Unit Conversion Broken

[tedh-shopbot]

The primary/basic unit conversion from inches to mm or mm to inches no longer updates the current position {pos: } after the unit change made with {gun: }.

This functionality worked correctly through version 101.57.32.
It was broken with commit 4d149970 "main item 34 faulty sr from feedhold or kill" 1/28/22 and remains broken through 101.57.35.

Sample after break (edited for readibility):
g2: --S-1647109441151----> {gun:n} [g2]
g2: <-S-1647109441158----- {"r":{"gun":0},"f":[1,0,9]} [g2]
g2: --S-1647109392429----> {mpo:n} [g2]
g2: <-S-1647109392440----- {"r":{"mpo":{"x":54.86393,"y":27.43183,"z":0,"a":0,"b":0,"c":0}},"f":[1,0,9]} [g2]
g2: --S-1647109418224----> {pos:n} [g2]
g2: <-S-1647109418235----- {"r":{"pos":{"x":2.16,"y":1.07999,"z":0,"a":0,"b":0,"c":0}},"f":[1,0,9]} [g2]
g2: --S-1647109452416----> {gun:1} [g2]
g2: <-S-1647109452423----- {"r":{"gun":1},"f":[1,0,9]} [g2]
g2: --S-1647109465315----> {mpo:n} [g2]
g2: <-S-1647109465325----- {"r":{"mpo":{"x":54.86393,"y":27.43183,"z":0,"a":0,"b":0,"c":0}},"f":[1,0,9]} [g2]
g2: --S-1647109472507----> {pos:n} [g2]
g2: <-S-1647109472520----- {"r":{"pos":{"x":2.16,"y":1.07999,"z":0,"a":0,"b":0,"c":0}},"f":[1,0,9]} [g2]
=> tool is now in metric and these values should reflect the change rather than just duplicating the inch values

(can't really call this a regression since it has always worked and is pretty basic)

(previously this was FabMo-Engine issue #827)

[Cosmic7373]

Is this the kind of behavior that is expected, but is not occurring?

g21
{"r":{},"f":[1,0,4]}
{"sr":{"posx":76.2,"stat":3,"unit":1}}
g20
{"r":{},"f":[1,0,4]}
{"sr":{"posx":3,"stat":3,"unit":0}}

[tedh-shopbot]

Yes.
FabMo uses gun:0/1 to set default units as well as the g20/21 which is run at the beginning of a file to specify the unit for the file.

[tedh-shopbot]

With respect to 'gun', FabMo-G2core operates differently than G2core. That's because we needed/wanted a system for setting a default and persistent Unit mode independently of running a gcode file. I believe this is basically accomplished in:
canonical_machine.cpp Lines: 2516 - 2525

It has worked pretty well up til now.

[tedh-shopbot]

Hmm ...
Just spent some time in distraction today because this bug still exists and I had forgotten I had already documented it. It is pretty ugly. Definitely one we can not release without fixing.

[msxmatt]

Hey @tedh-shopbot yea it's been a little while, but I remember it was a bit ugly. I think we discussed pulling the out the faulty SR fix, since it appeared to break a number of items. We did some squashing of bugs for a while, but others cropped up so it seemed like that fix should be pulled for a release.

I can hop back on the chat and we can discuss how you'd like to attack it.

[tedh-shopbot]

Thanks, Matt @msxconsulting. @RobMackie is pondering this one and a few other things. I'll let him get back to you on this one as he sorts plans out. =ted