无损转换 std::string 和 QByteArray 的正确方法
什么是在 std::string 和 QByteArray 之间无损转换的正确方法...主要是为了处理二进制数据?
What is the correct way to convert losslessly between std::string and QByteArray... mostly for the purpose of handling binary data?
我正在使用:
QByteArray qba = QString::fromStdString(stdString).toAscii();
和
QString(qba).toStdString();
但我想检查一下这是否真的正确.
but I wanted to check whether this is actually correct.
推荐答案
对于二进制数据，您的解决方案是有问题的，因为非 ASCII 字符会被 QString:: 转换为 .对于 '?'toAscii()QString 的内部表示，还有不必要的 UTF-16 转换开销.正如您可能已经猜到的那样，QString 仅应在数据是文本而非二进制数据时使用.
For binary data your solution is problematic since non-ASCII characters would be converted to '?' by QString::toAscii(). There is also the unnecessary overhead of UTF-16 conversion for the internal representation of QString. As you might have guessed, QString should only be used if the data is textual, not binary.
QByteArray 和 std::string 都有原始数据(C 字符串 + 长度)的构造函数，也有转换为 C 字符串 + 长度的构造函数.因此您可以使用它们进行转换:
Both QByteArray and std::string have constructors for raw data (C-string + length) and also a conversion to C-string + length. So you can use them for conversion:
// std::string => QByteArray
QByteArray byteArray(stdString.c_str(), stdString.length());
// QByteArray => std::string
std::string stdString(byteArray.constData(), byteArray.length());
它们都是二进制安全的，这意味着字符串可能包含 '' 字符并且不会被截断.数据也不会被触及(没有 UTF 转换)，所以这种转换是无损"的.
They are both binary-safe, meaning that the string may contain '' characters and doesn't get truncated. The data also doesn't get touched (there is no UTF conversion), so this conversion is "lossless".
确保使用长度作为第二个参数的构造函数(对于 QByteArray 和 std::string)，因为大多数其他构造函数会在之前截断数据第一次出现零.
Make sure to use the constructors with the length as the second argument (for both QByteArray and std::string), as most other constructors will truncate the data before the first occurrence of a zero.
相关文章