Re: [opensource-dev] Fix for Attachments displayed in mouselook bug

2010-10-10 Thread Ponzu
On Sat, Oct 9, 2010 at 11:30 PM, Tateru Nino tateru.n...@gmail.com wrote: Well, that depends. That element might be used for something else, elsewhere. If the _intended_ behaviour of the function (and of the array/container) is documented, then it's relatively easy to make decisions about

Re: [opensource-dev] Fix for Attachments displayed in mouselook bug

2010-10-10 Thread Ponzu
Well, this is pretty common idiom for code that deals with bit masks. Funny. The problem is that the bitmask (hard to understand?) was replaced with a simple array (easy to understand? and that is when the error was introduced. On Sat, Oct 9, 2010 at 2:59 PM, Zabb65 zab...@gmail.com wrote: I

Re: [opensource-dev] Fix for Attachments displayed in mouselook bug

2010-10-10 Thread Ponzu
This code below means it used to be inline, but now is a function call (unless the compiler is optimizing or something). You guys know the code wy better tahn I, but is it worth moving it back to the .h file? return type mRenderTypeEnabled[type] looks hard to beat for inlining, and it is

Re: [opensource-dev] Fix for Attachments displayed in mouselook bug

2010-10-10 Thread Tateru Nino
The original code looks very efficient, especially for inlining. Anyone have any idea if it actually does what it is supposed to do by accident or by design? On 11/10/2010 3:47 AM, Ponzu wrote: This code below means it used to be inline, but now is a function call (unless the compiler is