
/** Helper function for ParseFixedPoint */
static inline bool ProcessMantissaDigit(char ch, int64_t &mantissa, int &mantissa_tzeros)
{
if(ch == '0')
++mantissa_tzeros;
else {
for (int i=0; i<=mantissa_tzeros; ++i) {
if (mantissa > (UPPER_BOUND / 10LL))
return false; /* overflow */
mantissa *= 10;
}
mantissa += ch - '0';
mantissa_tzeros = 0;
}
return true;
}
bool ParseFixedPoint(const std::string &val, int decimals, int64_t *amount_out)
{
int64_t mantissa = 0;
int64_t exponent = 0;
int mantissa_tzeros = 0;
bool mantissa_sign = false;
bool exponent_sign = false;
int ptr = 0;
int end = val.size();
int point_ofs = 0;
if (ptr < end && val[ptr] == '-') {
mantissa_sign = true;
++ptr;
}
if (ptr < end)
{
if (val[ptr] == '0') {
/* pass single 0 */
++ptr;
} else if (val[ptr] >= '1' && val[ptr] <= '9') {
while (ptr < end && IsDigit(val[ptr])) {
if (!ProcessMantissaDigit(val[ptr], mantissa, mantissa_tzeros))
return false; /* overflow */
++ptr;
}
} else return false; /* missing expected digit */
} else return false; /* empty string or loose '-' */
if (ptr < end && val[ptr] == '.')
{
++ptr;
if (ptr < end && IsDigit(val[ptr]))
{
while (ptr < end && IsDigit(val[ptr])) {
if (!ProcessMantissaDigit(val[ptr], mantissa, mantissa_tzeros))
return false; /* overflow */
++ptr;
++point_ofs;
}
} else return false; /* missing expected digit */
}
if (ptr < end && (val[ptr] == 'e' || val[ptr] == 'E'))
{
++ptr;
if (ptr < end && val[ptr] == '+')
++ptr;
else if (ptr < end && val[ptr] == '-') {
exponent_sign = true;
++ptr;
}
if (ptr < end && IsDigit(val[ptr])) {
while (ptr < end && IsDigit(val[ptr])) {
if (exponent > (UPPER_BOUND / 10LL))
return false; /* overflow */
exponent = exponent * 10 + val[ptr] - '0';
++ptr;
}
} else return false; /* missing expected digit */
}
if (ptr != end)
return false; /* trailing garbage */
/* finalize exponent */
if (exponent_sign)
exponent = -exponent;
exponent = exponent - point_ofs + mantissa_tzeros;
/* finalize mantissa */
if (mantissa_sign)
mantissa = -mantissa;
/* convert to one 64-bit fixed-point value */
exponent += decimals;
if (exponent < 0)
return false; /* cannot represent values smaller than 10^-decimals */
if (exponent >= 18)
return false; /* cannot represent values larger than or equal to 10^(18-decimals) */
for (int i=0; i < exponent; ++i) {
if (mantissa > (UPPER_BOUND / 10LL) || mantissa < -(UPPER_BOUND / 10LL))
return false; /* overflow */
mantissa *= 10;
}
if (mantissa > UPPER_BOUND || mantissa < -UPPER_BOUND)
return false; /* overflow */
if (amount_out)
*amount_out = mantissa;
return true;
}
std::string ToLower(const std::string& str)
{
std::string r;
for (auto ch : str) r += ToLower((unsigned char)ch);
return r;
}
std::string ToUpper(const std::string& str)
{
std::string r;
for (auto ch : str) r += ToUpper((unsigned char)ch);
return r;
}
std::string Capitalize(std::string str)
{
if (str.empty()) return str;
str[0] = ToUpper(str.front());
return str;
}
std::string HexStr(const Span<const uint8_t> s)
{
std::string rv;
static constexpr char hexmap[16] = { '0', '1', '2', '3', '4', '5', '6', '7',
'8', '9', 'a', 'b', 'c', 'd', 'e', 'f' };
rv.reserve(s.size() * 2);
for (uint8_t v: s) {
rv.push_back(hexmap[v >> 4]);
rv.push_back(hexmap[v & 15]);
}
return rv;
}

302
src/util/strencodings.h Normal file
View File

@@ -0,0 +1,302 @@
// Copyright (c) 2009-2010 Satoshi Nakamoto
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
/**
* Utilities for converting data from/to strings.
