James Kerns

Will be fixed in 3.26. Many thanks for the suggested code. It worked perfectly.

I definitely missed the difference in reference temperature! Really sorry about that. I'll have a fix and warning in the docs.

Will be fixed in v3.29!

https://pylinac.readthedocs.io/en/latest/changelog.html#trs-398

If you're talking about 1D profiles, this section should help: https://pylinac.readthedocs.io/en/latest/topics/profiles.html. You can create your own types of metrics (e.g the mean of the central 1cm or something). Similar things...

For backwards compatibility I'll have to leave it the way it is and update the documentation. Thanks

Looks like possible datatype overflow error (e.g. the image might be 32 bit instead of 16). It's also possible that the specification has been updated since the original from Varian....

Got the image, thanks! Will take a look shortly.

Yup, datatype overflow. The assumption in the original XIM algorithm explanation was that there would only be uint8 or uint16 datatypes, but this file is trying to specify uint32. Easy...

Per the Dynalog File format specification document (2013):  Unless there is a newer spec.