7

I have another question yet. If I had a std::wstring looking like this:

ドイツ語で検索していてこちらのサイトにたどり着きました。

How could I possibly get it to be URL-Encoded (%nn, n = 0-9, a-f) to:

%E3%83%89%E3%82%A4%E3%83%84%E8%AA%9E%E3%81%A7%E6%A4%9C%E7%B4%A2%E3%81%97%E3%81%A6%E3%81%84%E3%81%A6%E3%81%93%E3%81%A1%E3%82%89%E3%81%AE%E3%82%B5%E3%82%A4%E3%83%88%E3%81%AB%E3%81%9F%E3%81%A9%E3%82%8A%E7%9D%80%E3%81%8D%E3%81%BE%E3%81%97%E3%81%9F%E3%80%82

... and also HTML-Encoded (&#nnn(nn);, n = 0-9(?)) to:

ドイツ語で検索していてこちらのサイトにたどり着きました。

Please help me as I am totally lost right now and don't even know where to start. By the way, performance isn't much important to me right now.

Thanks in advance!

3
  • 1
    Do you need a library or you want to code it yourself? Check this article with a sample. Commented Jul 21, 2010 at 14:19
  • Are you bound to a specific operating system or should the program be platform-independent? The encoding used by std::wstring varies between operating systems. Commented Jul 21, 2010 at 15:43
  • Kirill: Oh, I don't care. I might try to do it myself instead of using a possibly bloated library :/ Philipp: I don't need it to be platform-independent. It's okay if it works on WIN32 only. Commented Jul 21, 2010 at 15:46

5 Answers 5

4

Here is an example which shows two methods, one based on the Qt library and one based on the ICU library. Both should be fairly platform-independent:

#include <iostream>
#include <sstream>
#include <iomanip>
#include <stdexcept>

#include <boost/scoped_array.hpp>

#include <QtCore/QString>
#include <QtCore/QUrl>
#include <QtCore/QVector>

#include <unicode/utypes.h>
#include <unicode/ustring.h>
#include <unicode/unistr.h>
#include <unicode/schriter.h>

void encodeQt() {
  const QString str = QString::fromWCharArray(L"ドイツ語で検索していてこちらのサイトにたどり着きました。");
  const QUrl url = str;
  std::cout << "URL encoded: " << url.toEncoded().constData() << std::endl;
  typedef QVector<uint> CodePointVector;
  const CodePointVector codePoints = str.toUcs4();
  std::stringstream htmlEncoded;
  for (CodePointVector::const_iterator it = codePoints.constBegin(); it != codePoints.constEnd(); ++it) {
    htmlEncoded << "&#" << *it << ';';
  }
  std::cout << "HTML encoded: " << htmlEncoded.str() << std::endl;
}

void encodeICU() {
  const std::wstring cppString = L"ドイツ語で検索していてこちらのサイトにたどり着きました。";
  int bufSize = cppString.length() * 2;
  boost::scoped_array<UChar> strBuffer(new UChar[bufSize]);
  int size = 0;
  UErrorCode error = U_ZERO_ERROR;
  u_strFromWCS(strBuffer.get(), bufSize, &size, cppString.data(), cppString.length(), &error);
  if (error) return;
  const UnicodeString str(strBuffer.get(), size);
  bufSize = str.length() * 4;
  boost::scoped_array<char> buffer(new char[bufSize]);
  u_strToUTF8(buffer.get(), bufSize, &size, str.getBuffer(), str.length(), &error);
  if (error) return;
  const std::string urlUtf8(buffer.get(), size);
  std::stringstream urlEncoded;
  urlEncoded << std::hex << std::setfill('0');
  for (std::string::const_iterator it = urlUtf8.begin(); it != urlUtf8.end(); ++it) {
    urlEncoded << '%' << std::setw(2) << static_cast<unsigned int>(static_cast<unsigned char>(*it));
  }
  std::cout << "URL encoded: " << urlEncoded.str() << std::endl;
  std::stringstream htmlEncoded;
  StringCharacterIterator it = str;
  while (it.hasNext()) {
    const UChar32 pt = it.next32PostInc();
    htmlEncoded << "&#" << pt << ';';
  }
  std::cout << "HTML encoded: " << htmlEncoded.str() << std::endl;
}


int main() {
  encodeQt();
  encodeICU();
}
Sign up to request clarification or add additional context in comments.

1 Comment

Don't get me wrong! I appreciate your informative contribution and your time! But this looks way too bloated for my needs... :/
1

You see, before you can convert a char to a URL escape sequence, you have to convert your wstring* into ISO-Latin charset which is what is used for URLs. ICU could be a good place to start, where you can pass your wstring to it and get a ISO-Lantin sequence. Then, simply iterate through the resulting chars and convert them to the escape senquence:

std::stringstream URL;
URL << std::hex;
for(auto it = myWString.begin(); it != myWString.end(); ++it)
   URL << '%' << std::setfill('0') << std::setw(2) << (int)*it;

Take a look here for more info in how to format the string.

* I'm assuming that your wstring is a UTF-16, which usually is the case, although you didn't specify

This might help also.

