premiere-libtorrent/src/escape_string.cpp

642 lines
16 KiB
C++
Raw Normal View History

2004-03-26 12:39:17 +01:00
/*
Copyright (c) 2003-2012, Arvid Norberg
2004-03-26 12:39:17 +01:00
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in
the documentation and/or other materials provided with the distribution.
* Neither the name of the author nor the names of its
contributors may be used to endorse or promote products derived
from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
*/
#include "libtorrent/pch.hpp"
2004-03-26 12:39:17 +01:00
#include <string>
#include <cctype>
#include <algorithm>
#include <boost/limits.hpp>
2009-01-27 20:11:39 +01:00
#include <cstring>
#include <boost/optional.hpp>
#include <boost/array.hpp>
#include <boost/tuple/tuple.hpp>
2004-03-26 12:39:17 +01:00
#include "libtorrent/config.hpp"
2007-09-10 08:12:41 +02:00
#include "libtorrent/assert.hpp"
#include "libtorrent/escape_string.hpp"
#include "libtorrent/parse_url.hpp"
#include "libtorrent/random.hpp"
2007-09-10 08:12:41 +02:00
2009-06-23 03:57:53 +02:00
#ifdef TORRENT_WINDOWS
2009-11-23 00:55:54 +01:00
#ifndef WIN32_LEAN_AND_MEAN
#define WIN32_LEAN_AND_MEAN
#endif
#include <windows.h>
#endif
#include "libtorrent/utf8.hpp"
#include "libtorrent/thread.hpp"
#if TORRENT_USE_ICONV
#include <iconv.h>
#include <locale.h>
#endif
2004-03-26 12:39:17 +01:00
namespace libtorrent
{
// lexical_cast's result depends on the locale. We need
// a well defined result
2012-06-09 06:48:53 +02:00
boost::array<char, 4 + std::numeric_limits<size_type>::digits10> to_string(size_type n)
{
2012-06-09 06:48:53 +02:00
boost::array<char, 4 + std::numeric_limits<size_type>::digits10> ret;
char *p = &ret.back();
*p = '\0';
unsigned_size_type un = n;
if (n < 0) un = -un;
do {
*--p = '0' + un % 10;
un /= 10;
} while (un);
if (n < 0) *--p = '-';
2012-02-16 09:51:14 +01:00
std::memmove(&ret[0], p, &ret.back() - p + 1);
return ret;
}
std::string unescape_string(std::string const& s, error_code& ec)
2004-03-26 12:39:17 +01:00
{
std::string ret;
for (std::string::const_iterator i = s.begin(); i != s.end(); ++i)
{
if(*i == '+')
{
ret += ' ';
2004-03-26 12:39:17 +01:00
}
else if (*i != '%')
{
ret += *i;
}
else
{
++i;
if (i == s.end())
{
ec = errors::invalid_escaped_string;
return ret;
}
2004-03-26 12:39:17 +01:00
int high;
if(*i >= '0' && *i <= '9') high = *i - '0';
else if(*i >= 'A' && *i <= 'F') high = *i + 10 - 'A';
else if(*i >= 'a' && *i <= 'f') high = *i + 10 - 'a';
else
{
ec = errors::invalid_escaped_string;
return ret;
}
2004-03-26 12:39:17 +01:00
++i;
if (i == s.end())
{
ec = errors::invalid_escaped_string;
return ret;
}
2004-03-26 12:39:17 +01:00
int low;
if(*i >= '0' && *i <= '9') low = *i - '0';
else if(*i >= 'A' && *i <= 'F') low = *i + 10 - 'A';
else if(*i >= 'a' && *i <= 'f') low = *i + 10 - 'a';
else
{
ec = errors::invalid_escaped_string;
return ret;
}
2004-03-26 12:39:17 +01:00
ret += char(high * 16 + low);
}
}
return ret;
}
2009-04-04 11:52:25 +02:00
// http://www.ietf.org/rfc/rfc2396.txt
// section 2.3
2009-09-05 03:08:56 +02:00
static const char unreserved_chars[] =
// when determining if a url needs encoding
// % should be ok
"%+"
// reserved
";?:@=&,$/"
2009-09-05 03:08:56 +02:00
// unreserved (special characters) ' excluded,
// since some buggy trackers fail with those
"-_!.~*()"
2009-09-05 03:08:56 +02:00
// unreserved (alphanumerics)
2009-04-04 11:52:25 +02:00
"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"
"0123456789";
2009-09-05 03:08:56 +02:00
2009-04-04 11:52:25 +02:00
static const char hex_chars[] = "0123456789abcdef";
// the offset is used to ignore the first characters in the unreserved_chars table.
