Fix 25 year old typo

[kirberich]

I assume the code actually does the right thing, as I think it would be quite obvious if not, but I just noticed this typo in the original spec (https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=823374) and saw it's been copied to the LinuxCNC docs as well.

G94 is actually executed first and is duplicated here, instead of the much more sensible G92.3, which is missing.

[petterreinholdtsen]

The change look good to me, but I do not really know G-code that well and would prefer someone with higher confidence in the inner workings of g-code to decide if this is safe to go in. Is there documentation from other vendors about their order of execution?

[kirberich]

Judging from how every bit of documentation I can find has this same error, I think the spec is really the only place this is ever mentioned - I've attached a screenshot of the bit of the spec that I'm basing this on. I even checked the original version of the spec, but that doesn't have the order of execution in as much detail

[petterreinholdtsen]

I found the same duplication replicated on https://www.cnccookbook.com/g-code-basics-program-format-structure-blocks/ and https://docs.edingcnc.com/order-of-execution, so I guess it is widespread. Perhaps the issue should be reported as a errata for the specification too?

[kirberich]

I did wonder about that, but thought it might be a lost cause for a small thing like this 😅 as it doesn't look like it's been updated since 2000 (as far as I could tell, maybe I'm wrong about that!)