KCL: new 'hole' module for mech-eng #8530

adamchalmers · 2025-10-10T23:09:23Z

Add a new hole module with countersink, counterbore etc holes. This has been ready for two months or so, but I got distracted by constraint solvers. Clearly the ML team has found workarounds for converting holes in other CAD programs, but we should just add this as an experimental library and let people test it out in prod.

vercel · 2025-10-10T23:09:29Z

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Preview	Comments	Updated (UTC)
modeling-app	Ready	Preview	Comment	Oct 15, 2025 3:54am

rust/kcl-lib/std/hole.kcl

codspeed-hq · 2025-10-10T23:28:47Z

CodSpeed Performance Report

Merging #8530 will not alter performance

_{Comparing achalmers/hole-lib (e3325e1) with main (cff54a4)¹}

Summary

✅ 139 untouched
⏩ 92 skipped²

No successful run was found on main (9dbca34) during the generation of this report, so cff54a4 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩
92 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

jtran

I noticed that this isn't marked as experimental. Reminder that that's an option.

rust/kcl-lib/std/hole.kcl

jtran · 2025-10-11T14:39:30Z

rust/kcl-lib/std/hole.kcl

+/// From the hole's parts (bottom, middle, top), cut the hole into the given solid,
+/// at each of the given 2D positions on the given face.
+/// Basically like function `hole` but it takes multiple 2D positions in `cutsAt`.
+export fn holes(@solid, face, holeBottom, holeBody, holeType, cutsAt) {


Is having both the singular and plural to make KCL read nicely?

I almost don't want to remind anyone because I absolutely hate this feature, but if you add the parameter types and use an array, any single value will be automatically wrapped in an array. Part of the reason we made the automatic coercion to array was to address this exact use case where a function can take either a single thing or an array, without duplication.

That said, if you'd rather be explicit, and avoid all that terrible automatic coercion, I support it, but wonder if the singlular hole() flavored fn is necessary.

rust/kcl-lib/tests/holes_cube/input.kcl

jtran · 2025-10-11T14:50:07Z

rust/kcl-lib/std/hole.kcl

+
+/// End the hole flat.
+export fn flat() {
+  return { drillBitAngle = 179.99deg }


Is this leftover from the engine bug that didn't allow 180deg? I thought it was fixed.

It's actually a logic error, as tan(180) is 0 and somewhere in the math I divide by tan(drillBitAngle), so I kept it as 179.99

Seems strange to me to hardcode such a weird value. It's a form of a tolerance. It means that if anyone uses { drillBitAngle = 180deg } directly, without going through flat(), we'll needlessly fail.

Maybe we should move the check into sketchHoleProfile().

jtran · 2025-10-11T14:52:32Z

rust/kcl-lib/std/hole.kcl

+/// Place the given holes in a line.
+/// Basically like function `hole` but cuts multiple holes in a line.
+/// Works like linear patterns.
+export fn holesLinear(@solid, face, holeBottom, holeBody, holeType, cutAt, instances, distance, axis) {


Should we add parameter types? They would show up in the docs, and I think they're generally helpful to users.

pierremtb · 2025-10-15T13:36:50Z

rust/kcl-lib/std/hole.kcl

+  holeType,
+  /// Where to place the cut on the given face of the solid.
+  /// This controls how far up/down/left/right the hole is placed.
+  cutAt: [number(Length); 2],


@adamchalmers I know this is still experimental so we don't have to handle that case today, IMO it'd be really convenient to be able to select a sketch point here instead of giving relative coords to the face. One we have sketch v2 this would all integrate nicely with constraints!

Oh, the docs are actually wrong, it's actually absolute position. I'm opening up a PR to fix. #8560

pierremtb · 2025-10-15T13:47:29Z

rust/kcl-lib/std/hole.kcl

+  }
+}
+
+// Bodies


Through holes aren't implemented now, right?