static std::string escape_string_impl(const char* str, int len, int offset)
2004-03-26 12:39:17 +01:00
{
TORRENT_ASSERT(str != 0);
TORRENT_ASSERT(len >= 0);
TORRENT_ASSERT(offset >= 0);
2011-02-14 02:59:01 +01:00
TORRENT_ASSERT(offset < int(sizeof(unreserved_chars))-1);
2009-04-04 11:52:25 +02:00
std::string ret;
2004-03-26 12:39:17 +01:00
for (int i = 0; i < len; ++i)
{
2009-09-06 09:24:15 +02:00
if (std::strchr(unreserved_chars+offset, *str) && *str != 0)
2004-03-26 12:39:17 +01:00
{
2009-04-04 11:52:25 +02:00
ret += *str;
2004-03-26 12:39:17 +01:00
}
else
{
2009-04-04 11:52:25 +02:00
ret += '%';
2009-04-13 04:13:17 +02:00
ret += hex_chars[((unsigned char)*str) >> 4];
ret += hex_chars[((unsigned char)*str) & 15];
2004-03-26 12:39:17 +01:00
}
++str;
}
2009-04-04 11:52:25 +02:00
return ret;
2004-03-26 12:39:17 +01:00
}
std::string escape_string(const char* str, int len)
{
return escape_string_impl(str, len, 11);
}
std::string escape_path(const char* str, int len)
{
return escape_string_impl(str, len, 10);
}
2009-09-05 03:08:56 +02:00
bool need_encoding(char const* str, int len)
{
for (int i = 0; i < len; ++i)
{
2009-09-06 09:24:15 +02:00
if (std::strchr(unreserved_chars, *str) == 0 || *str == 0)
return true;
++str;
}
return false;
}
2009-10-30 04:42:29 +01:00
void convert_path_to_posix(std::string& path)
{
for (std::string::iterator i = path.begin()
, end(path.end()); i != end; ++i)
if (*i == '\\') *i = '/';
}
std::string read_until(char const*& str, char delim, char const* end)
{
TORRENT_ASSERT(str <= end);
std::string ret;
while (str != end && *str != delim)
{
ret += *str;
++str;
}
// skip the delimiter as well
while (str != end && *str == delim) ++str;
return ret;
}
std::string maybe_url_encode(std::string const& url)
{
std::string protocol, host, auth, path;
int port;
error_code ec;
boost::tie(protocol, auth, host, port, path) = parse_url_components(url, ec);
if (ec) return url;
// first figure out if this url contains unencoded characters
if (!need_encoding(path.c_str(), path.size()))
return url;
2009-09-10 19:12:17 +02:00
char msg[TORRENT_MAX_PATH*4];
2013-02-26 06:57:29 +01:00
snprintf(msg, sizeof(msg), "%s://%s%s%s%s%s%s", protocol.c_str(), auth.c_str()
, auth.empty()?"":"@", host.c_str()
, port == -1 ? "" : ":"
, port == -1 ? "" : to_string(port).elems
, escape_path(path.c_str(), path.size()).c_str());
return msg;
}
std::string base64encode(const std::string& s)
{
static const char base64_table[] =
{
'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H',
'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P',
'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X',
'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f',
'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n',
'o', 'p', 'q', 'r', 's', 't', 'u', 'v',
'w', 'x', 'y', 'z', '0', '1', '2', '3',
'4', '5', '6', '7', '8', '9', '+', '/'
};
unsigned char inbuf[3];
unsigned char outbuf[4];
std::string ret;
for (std::string::const_iterator i = s.begin(); i != s.end();)
{
// available input is 1,2 or 3 bytes
// since we read 3 bytes at a time at most
int available_input = (std::min)(3, int(s.end()-i));
