All pull requests

Pull requests

• fix: preserve cancellation through boolean status handlers (#2870)

  #2873·tdamsma opened Nov 5, 2026Contributor·

  ·
  1

• [WIP] fillet_variable: a radius law with a corner is its balls' union (split in do_filleting)

  #2872·MarkusReichart opened Nov 5, 2026Contributor·

  ·
  1

  1

• [WIP] blend: a round whose edge runs on across a seam's vertex is refused, not ended in a step (#2864, part 1)

  #2869·MarkusReichart opened Nov 5, 2026Contributor·

  ·
  1

• A round of a two-cylinder join builds when both seams lie on the join: the solver's guess comes from the edge use's own UV trim (part of #2826)

  #2863·khanahmad4527 opened Nov 5, 2026Contributor·

  ·
  3

• [WIP] to_smb: write hyperbolas, composite and offset curves, STEP surfaces and surfaces in volumes

  #2860·MarkusReichart opened Nov 5, 2026Contributor·

  ·
  1

  2

• fix: tests 13/14 compare a radius of curvature with a length, not a speed (#2785)

  #2859·saosebastiao opened Nov 5, 2026Contributor·

  ·
  1

  1

• wire: replay recorded all-hole untrim operations (#2855)

  #2856·tdamsma opened Nov 5, 2026Contributor·

  ·
  1

  7

• wire: replay union_all in the kernel rig (#2850)

  #2853·tdamsma opened Nov 5, 2026Contributor·

  ·
  1

  1

• [WIP] test: three SolidCraft 0.40 RC fillet failures as known-failing kernel cases (#2843, #2844, #2845)

  #2846·MarkusReichart opened Nov 5, 2026Contributor·

  ·
  1

• tools: replay SolidCraft benchmark recipes with sanitized evidence

  #2841·ChrisNourse opened Nov 5, 2026Contributor·

  ·
  13

• fix: projected extrema used by small-profile mitres

  urgent
  #2840·ChrisNourse opened Nov 5, 2026Contributor·

  ·
  26

• Kernel rig: replay stitch (#2836)

  #2837·MarkusReichart opened Nov 5, 2026Contributor·

  ·
  1

  2

• difftest: handle parked diagnostic value share (#2796)

  #2833·tdamsma opened Nov 5, 2026Contributor·

  ·
  1

  3

• test: honor guarded offset paths in native library checks (#2798)

  #2831·tdamsma opened Nov 5, 2026Contributor·

  ·
  3

• chore: retire the dead offset_extends_cones_as_themselves and offset_extends_ball_patches_as_themselves flags (WR-532, WR-728)

  #2825·zorak1103 opened Nov 5, 2026Contributor·

  ·
  2

• fix: restore diagnostics formatting on main

  #2820·altvnk opened Nov 5, 2026Contributor·

  ·
  3

• test: select every edge of the rim in the exact-class cases (#2441)

  #2818·MarkusReichart opened Nov 5, 2026Contributor·

  ·
  1

  6

• Kernel tests become goldens: each case checked against its own expected output, no C++

  #2816·jyviko opened Nov 5, 2026Collaborator·

  ·
  3

• difftest(primitive.primitive_creation_p2): gen0092 and gen0189 recorded, the exact net's small elbow is an invalid body

  #2811·ablasik opened Nov 5, 2026Contributor·

  ·
  4

• fix(blend): a knife edge chamfered at equal distances builds, as the chord between the ball's contacts

  #2804·bherbruck opened Nov 5, 2026Collaborator·

  ·
  6

• difftest: record seven heal.gap append-state changes

  #2803·altvnk opened Nov 5, 2026Contributor·

  ·
  1

  4

• blend: a closed edge meeting itself at an angle is a Mitred Turn (part of #2349)

  #2800·Karl-VonRichter opened Nov 5, 2026Contributor·

  ·
  3

• test: exact tests of the distance bound's rounding charges

  #2795·MarkusReichart opened Nov 5, 2026Contributor·

  ·
  1

• fix: reject failed N-D solver evaluations by default (#2782)

  #2791·fdalmaso opened Nov 5, 2026Contributor·

  ·
  1

  2

• fix: include every point-set block in the default centroid (WR-15)

  #2788·altvnk opened Nov 5, 2026Contributor·

  ·
  1