On 13-5-2015 2:44, Thiago Macieira wrote: > On Wednesday 13 May 2015 02:34:28 Jan Kundrát wrote: >> Hi, >> this commit [1] added a new overload to QTextStream::readLine. As a result >> of that, calling stream.readLine(0) is now ambiguous: >> >> QString readLine(qint64 maxlen = 0); >> bool readLine(QString *line, qint64 maxlen = 0); >> >> While I can easily fix this in the caller (Konsole in this case), I'm >> wodnering whether this effect was understood at the time the change was >> merged. IMHO it's a bit more user-friendly to preserve source compatibility >> by removing the default value in the newly added overload. >> >> Should I send such a patch? > Why write readLine(0) when it's the same as readLine() ? readLine(0) seems to > me that it's asking for a line of at most zero bytes -- we should have had > this argument default to -1 to indicate maximum length, not 0. > > Removing the default argument for the new overload will make it worse, by > making people have to write the zero when they mean unlimited. I don't like > that. The two options I will consider are: > > a) do nothing, accept and document the source incompatibility > b) modify differently so that the new API isn't ambiguous but doesn't require > people to write 0 either. In the spirit of option b), would it be an option to have the method take a QString& instead of a QString*? That would resolve the ambiguity. It also makes it clear that it makes little sense to call the method without an actual QString to store the results in.
André _______________________________________________ Development mailing list Development@qt-project.org http://lists.qt-project.org/mailman/listinfo/development