// clear input buffer
std::fill(inbuf, inbuf+3, 0);
// read a chunk of input into inbuf
std::copy(i, i + available_input, inbuf);
i += available_input;
// encode inbuf to outbuf
outbuf[0] = (inbuf[0] & 0xfc) >> 2;
outbuf[1] = ((inbuf[0] & 0x03) << 4) | ((inbuf [1] & 0xf0) >> 4);
outbuf[2] = ((inbuf[1] & 0x0f) << 2) | ((inbuf [2] & 0xc0) >> 6);
outbuf[3] = inbuf[2] & 0x3f;
// write output
for (int j = 0; j < available_input+1; ++j)
{
ret += base64_table[outbuf[j]];
}
// write pad
for (int j = 0; j < 3 - available_input; ++j)
{
ret += '=';
}
}
return ret;
}
std::string base32encode(std::string const& s)
{
static const char base32_table[] =
{
'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H',
'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P',
'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X',
'Y', 'Z', '2', '3', '4', '5', '6', '7'
};
int input_output_mapping[] = {0, 2, 4, 5, 7, 8};
unsigned char inbuf[5];
unsigned char outbuf[8];
std::string ret;
for (std::string::const_iterator i = s.begin(); i != s.end();)
{
int available_input = (std::min)(5, int(s.end()-i));
// clear input buffer
std::fill(inbuf, inbuf+5, 0);
// read a chunk of input into inbuf
std::copy(i, i + available_input, inbuf);
i += available_input;
// encode inbuf to outbuf
outbuf[0] = (inbuf[0] & 0xf8) >> 3;
outbuf[1] = ((inbuf[0] & 0x07) << 2) | ((inbuf[1] & 0xc0) >> 6);
outbuf[2] = ((inbuf[1] & 0x3e) >> 1);
outbuf[3] = ((inbuf[1] & 0x01) << 4) | ((inbuf[2] & 0xf0) >> 4);
outbuf[4] = ((inbuf[2] & 0x0f) << 1) | ((inbuf[3] & 0x80) >> 7);
outbuf[5] = ((inbuf[3] & 0x7c) >> 2);
outbuf[6] = ((inbuf[3] & 0x03) << 3) | ((inbuf[4] & 0xe0) >> 5);
outbuf[7] = inbuf[4] & 0x1f;
// write output
int num_out = input_output_mapping[available_input];
for (int j = 0; j < num_out; ++j)
{
ret += base32_table[outbuf[j]];
}
// write pad
for (int j = 0; j < 8 - num_out; ++j)
{
ret += '=';
}
}
return ret;
}
std::string base32decode(std::string const& s)
{
unsigned char inbuf[8];
unsigned char outbuf[5];
std::string ret;
for (std::string::const_iterator i = s.begin(); i != s.end();)
{
int available_input = (std::min)(8, int(s.end()-i));
int pad_start = 0;
if (available_input < 8) pad_start = available_input;
// clear input buffer
std::fill(inbuf, inbuf+8, 0);
for (int j = 0; j < available_input; ++j)
{
char in = std::toupper(*i++);
if (in >= 'A' && in <= 'Z')
inbuf[j] = in - 'A';
else if (in >= '2' && in <= '7')
inbuf[j] = in - '2' + ('Z' - 'A') + 1;
else if (in == '=')
{
inbuf[j] = 0;
if (pad_start == 0) pad_start = j;
}
else if (in == '1')
inbuf[j] = 'I' - 'A';
else
return std::string();
TORRENT_ASSERT(inbuf[j] == (inbuf[j] & 0x1f));
}
// decode inbuf to outbuf
outbuf[0] = inbuf[0] << 3;
outbuf[0] |= inbuf[1] >> 2;
outbuf[1] = (inbuf[1] & 0x3) << 6;
outbuf[1] |= inbuf[2] << 1;
outbuf[1] |= (inbuf[3] & 0x10) >> 4;
outbuf[2] = (inbuf[3] & 0x0f) << 4;
outbuf[2] |= (inbuf[4] & 0x1e) >> 1;
outbuf[3] = (inbuf[4] & 0x01) << 7;
outbuf[3] |= (inbuf[5] & 0x1f) << 2;
outbuf[3] |= (inbuf[6] & 0x18) >> 3;
outbuf[4] = (inbuf[6] & 0x07) << 5;
outbuf[4] |= inbuf[7];
int input_output_mapping[] = {5, 1, 1, 2, 2, 3, 4, 4, 5};
