Add support for text FBX

[200sc]

We currently only have working (tested) code for binary FBX files.

[mewmew]

To enable round-trip encoding and decoding of FBX files, and support further use cases (e.g. in-memory manipulation of existing FBX models), I'd like to propose adding support for text output of the FBX ASCII format.

Personally, I find support for input to be less critical as the binary and ASCII formats are isomorphic, and ofbx already support binary decoding.

The use case I have which drives this is an existing set of FBX models which I would like to further extend (for instance, add animations, textures, etc). The approach I envision using to solve this is to decode the binary FBX models into in-memory Go structures using ofbx, and then to extend these structures as see fit (e.g. adding AnimationStacks to the parsed Scene). I would then invoke scene.FBXString() (or ofbx.WriteText(w io.Writer, scene *Scene)) to output the modified FBX model (including the animation) in FBX ASCII format.

Existing conversion tools such as FBX Converter Archives may be used to convert back and forth between the isomorphic binary and ASCII representations of FBX.

Adding this as a data point for use cases and as a proposal to add support for ASCII FBX output to ofbx.

Cheers,
Robin

Edit: note, this approach is also taken by https://github.com/llir/llvm which supports output of the ASCII LLVM IR format and rely on the official LLVM opt tool for converting between isomorphic versions of the file format.

[200sc]

Fundamentally I'm not opposed to this, I don't know when I'll personally have time to work on it however. I'll tentatively say that I'll start looking into it in a couple of days.

[200sc]

Just to give an update, because I did look at this and completely forgot to come back and comment on it: to be able to output in text format we need to be able to parse in text format, otherwise testing that it works requires bringing in another tool to perform that parsing in the tests, and that's really not ideal. Obviously based on response time this project is low on my priority list, but I can review PRs and will still attempt to get around to this.

[mewmew]

Thanks for the update @200sc! I agree, having a parser would facilitate round-trip testing a lot. No rush with the implementation, I understand if these features in ofbx are not top on the priority list. Still, would be great to have one day :)

Cheers,
Robin