Potential discrepancy in gdft-builder validation of multiple LogicalChannels

[PhotonDevelopment]

Hi everyone,

I built a program that goes through all GDTFs in the gdtf-share-library via the gdtf-share-api and looks for fixtures with more than one LogicalChannel within a DMX channel.

I went through the entire library, which currently contains around 13,000 GDTFs, and found 31 GDTFs with multiple LogicalChannels.

When opening these files in the gdtf-builder, most of them result in an error. Looking at the affected files, it appears that the DMX ranges covered by the ChannelFunctions of the different LogicalChannels overlap.

I've attached a text document containing links to all 31 GDTFs.

This made me wonder whether there might be a discrepancy between what the gdtf-builder considers valid and how multiple LogicalChannels are actually intended to be used.

The GDTF Share documentation seems to explicitly describe multiple LogicalChannels overlapping with eachother within a DMX channel as a valid use case. (GdtfShare-help-page)

So I'm wondering whether the errors reported by the builder could be caused by a limitation or bug in its validation logic, rather than the GDTFs themselves being invalid.

Has anyone encountered this before, or could clarify how these cases are supposed to be handled according to the GDTF specification?

Thanks!

multpile_logical_channel_gdtfs.txt

[Petr Vanek - Robe]

Hello,

i checked some of these files and they are faulty. If multiple Logical Channels exist, their DMX ranges must not overlap.

[PhotonDevelopment]

I’ve now also reproduced the same error with the example from the GDTF Share Help Page itself:
https://gdtf-share.com/share.php?v=gdtf&page=home&manu=PhotonDevelopment&fix=LogicalChannelExample

So this seems to indicate that either the example on the Help Page is incorrect, or the Builder’s validation logic is rejecting a case that is supposed to be valid according to the documentation.

It’s of course possible that I misunderstood something when reproducing the example. If so, I’d be happy to know where I went wrong. But there seems to be some discrepancy here somewhere, since following the example as described also results in a validation error.

[Petr Vanek - Robe]

OK, i have checked that and the Builder is actually not checking it correctly, i have now opened an issue for the GDTF Builder.

The documentation is correct and your example is also built correctly. This is based on the principle that a  ChannelFunctions of a logical channel must describe complete value range of a DMX channel (this is intentionally ensured by the DMXFrom (and the lack of DMXTo).

Thank you for reporting.