int num_out = input_output_mapping[pad_start];
// write output
std::copy(outbuf, outbuf + num_out, std::back_inserter(ret));
}
return ret;
}
std::string url_has_argument(
2010-03-06 04:57:48 +01:00
std::string const& url, std::string argument, std::string::size_type* out_pos)
{
size_t i = url.find('?');
if (i == std::string::npos) return std::string();
++i;
argument += '=';
if (url.compare(i, argument.size(), argument) == 0)
{
size_t pos = i + argument.size();
if (out_pos) *out_pos = pos;
return url.substr(pos, url.find('&', pos) - pos);
}
argument.insert(0, "&");
i = url.find(argument, i);
if (i == std::string::npos) return std::string();
size_t pos = i + argument.size();
if (out_pos) *out_pos = pos;
return url.substr(pos, url.find('&', pos) - pos);
}
TORRENT_EXTRA_EXPORT std::string to_hex(std::string const& s)
{
std::string ret;
for (std::string::const_iterator i = s.begin(); i != s.end(); ++i)
{
2009-04-04 11:52:25 +02:00
ret += hex_chars[((unsigned char)*i) >> 4];
ret += hex_chars[((unsigned char)*i) & 0xf];
}
return ret;
}
TORRENT_EXTRA_EXPORT void to_hex(char const *in, int len, char* out)
2009-04-04 11:52:25 +02:00
{
for (char const* end = in + len; in < end; ++in)
{
*out++ = hex_chars[((unsigned char)*in) >> 4];
*out++ = hex_chars[((unsigned char)*in) & 0xf];
}
*out = '\0';
}
2013-02-24 11:02:26 +01:00
TORRENT_EXTRA_EXPORT int hex_to_int(char in)
2009-04-04 11:52:25 +02:00
{
if (in >= '0' && in <= '9') return int(in) - '0';
if (in >= 'A' && in <= 'F') return int(in) - 'A' + 10;
if (in >= 'a' && in <= 'f') return int(in) - 'a' + 10;
return -1;
}
TORRENT_EXTRA_EXPORT bool is_hex(char const *in, int len)
{
for (char const* end = in + len; in < end; ++in)
{
int t = hex_to_int(*in);
if (t == -1) return false;
}
return true;
}
TORRENT_EXTRA_EXPORT bool from_hex(char const *in, int len, char* out)
2009-04-04 11:52:25 +02:00
{
for (char const* end = in + len; in < end; ++in, ++out)
{
int t = hex_to_int(*in);
if (t == -1) return false;
*out = t << 4;
2009-04-04 23:44:04 +02:00
++in;
2009-04-04 11:52:25 +02:00
t = hex_to_int(*in);
if (t == -1) return false;
*out |= t & 15;
}
return true;
}
2010-02-15 06:49:10 +01:00
#if defined TORRENT_WINDOWS && TORRENT_USE_WSTRING
std::wstring convert_to_wstring(std::string const& s)
{
std::wstring ret;
int result = libtorrent::utf8_wchar(s, ret);
if (result == 0) return ret;
ret.clear();
const char* end = &s[0] + s.size();
for (const char* i = &s[0]; i < end;)
{
wchar_t c = '.';
2012-06-09 06:48:53 +02:00
result = std::mbtowc(&c, i, end - i);
if (result > 0) i += result;
else ++i;
ret += c;
}
return ret;
}
std::string convert_from_wstring(std::wstring const& s)
{
std::string ret;
int result = libtorrent::wchar_utf8(s, ret);
if (result == 0) return ret;
ret.clear();
const wchar_t* end = &s[0] + s.size();
for (const wchar_t* i = &s[0]; i < end;)
{
char c[10];
TORRENT_ASSERT(sizeof(c) >= MB_CUR_MAX);
2012-06-09 06:48:53 +02:00
result = std::wctomb(c, *i);
if (result > 0)
{
i += result;
ret.append(c, result);
}
else
{
++i;
ret += ".";
}
}
return ret;
}
#endif
2011-02-07 18:25:24 +01:00
#if TORRENT_USE_ICONV
