Re: [Crystal-cvs] SF.net SVN: crystal:[38394] CS/trunk/include/csutil/vararg.h
Eric Sunshine <[email protected]>
| Newsgroups | gmane.comp.graphics.crystalspace.devel |
|---|---|
| Message-ID | <CAPig+cRXrV=AFXwjxrkyyMzX+=DaxhZ6c8XsCDOi0vrWRKbA7w@mail.gmail.com> |
On Tue, Jul 17, 2012 at 9:13 AM, res <[email protected]> wrote: > Probably unsurprising: I'm fine with having SCF interfaces that contain inline functions. ;) > (I changed the docs to make note of the acceptability of small helper/convenience inline methods.) The case you documented seems reasonable, especially the key bit saying "would work for any implementation". > Don't forget that the SCF interface version should be bumped if (virtual) methods are removed. (I don't think you did when implementing the va_call() approach.) Indeed, I bumped the interface version numbers in this revised implementation. The failure to do so with the va_call() approach was an intentional, if questionable, choice. Considering that, with the va_call() solution, the interface changed only on OSX, _and_ it was and has long been horribly broken anyhow by Apple's shortcoming, bumping the version number did not seem especially meaningful. -- ES ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/