*/
#ifndef BITCOIN_UTIL_STRENCODINGS_H
#define BITCOIN_UTIL_STRENCODINGS_H
#include "attributes.h"
#include "span.h"
#include <cstdint>
#include <iterator>
#include <string>
#include <vector>
#define BEGIN(a) ((char*)&(a))
#define END(a) ((char*)&((&(a))[1]))
#define UBEGIN(a) ((unsigned char*)&(a))
#define UEND(a) ((unsigned char*)&((&(a))[1]))
#define ARRAYLEN(array) (sizeof(array)/sizeof((array)[0]))
/** This is needed because the foreach macro can't get over the comma in pair<t1, t2> */
#define PAIRTYPE(t1, t2) std::pair<t1, t2>
/** Used by SanitizeString() */
enum SafeChars
{
SAFE_CHARS_DEFAULT, //!< The full set of allowed chars
SAFE_CHARS_UA_COMMENT, //!< BIP-0014 subset
SAFE_CHARS_FILENAME, //!< Chars allowed in filenames
SAFE_CHARS_URI, //!< Chars allowed in URIs (RFC 3986)
};
std::string SanitizeFilename(const std::string& str);
/**
* Remove unsafe chars. Safe chars chosen to allow simple messages/URLs/email
* addresses, but avoid anything even possibly remotely dangerous like & or >
* @param[in] str The string to sanitize
* @param[in] rule The set of safe chars to choose (default: least restrictive)
* @return A new string without unsafe chars
*/
std::string SanitizeString(const std::string& str, int rule = SAFE_CHARS_DEFAULT);
std::string HexInt(uint32_t val);
uint32_t ParseHexToUInt32(const std::string& str);
std::vector<unsigned char> ParseHex(const char* psz);
std::vector<unsigned char> ParseHex(const std::string& str);
signed char HexDigit(char c);
/* Returns true if each character in str is a hex character, and has an even
* number of hex digits.*/
bool IsHex(const std::string& str);
/**
* Return true if the string is a hex number, optionally prefixed with "0x"
*/
bool IsHexNumber(const std::string& str);
std::vector<unsigned char> DecodeBase64(const char* p, bool* pf_invalid = nullptr);
std::string DecodeBase64(const std::string& str, bool* pf_invalid = nullptr);
std::string EncodeBase64(Span<const unsigned char> input);
std::string EncodeBase64(const unsigned char* pch, size_t len);
std::string EncodeBase64(const std::string& str);
std::vector<unsigned char> DecodeBase32(const char* p, bool* pf_invalid = nullptr);
std::string DecodeBase32(const std::string& str, bool* pf_invalid = nullptr);
/**
* Base32 encode.
* If `pad` is true, then the output will be padded with '=' so that its length
* is a multiple of 8.
*/
std::string EncodeBase32(Span<const unsigned char> input, bool pad = true);
/**
* Base32 encode.
* If `pad` is true, then the output will be padded with '=' so that its length
* is a multiple of 8.
*/
std::string EncodeBase32(const std::string& str, bool pad = true);
void SplitHostPort(std::string in, int& portOut, std::string& hostOut);
std::string i64tostr(int64_t n);
std::string itostr(int n);
int64_t atoi64(const char* psz);
int64_t atoi64(const std::string& str);
int atoi(const std::string& str);
/**
* Tests if the given character is a decimal digit.
* @param[in] c character to test
* @return true if the argument is a decimal digit; otherwise false.
*/
constexpr bool IsDigit(char c)
{
return c >= '0' && c <= '9';
}
/**
* Tests if the given character is a whitespace character. The whitespace characters
* are: space, form-feed ('\f'), newline ('\n'), carriage return ('\r'), horizontal
* tab ('\t'), and vertical tab ('\v').
*
* This function is locale independent. Under the C locale this function gives the
* same result as std::isspace.
*
* @param[in] c character to test
* @return true if the argument is a whitespace character; otherwise false
*/
constexpr inline bool IsSpace(char c) noexcept {
return c == ' ' || c == '\f' || c == '\n' || c == '\r' || c == '\t' || c == '\v';
}
/**
* Convert string to signed 32-bit integer with strict parse error feedback.
* @returns true if the entire string could be parsed as valid integer,
* false if not the entire string could be parsed or when overflow or underflow occurred.
*/
bool ParseInt32(const std::string& str, int32_t *out);
/**
* Convert string to signed 64-bit integer with strict parse error feedback.
* @returns true if the entire string could be parsed as valid integer,
* false if not the entire string could be parsed or when overflow or underflow occurred.
