我有一些* .bz2文件，其中包含* .csv文本文件。
我需要处理它 - 解压缩它并改变它的编码。
现在我有了代码，解压缩此文件并通过将原始文件读取到另一个文件来更改* .csv文件编码。但是这一刻我“得到了情况” - * .csv文件很大（5 GB），但服务器只有8 GB。 IE，如果我通过将文件编码复制到另一个来更改文件编码，我将不会有它的硬盘空间。
因此，我需要“动态”更改文件的编码。
目前，我有两段代码：
启封
bool NetIO::unBZ2(string PathToBZ2, string PathToCSV) {
try {
boost::locale::generator gen;
ofstream outFile("c:/temp/sqlShare/qqq.csv", std::ofstream::out | std::ofstream::binary);
ifstream file(PathToBZ2, ios_base::in | ios_base::binary);
filtering_streambuf<input> in;
in.push(bzip2_decompressor());
in.push(file);
boost::iostreams::copy(in, outFile);
file.close();
outFile.close();
}
catch (const bzip2_error& exception) {
int error = exception.error();
if (error == boost::iostreams::bzip2::data_error) {
// compressed data stream is corrupted
cout << "compressed data stream is corrupted";
}
else if (error == boost::iostreams::bzip2::data_error_magic)
{
// compressed data stream does not begin with the 'magic' sequence 'B' 'Z' 'h'
cout << "compressed data stream does not begin with the 'magic' sequence 'B' 'Z' 'h'";
}
else if (error == boost::iostreams::bzip2::config_error) {
// libbzip2 has been improperly configured for the current platform
cout << "libbzip2 has been improperly configured for the current platform";
}
return false;
}
return true;
}
和用于更改文件编码的代码
bool NetIO::replaceLineBreaks(const string& inFilePath, const string& searchStr, const string& replacement) {
try {
boost::locale::generator gen;
ifstream iss(inFilePath.c_str(), std::ios::in | std::ios::binary);
std::locale lru = gen("ru_RU.CP1251");
std::locale lru2 = gen("ru_RU.UTF-8");
iss.imbue(lru2);
//string tempFilePath2 = inFilePath + ".tmp3";
string tempFilePath2 = "d:/temp/qqq.tmp3";
ofstream os(tempFilePath2, std::ios::out | std::ios::binary);
os.imbue(lru);
const int buffSize= 500000;
char qqq[buffSize];
//wchar_t wqqq[buffSize];
while (iss) {
iss.read(qqq, buffSize);
// MultiByteToWideChar(CP_UTF8, 0, qqq, buffSize, wqqq, buffSize);
os << qqq;
//cnt++;
/*if (cnt > 80) {
break;
}*/
}
os.close();
/*boost::iostreams::mapped_file istm(inFilePath.c_str());
if (!istm.is_open())
return false;
string tempFilePath = inFilePath + ".tmp";
ofstream ostm(tempFilePath.c_str(), std::ios::out | std::ios::binary);
if (!ostm.is_open()) {
istm.close();
return false;
}
boost::regex regexp(searchStr);
ostreambuf_iterator<char> it(ostm);
boost::regex_replace(it, istm.begin(), istm.end(), regexp, replacement, boost::match_default | boost::format_all);
istm.close();
ostm.close();
boost::filesystem::rename(tempFilePath, inFilePath);*/
}
catch (exception ex) {
cout << "Problem in line endings replacer" << std::endl;
cout << ex.what() << std::endl;
cout << "-----------------" << std::endl;
return false;
}
return true;
}
我想了解，如何将这两段代码组合在一起？
据我了解，可能的解决方案可能是创建一个自定义ofstream对象，该对象将接收数据并转换其编码，并使用boost::iostreams::copy将数据传递给它。
如何组合这两段代码？
非常感谢！
答案 0 :(得分：2)
第二个功能中的代码有问题，您可以在哪里执行
iss.read(qqq, buffSize);
os << qqq;
它不仅假设始终读取qqq的完整大小，而且qqq从char(&)[500000]衰减到char*并且将被解释为以NUL结尾的字符串。这至少打破了嵌入式NUL字符。喜欢：
auto bytes_read = iss.read(qqq, buffSize);
os.write(qqq, bytes_read);
注意：
boost::iostreams::copy使用默认的buffersize（如果未指定）已经做了正确的事情#ifndef BOOST_IOSTREAMS_DEFAULT_DEVICE_BUFFER_SIZE # define BOOST_IOSTREAMS_DEFAULT_DEVICE_BUFFER_SIZE 4096 #endif
接下来，循环很容易组合成类似
的东西// input
std::ifstream file(PathToBZ2, std::ios::binary);
bio::filtering_stream<bio::input> in;
in.push(bio::bzip2_decompressor());
in.push(file);
std::locale lru2 = gen("ru_RU.UTF-8");
in.imbue(lru2);
// output
std::ofstream outFile("c:/temp/sqlShare/qqq.csv", std::ios::binary);
std::locale lru = gen("ru_RU.CP1251");
outFile.imbue(lru);
// copy
const int buffSize = 500000;
