<br><br><div class="gmail_quote">On Fri, Jan 9, 2009 at 1:20 PM, Marco Martin <span dir="ltr"><<a href="mailto:notmart@gmail.com">notmart@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="Ih2E3d">On Friday 09 January 2009, Marco Martin wrote:<br>
> on a slightly unrelated note about ruby bindings: on what criteria some<br>
> methods seems to be present others not?<br>
> i see for example that Svg has resize(int, int) and not resize(qsize)<br>
> and FrameSvg has no resizeFrame at all (making the whole class a bit well,<br>
> useless :/)<br>
<br>
</div>eeeh, just forget what i said here, was just an issue of QSize->QSizeF<br>
so it actually works :p</blockquote><div>Yes in C++ will use 'implicit constructors' (is that the right term?) to convert a type you pass as an argument to the real type that the method expects. <br><br>At the moment QtRuby doesn't emulate that behaviour, and it currently needs to be special cased. For Plasma I've done a bit of special casing for Qt::Variants in the scripting Data Engine methods, but I think special casing all the QSizeF methods is more work than it's worth.<br>
<br>-- Richard<br></div></div><br>