Community/Sheetcam job import

Sheetcam job import

Closed
Feature Request

Joel Steiner27 days ago19 comments

Just an idea, if it's feasible to develop -- I have a bunch of Sheetcam job files for partially-cut sheets in my rack.

Attached a 3/8 plate job file from the shop if it helps. 60" (Y) x 120" (X)

19 Comments

showing 1619 of 19
Travis GillinStaff25 days ago

I published the release, it's live now with SheetCAM job file import support. Update and give it a go. It's experimental (which is noted in the dialog at import time) and I'm aware that some jobs are not importing correctly. A bunch of them are though. I have not been able to determine what the breaking difference between them are yet. I'm going to revisit this next month and see if I can narrow it down. In the mean time it would be very helpful if you could find 5-10 jobs that import perfectly and 5-10 that are broken on import. Compile a zip archive with two folders (good, bad). Then email that over too me. I'm missing something subtle, maybe a few things. In my opinion (not bash on Les Newell at all, he's incredibly talented!) the way SheetCAM's job files are structured is archaically organized in the most frustrating way lol

Travis GillinStaff17 days ago

I'm marking this as closed. V2.0.5 has significant work done for SheetCAM Job file imports. Every job file I have on hand which includes the ones you emailed me are now importing correctly. This is the result of over 50 hours spent. Honestly if there's any edge cases left, I'm not going to address them at any sort of priority. SheetCAM's job file format is a dumpster fire full of unordered contour soup. It should be good enough to be useful for you now.

Joel SteinerStaff16 days ago

Sounds great! I'll give this a spin when 2.0.5 is released.

Travis GillinStaff16 days ago

Two of your Sheetcam job files had geometry that found edge cases in my Voronoi offsetting algorithm. Those two parts are super arc dense. I have fallbacks in place that are properly catching the errors but that means falling back to just line segments on those toolpaths and I don't like that because if there's geometry that the validation check misses, it means broken toolpaths. These edge cases would likely never be exposed with mechanical parts but art work would find them. I thought my Voronoi algorithm was stable 10 years ago lol. That was ported from JetCad2 which was never released to the public. I've been working on sorting this out since 7pm yesterday evening. Still at it right now. If I can't nail it down by midnight, I'll probably still push the release because the fail-overs do catch it.

Sign in to leave a comment or vote.

Sign In