Impelmentation updates to AMF spec

Discussions - ACESNext VWG – ACES Metadata File (AMF)
1

Attached is the latest draft of the AMF spec with the new UML in it.

@dtatut and @CClark let me know how this looks. Happy to update if anything looks amiss.

S-2019-001.pdf (1.2 MB)

2

Hi Alex,

I think it’s important to highlight the changes made in addition to updating the new UML diagram. The main change is the addition of <file> and <uuid> for linking to external transforms that do not have an official <transformID> (i.e custom IDTs, LMTs, etc) which we feel will continue to be an important use case for AMF. As it stands, this should now part of the transform types for input, look, and output transforms.

I see those additions in the diagram, but are they missing from Section 6 where each element and type are described? I may be reading it wrong since I’m really not an xml schema expert.

I do see them here under outputTransformType - is aces:uuid there twice a typo?

Screen Shot 2021-02-12 at 2.30.23 PM
Screen Shot 2021-02-12 at 2.30.23 PM1048×436 53.5 KB

But it seems to be missing from inputTransformType and lookTransformType? Or am I missing something?

As we discussed on the call yesterday, I think the UML diagram, while the text is very small, looks good once zoomed in.

Thanks,
Chris

3

Thanks @CClark. I really appreciate you looking closely at this.
I’ll go through each point, address, and post a new version.

4

hey @CClark

Thanks for the input … I’ve corrected the problem you found in 6.2.2.13. Good catch.

As for the file and uuid tags. Those are defined in the following sections.

  • Input transforms sections
    aces:inputTransformType / aces:file is in section 6.3.7.1
    aces:inputTransformType / aces:uuid is in section 6.3.7.6
    aces:inverseOutputDeviceTransformType / aces:file is in section 6.3.8.1
    aces:inverseOutputDeviceTransformType / aces:uuid is in section 6.3.8.3
    aces:inverseOutputTransformType / aces:file is in section 6.3.9.1
    aces:inverseOutputTransformType / aces:uuid is in section 6.3.9.3
    aces:inverseReferenceRenderingTransformType / aces:file is in section 6.3.10.1
    aces:inverseReferenceRenderingTransformType / aces:uuid is in section 6.3.10.3
  • Look transforms sections
    aces:lookTransformType / aces:file is in section 6.3.11.2
    aces:lookTransformType / aces:uuid is in section 6.3.11.4
  • Output transforms sections
    aces:outputDeviceTransformType / aces:file is in 6.3.12.1
    aces:outputDeviceTransformType / aces:uuid is in 6.3.12.3
    aces:outputTransformType / aces:file is in 6.3.13.1
    aces:outputTransformType / aces:uuid is in 6.3.12.5
    aces:referenceRenderingTransformType / aces:file is in 6.3.16.1
    aces:referenceRenderingTransformType / aces:uuid is in 6.3.16.3

I did correct the following typos:

  • section 6.3.16 erroneously had aces:referenceRenderingTransform instead of aces:referenceRenderingTransformType
  • a couple of elements in 6.3.7 where the examples were in the children section and children were missing
  • in the example in 6.3.7.2 where it said <aces:inverseOutputTransform> instead of <aces:inverseOutputDeviceTransform>

The one last thing I’m working on is breaking the UML into smaller chucks. Hope to have that done later today. For now, here’s the latest.

S-2019-001.pdf (1.2 MB)

5

Here’s the version with the update UML diagrams.

This should be final assuming no one catches typos or copy/paste bombs.

Alex

S-2019-001.pdf (1.2 MB)

6

@Alexander_Forsythe The new diagrams look great. Much more readable!

Thanks,
Chris

7

Hey Alex, I was recently writing an AMF parsing tool and found that in the example of “aces : clipIdType / aces : sequence”, there seems to be a bit of an error where you need to add spaces between each attribute, otherwise there will be an XML parsing error.

such as：
<aces:sequence idx=“#" min=“1" max="240”>

Screen Shot 2021-09-25 at 14.28.21
Screen Shot 2021-09-25 at 14.28.212418×336 23.7 KB

and In the example on page 68, an unnecessary " may have been added.

best,
Yu

8

@youdrewxuan Thanks for catching this … looks like that’s an issue in the document formatting. I’ll fix it.

The schema has it correct.

Alex

1 Like