Qianqian Fang

@frsyuki, I concur with @MaxNoe, it is fairly common these days to use arrays more than 4GB. Most software, if compiled with the right flag (such as `-m64` with gcc)...

Re @kuenishi: > By the way, the size limit of 32bit width is a bit incomplete, I agree. This can be replaced by a sequence of `[int_type, int_value]` pairs if...

> But I'm not a supporter of the idea to use the combination of bytes because ... it doesn't solve the > problem - users cannot define their own data...

Re: @frsyuki > An example use of 0xc1: https://github.com/msgpack/msgpack-ruby/blob/1e35fb8a771339fc51a9a9c96e77046dfc086954/ext/msgpack/unpacker.c#L53-L61 reading your above code, and it appears to me that `0xc1` was used as a marker to help decide if one...

Just to explain this a little bit more, I propose to use the unused label `0xc1` to represent a new array class that allows one to store N-D numerical arrays...

I also suggest to use `[0xc1] [0xc3=true]` and `[0xc1][0xc2=false]` pair to define an array with dynamic length, i.e. ``` [ 0xc1 ] [ 0xc3 ] [type1] [data1] [type2] [data2] ......

Just want to let you know I decided to switch from column-major to row-major order when serializing an N-D packed array using the proposed array construct. This change was a...

> Using 0xc1 for a new type removes the ability to use it as an _escape character_ for other possibilities (potential preferred extensions). You did realize that with your dynamic...

A proposal for N-D typed packed array can be found in Issue #268 and Pull request https://github.com/msgpack/msgpack/pull/267, along side with a proposal for dynamically sized array/map. The N-D packed array...

> Thanks @fangq! This is great to see. If I'm reading this right, jmsh/bmsh are specifically surface meshes, not data sampled to surface meshes? If so, then it corresponds to...