> On 3 Jul 2020, at 02:49, Thiago Macieira <thiago.macie...@intel.com> wrote: > > On Thursday, 2 July 2020 15:46:24 PDT Tor Arne Vestbø wrote: >> I think I prefer permanently, and then clearly documenting that the fallback >> conversion to a string for non-listed types may be lossy, teaching the user >> to either encode the data into one of the well defined and non-lossy types >> (e.g. use a QString if what you're representing is a string, or >> base64-encode raw binary data), or use CBOR instead. > >> That leaves less amount of surprises when upgrading to Qt 6 while being >> clear about the possible downsides. > > Thank you. > > For Qt 6, I think we should look at the list of meta types and explicitly > list > how they are converted to CBOR and JSON, so for QByteArray it would be clear > that it's a lossy conversion using fromUtf8. Any type not in the white list > gets converted to JSON null or CBOR undefined but subject to change. > > And unit-test them.
That sounds like a great way forward. Thank you Thiago! Cheers, Tor Arne _______________________________________________ Development mailing list Development@qt-project.org https://lists.qt-project.org/listinfo/development