*/
bool ParseInt64(const std::string& str, int64_t *out);
/**
* Convert decimal string to unsigned 8-bit integer with strict parse error feedback.
* @returns true if the entire string could be parsed as valid integer,
* false if not the entire string could be parsed or when overflow or underflow occurred.
*/
bool ParseUInt8(const std::string& str, uint8_t *out);
/**
* Convert decimal string to unsigned 32-bit integer with strict parse error feedback.
* @returns true if the entire string could be parsed as valid integer,
* false if not the entire string could be parsed or when overflow or underflow occurred.
*/
bool ParseUInt32(const std::string& str, uint32_t *out);
/**
* Convert decimal string to unsigned 64-bit integer with strict parse error feedback.
* @returns true if the entire string could be parsed as valid integer,
* false if not the entire string could be parsed or when overflow or underflow occurred.
*/
bool ParseUInt64(const std::string& str, uint64_t *out);
/**
* Convert string to double with strict parse error feedback.
* @returns true if the entire string could be parsed as valid double,
* false if not the entire string could be parsed or when overflow or underflow occurred.
*/
bool ParseDouble(const std::string& str, double *out);
template<typename T>
std::string HexStr(const T itbegin, const T itend, bool fSpaces=false)
{
std::string rv;
static const char hexmap[16] = { '0', '1', '2', '3', '4', '5', '6', '7',
'8', '9', 'a', 'b', 'c', 'd', 'e', 'f' };
rv.reserve((itend-itbegin)*3);
for(T it = itbegin; it < itend; ++it)
{
unsigned char val = (unsigned char)(*it);
if(fSpaces && it != itbegin)
rv.push_back(' ');
rv.push_back(hexmap[val>>4]);
rv.push_back(hexmap[val&15]);
}
return rv;
}
template<typename T>
inline std::string HexStr(const T& vch, bool fSpaces=false)
{
return HexStr(vch.begin(), vch.end(), fSpaces);
}
/**
* Convert a span of bytes to a lower-case hexadecimal string.
*/
std::string HexStr(const Span<const uint8_t> s);
inline std::string HexStr(const Span<const char> s) { return HexStr(MakeUCharSpan(s)); }
/**
* Format a paragraph of text to a fixed width, adding spaces for
* indentation to any added line.
*/
std::string FormatParagraph(const std::string& in, size_t width = 79, size_t indent = 0);
/**
* Timing-attack-resistant comparison.
* Takes time proportional to length
* of first argument.
*/
template <typename T>
bool TimingResistantEqual(const T& a, const T& b)
{
if (b.size() == 0) return a.size() == 0;
size_t accumulator = a.size() ^ b.size();
for (size_t i = 0; i < a.size(); i++)
accumulator |= a[i] ^ b[i%b.size()];
return accumulator == 0;
}
/** Parse number as fixed point according to JSON number syntax.
* See http://json.org/number.gif
* @returns true on success, false on error.
* @note The result must be in the range (-10^18,10^18), otherwise an overflow error will trigger.
*/
bool ParseFixedPoint(const std::string &val, int decimals, int64_t *amount_out);
/** Convert from one power-of-2 number base to another. */
template<int frombits, int tobits, bool pad, typename O, typename I>
bool ConvertBits(const O& outfn, I it, I end) {
size_t acc = 0;
size_t bits = 0;
constexpr size_t maxv = (1 << tobits) - 1;
constexpr size_t max_acc = (1 << (frombits + tobits - 1)) - 1;
while (it != end) {
acc = ((acc << frombits) | *it) & max_acc;
bits += frombits;
while (bits >= tobits) {
bits -= tobits;
outfn((acc >> bits) & maxv);
}
++it;
}
if (pad) {
if (bits) outfn((acc << (tobits - bits)) & maxv);
} else if (bits >= frombits || ((acc << (tobits - bits)) & maxv)) {
return false;
}
return true;
}
/**
* Converts the given character to its lowercase equivalent.
* This function is locale independent. It only converts uppercase
* characters in the standard 7-bit ASCII range.
* This is a feature, not a limitation.
*
* @param[in] c the character to convert to lowercase.
* @return the lowercase equivalent of c; or the argument
* if no conversion is possible.
*/
constexpr char ToLower(char c)
{
return (c >= 'A' && c <= 'Z' ? (c - 'A') + 'a' : c);
}
/**
* Returns the lowercase equivalent of the given string.
* This function is locale independent. It only converts uppercase
* characters in the standard 7-bit ASCII range.