std::string iconv_convert_impl(std::string const& s, iconv_t h)
{
std::string ret;
size_t insize = s.size();
size_t outsize = insize * 4;
ret.resize(outsize);
2010-03-04 04:28:18 +01:00
char const* in = s.c_str();
char* out = &ret[0];
2011-07-23 21:30:23 +02:00
// posix has a weird iconv signature. implementations
// differ on what this signature should be, so we use
// a macro to let config.hpp determine it
size_t retval = iconv(h, TORRENT_ICONV_ARG &in, &insize,
&out, &outsize);
if (retval == (size_t)-1) return s;
2010-03-04 04:28:18 +01:00
// if this string has an invalid utf-8 sequence in it, don't touch it
if (insize != 0) return s;
// not sure why this would happen, but it seems to be possible
if (outsize > s.size() * 4) return s;
// outsize is the number of bytes unused of the out-buffer
TORRENT_ASSERT(ret.size() >= outsize);
ret.resize(ret.size() - outsize);
return ret;
}
std::string convert_to_native(std::string const& s)
{
static mutex iconv_mutex;
// only one thread can use this handle at a time
mutex::scoped_lock l(iconv_mutex);
// the empty string represents the local dependent encoding
static iconv_t iconv_handle = iconv_open("", "UTF-8");
if (iconv_handle == iconv_t(-1)) return s;
return iconv_convert_impl(s, iconv_handle);
}
std::string convert_from_native(std::string const& s)
{
static mutex iconv_mutex;
// only one thread can use this handle at a time
mutex::scoped_lock l(iconv_mutex);
// the empty string represents the local dependent encoding
static iconv_t iconv_handle = iconv_open("UTF-8", "");
if (iconv_handle == iconv_t(-1)) return s;
return iconv_convert_impl(s, iconv_handle);
}
2011-02-22 04:09:13 +01:00
2013-01-21 19:54:45 +01:00
#elif defined TORRENT_WINDOWS
std::string convert_to_native(std::string const& s)
{
std::wstring ws;
libtorrent::utf8_wchar(s, ws);
std::string ret;
ret.resize(ws.size() * 4 + 1);
2013-01-21 19:54:45 +01:00
std::size_t size = WideCharToMultiByte(CP_ACP, 0, ws.c_str(), -1, &ret[0], ret.size(), NULL, NULL);
if (size == std::size_t(-1)) return s;
ret.resize(size);
return ret;
}
std::string convert_from_native(std::string const& s)
{
std::wstring ws;
ws.resize(s.size() + 1);
2013-01-21 19:54:45 +01:00
std::size_t size = MultiByteToWideChar(CP_ACP, 0, s.c_str(), -1, &ws[0], ws.size());
if (size == std::size_t(-1)) return s;
ws.resize(size);
std::string ret;
libtorrent::wchar_utf8(ws, ret);
return ret;
}
2011-02-22 04:09:13 +01:00
#elif TORRENT_USE_LOCALE
2011-02-07 18:25:24 +01:00
std::string convert_to_native(std::string const& s)
{
std::wstring ws;
libtorrent::utf8_wchar(s, ws);
std::size_t size = wcstombs(0, ws.c_str(), 0);
if (size == std::size_t(-1)) return s;
std::string ret;
ret.resize(size);
size = wcstombs(&ret[0], ws.c_str(), size + 1);
if (size == std::size_t(-1)) return s;
ret.resize(size);
return ret;
}
std::string convert_from_native(std::string const& s)
{
std::wstring ws;
ws.resize(s.size());
std::size_t size = mbstowcs(&ws[0], s.c_str(), s.size());
if (size == std::size_t(-1)) return s;
std::string ret;
libtorrent::wchar_utf8(ws, ret);
return ret;
}
2011-02-22 04:09:13 +01:00
#endif
2004-03-26 12:39:17 +01:00
}