/*auto totalWritten =*/ boost::iostreams::copy(in, outFile, buffSize);
请注意，这使用的是filtering_stream，而不是filtering_streambuf
使用code_converter<>强制执行所需的转化：
<强> Live On Coliru
#include <boost/iostreams/filtering_stream.hpp>
#include <boost/iostreams/filter/bzip2.hpp>
#include <boost/iostreams/stream.hpp>
#include <boost/iostreams/copy.hpp>
#include <boost/iostreams/code_converter.hpp>
#include <boost/locale.hpp>
#include <iostream>
#include <fstream>
namespace bio = boost::iostreams;
bool unBZ2(std::string PathToBZ2, std::string PathToCSV) {
try {
boost::locale::generator gen;
// input
std::ifstream file(PathToBZ2, std::ios::binary);
bio::filtering_stream<bio::input> in;
in.push(bio::bzip2_decompressor());
in.push(file);
std::locale lru2 = gen("ru_RU.UTF-8");
in.imbue(lru2);
// output
std::ofstream outFile(PathToCSV, std::ios::binary);
std::locale lru = gen("ru_RU.CP1251");
outFile.imbue(lru);
{
bio::code_converter<bio::filtering_stream<bio::input> > win(in);
bio::code_converter<decltype(outFile)> wout(outFile);
win.imbue(lru2);
wout.imbue(lru);
const int buffSize = 500000;
/*auto totalWritten =*/ boost::iostreams::copy(win, wout, buffSize);
}
} catch (const bio::bzip2_error &exception) {
int error = exception.error();
if (error == boost::iostreams::bzip2::data_error) {
// compressed data stream is corrupted
std::cout << "compressed data stream is corrupted";
} else if (error == boost::iostreams::bzip2::data_error_magic) {
// compressed data stream does not begin with the 'magic' sequence 'B' 'Z'
// 'h'
std::cout << "compressed data stream does not begin with the 'magic' sequence " "'B' 'Z' 'h'";
} else if (error == boost::iostreams::bzip2::config_error) {
// libbzip2 has been improperly configured for the current platform
std::cout << "libbzip2 has been improperly configured for the current platform";
}
return false;
}
return true;
}
int main() {
unBZ2("input.txt.bz2", "output.txt");
}
将"Lorem Ipsum".ru翻译成
00000000: cbee f0e5 ec20 e8ef f1f3 ec20 e4ee ebee ..... ..... ....
00000010: f020 f1e8 f220 e0ec e5f2 2c20 f1ee ebf3 . ... ...., ....
00000020: ec20 eff0 e8ec e8f1 20e0 ed20 f1e5 e42e . ...... .. ....
00000030: 20cd e5f6 20e5 f220 eee4 e8ee 20e4 e8f1 ... .. .... ...
00000040: eff3 f2e0 ede4 ee2c 20ef f0e8 20ed e50a ......., ... ...
00000050: f4ee f0e5 edf1 e8e1 f3f1 20e2 eeeb f3ef .......... .....
00000060: f2e0 f2e8 e1f3 f12e 20c5 efe8 f6f3 f0e8 ........ .......
00000070: 20f6 eeec ecf3 ede5 20f1 f3f1 f6e8 efe8 ....... .......
00000080: f220 e5f5 20ef f0ee 2e20 cff0 eee1 ee20 . .. .... .....
00000090: f4e5 f3e3 e0e8 f20a f0e5 f6f2 e5ff f3e5 ................
000000a0: 20f3 f220 eff0 e82c 20ec e5e8 20f2 eeeb .. ..., ... ...
000000b0: ebe8 f220 ecee ebe5 f1f2 e8e5 20e4 e8f1 ... ........ ...
000000c0: eff3 f2e0 ede4 ee20 f6f3 2e20 d1e8 edf2 ....... ... ....
000000d0: 20e1 f0f3 f2e5 20ec e0e7 e8ec 20e5 eef1 ..... ..... ...
000000e0: 20e5 f32c 0ae1 f0f3 f2e5 20e0 ebf2 e5f0 ..,...... .....
000000f0: e020 eced e5f1 e0f0 f7f3 ec20 f6f3 20ec . ......... .. .
00000100: e5eb 2c20 e5e8 20f5 e0f1 20eb f3ef f2e0 .., .. ... .....
00000110: f2f3 ec20 f6ee edf1 e5f6 f2e5 f2f3 e5f0 ... ............
00000120: 2e0a 0acd e520 e0eb e8e8 20ff f3e0 f120 ..... .... ....
00000130: e0f1 f1e5 edf2 e8ee f020 e2e8 f12c 20e5 ......... ..., .
00000140: e820 e2e8 ec20 ecf3 f6e8 f3f1 20ec e5e4 . ... ...... ...
00000150: e8ee f6f0 e8f2 e0f2 e5ec 2c20 eff0 ee20 .........., ...
00000160: e5e0 20e0 e5ff f3e5 20eb e0ee f0e5 e5f2 .. ..... .......
00000170: 0aef f5e0 e5e4 f0f3 ec2e 20c5 eef1 20f1 .......... ... .
00000180: f3ec ee20 e0e4 f5f3 f620 f6ee edf1 e5ff ... ..... ......
00000190: f3e0 f220 e5f5 2c20 e8e4 20f5 e0f1 20eb ... .., .. ... .