* This is a feature, not a limitation.
*
* @param[in] str the string to convert to lowercase.
* @returns lowercased equivalent of str
*/
std::string ToLower(const std::string& str);
/**
* Converts the given character to its uppercase equivalent.
* This function is locale independent. It only converts lowercase
* characters in the standard 7-bit ASCII range.
* This is a feature, not a limitation.
*
* @param[in] c the character to convert to uppercase.
* @return the uppercase equivalent of c; or the argument
* if no conversion is possible.
*/
constexpr char ToUpper(char c)
{
return (c >= 'a' && c <= 'z' ? (c - 'a') + 'A' : c);
}
/**
* Returns the uppercase equivalent of the given string.
* This function is locale independent. It only converts lowercase
* characters in the standard 7-bit ASCII range.
* This is a feature, not a limitation.
*
* @param[in] str the string to convert to uppercase.
* @returns UPPERCASED EQUIVALENT OF str
*/
std::string ToUpper(const std::string& str);
/**
* Capitalizes the first character of the given string.
* This function is locale independent. It only converts lowercase
* characters in the standard 7-bit ASCII range.
* This is a feature, not a limitation.
*
* @param[in] str the string to capitalize.
* @returns string with the first letter capitalized.
*/
std::string Capitalize(std::string str);
#endif // BITCOIN_UTIL_STRENCODINGS_H

6
src/util/string.cpp Normal file
View File

@@ -0,0 +1,6 @@
// Copyright (c) 2019 The Bitcoin Core developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
#include "util/string.h"

99
src/util/string.h Normal file
View File

@@ -0,0 +1,99 @@
// Copyright (c) 2019-2020 The Bitcoin Core developers
// Copyright (c) 2016-2022 The Hush developers
// Distributed under the GPLv3 software license, see the accompanying
// file COPYING or https://www.gnu.org/licenses/gpl-3.0.en.html
#ifndef BITCOIN_UTIL_STRING_H
#define BITCOIN_UTIL_STRING_H
#include "attributes.h"
#include <algorithm>
#include <array>
#include <cstring>
#include <locale>
#include <sstream>
#include <string>
#include <vector>
[[nodiscard]] inline std::string TrimString(const std::string& str, const std::string& pattern = " \f\n\r\t\v")
{
std::string::size_type front = str.find_first_not_of(pattern);
if (front == std::string::npos) {
return std::string();
}
std::string::size_type end = str.find_last_not_of(pattern);
return str.substr(front, end - front + 1);
}
[[nodiscard]] inline std::string RemovePrefix(const std::string& str, const std::string& prefix)
{
if (str.substr(0, prefix.size()) == prefix) {
return str.substr(prefix.size());
}
return str;
}
/**
* Join a list of items
*
* @param list The list to join
* @param separator The separator
* @param unary_op Apply this operator to each item in the list
*/
template <typename T, typename BaseType, typename UnaryOp>
auto Join(const std::vector<T>& list, const BaseType& separator, UnaryOp unary_op)
-> decltype(unary_op(list.at(0)))
{
decltype(unary_op(list.at(0))) ret;
for (size_t i = 0; i < list.size(); ++i) {
if (i > 0) ret += separator;
ret += unary_op(list.at(i));
}
return ret;
}
template <typename T>
T Join(const std::vector<T>& list, const T& separator)
{
return Join(list, separator, [](const T& i) { return i; });
}
// Explicit overload needed for c_str arguments, which would otherwise cause a substitution failure in the template above.
inline std::string Join(const std::vector<std::string>& list, const std::string& separator)
{
return Join<std::string>(list, separator);
}
/**
* Check if a string does not contain any embedded NUL (\0) characters
*/
[[nodiscard]] inline bool ValidAsCString(const std::string& str) noexcept
{
return str.size() == strlen(str.c_str());
}
/**
* Locale-independent version of std::to_string
*/
template <typename T>
std::string ToString(const T& t)
{
std::ostringstream oss;
oss.imbue(std::locale::classic());
oss << t;
return oss.str();
}
/**
* Check whether a container begins with the given prefix.
*/
template <typename T1, size_t PREFIX_LEN>
[[nodiscard]] inline bool HasPrefix(const T1& obj,
const std::array<uint8_t, PREFIX_LEN>& prefix)
{
return obj.size() >= PREFIX_LEN &&
std::equal(std::begin(prefix), std::end(prefix), std::begin(obj));
}
#endif // BITCOIN_UTIL_STRENCODINGS_H