adamchalmers requested a review from a team as a code owner October 10, 2025 23:09

adamchalmers requested a review from pierremtb October 10, 2025 23:09

graphite-app bot reviewed Oct 10, 2025

View reviewed changes

rust/kcl-lib/std/hole.kcl Outdated Show resolved Hide resolved

vercel bot deployed to Preview October 10, 2025 23:22 View deployment

adamchalmers requested a review from a team as a code owner October 10, 2025 23:40

vercel bot deployed to Preview October 10, 2025 23:52 View deployment

adamchalmers force-pushed the achalmers/hole-lib branch 2 times, most recently from 7f11d1f to 4faa15d Compare October 11, 2025 00:22

adamchalmers changed the base branch from main to achalmers/no-empty-modules-in-sim-tests October 11, 2025 00:22

adamchalmers force-pushed the achalmers/no-empty-modules-in-sim-tests branch from f7cd818 to df4831d Compare October 11, 2025 00:32

vercel bot deployed to Preview October 11, 2025 00:35 View deployment

adamchalmers force-pushed the achalmers/hole-lib branch from 4faa15d to 7fe91b4 Compare October 11, 2025 00:38

Base automatically changed from achalmers/no-empty-modules-in-sim-tests to main October 11, 2025 00:49

adamchalmers force-pushed the achalmers/hole-lib branch from 527ee29 to 04d5f29 Compare October 11, 2025 00:50

adamchalmers changed the title ~~KCL: hole support~~ KCL: new 'hole' module for mech-eng Oct 11, 2025

adamchalmers enabled auto-merge (squash) October 11, 2025 01:09

vercel bot deployed to Preview October 11, 2025 01:24 View deployment

jtran reviewed Oct 11, 2025

View reviewed changes

adamchalmers force-pushed the achalmers/hole-lib branch 3 times, most recently from dc4fd1d to b1b8f6b Compare October 12, 2025 20:53

vercel bot deployed to Preview October 12, 2025 21:06 View deployment

adamchalmers force-pushed the achalmers/hole-lib branch from b1b8f6b to b6f0fbd Compare October 12, 2025 21:48

vercel bot deployed to Preview October 12, 2025 22:01 View deployment

adamchalmers force-pushed the achalmers/hole-lib branch 2 times, most recently from b43b21b to 9c072c4 Compare October 14, 2025 18:55

vercel bot deployed to Preview October 14, 2025 19:10 View deployment

vercel bot deployed to Preview October 14, 2025 20:34 View deployment

adamchalmers added 6 commits October 14, 2025 20:52

New KCL module 'hole' added

0ebdb7c

We do want to use the prelude in this module actually

6b04bac

Update docs and test artifacts

17ddea9

Make the hole module experimental

8ccda10

Add signatures

9c13637

Add signatures

2a69014

adamchalmers force-pushed the achalmers/hole-lib branch from cc21dd8 to 2a69014 Compare October 15, 2025 01:52

vercel bot deployed to Preview October 15, 2025 02:08 View deployment

adamchalmers force-pushed the achalmers/hole-lib branch from e9d4f89 to 26c0cd9 Compare October 15, 2025 02:33

Handle experimental KCL stdlib fns defined in KCL.

24f1554

adamchalmers force-pushed the achalmers/hole-lib branch from 26c0cd9 to 24f1554 Compare October 15, 2025 02:34

vercel bot deployed to Preview October 15, 2025 02:49 View deployment

Fix test

e3325e1

vercel bot deployed to Preview October 15, 2025 03:54 View deployment

pierremtb reviewed Oct 15, 2025

View reviewed changes

pierremtb mentioned this pull request Oct 15, 2025

New Modeling Command Dialogs Instead of Individual Args #8522

Open

jtran approved these changes Oct 15, 2025

View reviewed changes

adamchalmers merged commit ca50eb0 into main Oct 15, 2025
77 checks passed

adamchalmers deleted the achalmers/hole-lib branch October 15, 2025 14:58

pierremtb linked an issue Oct 15, 2025 that may be closed by this pull request

Generic hole stdlib functions #7625

Closed

KCL: new 'hole' module for mech-eng #8530

KCL: new 'hole' module for mech-eng #8530

Conversation

adamchalmers commented Oct 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

vercel bot commented Oct 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

codspeed-hq bot commented Oct 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merging #8530 will not alter performance

Summary

Footnotes

Uh oh!

jtran left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

adamchalmers Oct 15, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

adamchalmers commented Oct 10, 2025 •

edited

Loading

vercel bot commented Oct 10, 2025 •

edited

Loading

codspeed-hq bot commented Oct 10, 2025 •

edited

Loading

adamchalmers Oct 15, 2025 •

edited

Loading