On sexta-feira, 15 de maio de 2020 10:52:49 PDT Matthew Woehlke wrote: > > Like that, it's just "array of bytes of an arbitrary encoding (or none)". > > There's still a reason to have QByteArray and it'll need to exist in > > networking and file I/O code. That means the string classes, if any, need > > to be convertible to QByteArray anyway. > > I think we can learn from Python 3 here... QByteArray should go the way > of QStringList, i.e. yes, it *should* be a QVector<byte>. Like > QVector<QString>, it might (should) have additional methods, such as > explicit conversion to/from QString (a la Python's encode/decode), but > it should *not* have string-like manipulation (e.g. toUpper).
Those are all Qt 7 work. We can deprecate those methods in Qt 6, but not remove them in 6.0. Python's bstr still behaves string-like and has methods like QByteArray::indexOf(const char *). QVector has no such methods. But since we do have QListSpecialMethods, we can add inject them into QVector too. > >> So, assuming the premise that QByteArray should not be string-ish > >> anymore, what do we want to have as the result type of QString::toUtf8() > >> and QString::toLatin1()? Do we really want mere bytes? > > Yes. Maybe. Again, this is how Python 3 works. > > It might make sense to have a QUtf8String class, but that should be > distinct from, and not implicitly constructible from, QByteArray a.k.a. > QVector<byte>. (Implicit conversion *to* QByteArray might be okay.) > > (BTW, is 'byte' QByte or std::byte? Can we possibly achieve the latter?) There's no QByte and we shouldn't have that type. std::byte is an enum around the actual byte type (unsigned char) and char is also a byte. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel System Software Products _______________________________________________ Development mailing list Development@qt-project.org https://lists.qt-project.org/listinfo/development