> On 24 Jun 2020, at 11:27, Edward Welbourne <[email protected]> wrote: > > Marc Mutz (24 June 2020 09:32) wrote >> My Qt 5-era changes to QLatin1String (adding QL1S::arg(), enabling >> QL1S as a type for date/time formats, overloading QtJSON functions for >> QL1S) have shown how dramatic the effect of 8-bit string values >> without constructing a QString first really is, without actually >> limiting functionality. > > wait ... "enabling QL1S as a type for date/time formats" ? > A quick search for "QLatin1String" in qdatetime.h and qlocale.h shows a > total of zero matches. We have QStringView overloads, but not QL1S.
I certainly don’t want to add more QL1S overloads to existing methods. If using QAnyStringView makes sense in that API, they could come implicitly, but I really want to deprecate QL1S over time. Cheers, Lars > > Not that I have any objection to adding such overloads, but they can't be > claimed as qt5-era wins of viewification. > > Eddy. > _______________________________________________ > Development mailing list > [email protected] > https://lists.qt-project.org/listinfo/development _______________________________________________ Development mailing list [email protected] https://lists.qt-project.org/listinfo/development