4 Comments

Thanks for your time, but the example doesn't work well together with Unicode for me (results look like %n, %nn and even %nnn as well as %nnnn for some characters) :(
@Oliver sry, forgot about the width and fill guys. I've included in the code above. And like I said, you need to convert to ISO-Latin with ICU before using the sstream.
You cannot convert Chinese characters to a Latin encoding. Using UTF-8 is the common choice.
@Phillip Sure, it is most likely a mistake or wrongly documented, but URLs by default use ISO-Lantin (which is a subset of UTF-8) instead of UTF-8. However, ignoring that and just using UTF-8 might still be safe.
0

Here's a version that converts from UTF-16 (wchar) to hex-encoded UTF-8 using the Win32-specific WideCharToMultiByte() function.

#include <string>
#include <iostream>
#include <stdio.h>
#include <windows.h>


std::string wstring_to_utf8_hex(const std::wstring &input)
{
  std::string output;
  int cbNeeded = WideCharToMultiByte(CP_UTF8, 0, input.c_str(), -1, NULL, 0, NULL, NULL);
  if (cbNeeded > 0) {
    char *utf8 = new char[cbNeeded];
    if (WideCharToMultiByte(CP_UTF8, 0, input.c_str(), -1, utf8, cbNeeded, NULL, NULL) != 0) {
      for (char *p = utf8; *p; *p++) {
        char onehex[5];
        _snprintf(onehex, sizeof(onehex), "%%%02.2X", (unsigned char)*p);
        output.append(onehex);
      }
    }
    delete[] utf8;
  }
  return output;
}


int main(int, char*[])
{
  std::wstring ja = L"ドイツ語で検索していてこちらのサイトにたどり着きました。";
  std::cout << "result=" << wstring_to_utf8_hex(ja) << std::endl;
  return 0;
}

To go the other way, you'll need to use some parsing to decode the hex values into a UTF-8 buffer, and then call the complimentary MultiByteToWideChar() to get it back into a wchar array.

#include <string>
#include <iostream>
#include <ctype.h>
#include <stdio.h>
#include <stdlib.h>
#include <windows.h>

std::string unhexlify(const std::string &input)
{
  std::string output;
  for (const char *p = input.c_str(); *p; ) {
    if (p[0] == '%' && isxdigit(p[1]) && isxdigit(p[2])) {
      int ch = (isdigit(p[1]) ? p[1] - '0' : toupper(p[1]) - 'A' + 10) * 16 + 
               (isdigit(p[2]) ? p[2] - '0' : toupper(p[2]) - 'A' + 10);
      output.push_back((char)ch);
      p += 3;
    } else if (p[0] == '%' && p[1] == '#' && isdigit(p[2])) {
      int ch = atoi(p + 2);
      output.push_back((char)ch);
      p += 2;
      while (*p && isdigit(*p)) p++;
      if (*p == ';') p++;
    } else {
      output.push_back(*p++);
    }
  }
  return output;
}


std::wstring utf8_hex_to_wstring(const std::string &input)
{
  std::wstring output;
  std::string utf8 = unhexlify(input);
  int cchNeeded = MultiByteToWideChar(CP_UTF8, 0, utf8.c_str(), -1, NULL, 0);
  if (cchNeeded > 0) {
    wchar_t *widebuf = new wchar_t[cchNeeded];
    if (MultiByteToWideChar(CP_UTF8, 0, utf8.c_str(), -1, widebuf, cchNeeded) != 0) {
      output = widebuf;
    }
    delete[] widebuf;
  }
  return output;
}

int main(int, char*[])
{
  std::wstring ja = L"ドイツ語で検索していてこちらのサイトにたどり着きました。";
  std::string hex = "%E3%83%89%E3%82%A4%E3%83%84%E8%AA%9E%E3%81%A7%E6%A4%9C%E7%B4%A2%E3%81%97%E3%81%A6%E3%81%84%E3%81%A6%E3%81%93%E3%81%A1%E3%82%89%E3%81%AE%E3%82%B5%E3%82%A4%E3%83%88%E3%81%AB%E3%81%9F%E3%81%A9%E3%82%8A%E7%9D%80%E3%81%8D%E3%81%BE%E3%81%97%E3%81%9F%E3%80%82";
  std::wstring newja = utf8_hex_to_wstring(hex);
  std::cout << "match?=" << (newja == ja ? "yes" : "no") << std::endl;
  return 0;
}

Comments

0

First, convert to UTF-8. Then, normal URL/HTML encode would do the right thing.

Comments

0

I find in C# it's simple, so I use C++\CLI as wrapper, wrap C# code:

string encodedStr = System.Web.HttpUtility.UrlEncode(inputstr);`

in C++\CLI make a method as __declspec(dllexport) so in C++ can call it, the C++\CLI syntax is:

String^ encodedStr  = System::Web::HttpUtility::UrlEncode(inputStr);`.

this is a tutorial about how to call C++\CLI from C++: How to call a C# library from Native C++ (using C++\CLI and IJW)

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.