Preview / interim / grip-drag overlay wires were drawn in the main wire
pass with the depth-tested wire_pipeline (depth_compare=LessEqual), so a
live command preview could be hidden behind occluding geometry — a 3D
solid, or 2D geometry drawn in front of it.
Draw them with the existing wire_xray_pipeline instead (depth_compare=
Always, no depth write), the same always-on-top path selection highlights
already use. Previews are now never occluded, for every command.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
TRIM ignored circles: the target dispatch handled Line/Arc/Ellipse/
Spline/Ray/XLine/LwPolyline but had no Circle arm, so clicking a circle
fell through to a no-op. Add `trim_circle()` — a full circle needs ≥2
boundary crossings; the arc segment under the click is removed and the
circle becomes a single Arc (center/radius/thickness/normal preserved).
Wire it into both the pick (apply) and hover-preview paths.
Also fix `arc_pts`, which emitted points as [x, y, cy+r·sin] — the XZ
plane — so arc and circle previews drew in the wrong plane (obvious on a
full circle). Match the XY convention used by line_pts and fillet's
arc_pts: [cx+r·cos, cy+r·sin, z]; callers now pass center.z.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Tracks acadrust main at 0.4.x: plotstyle/material common-entity
ordering fix plus in-flight DWG reader work. Patch comment updated
to match; drop the git patch once these land in a published 0.4 crate.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Follow-up cleanups on the merged TTF-fill work:
- Drop the unconditional resolve_text_style log line — it fired for every
text style and spammed ~7k stderr lines opening a single real drawing.
- Drop the per-glyph tessellation-error eprintln; a malformed font would
otherwise log once per character. The glyph still falls back silently.
- Tessellate glyph fills with the nonzero winding rule (lyon defaults to
even-odd, which mis-fills TrueType outlines whose contours overlap).
- Make the sysfont prefix fallback deterministic (sorted) and require ≥3
chars so a 1–2 letter request can't grab an arbitrary family.
- Remove the unrelated windows-gnu --exclude-all-symbols linker flag that
slipped into .cargo/config.toml.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A drawing opened with a twisted view made a TOP-face click keep the
twist, so north ended up at the bottom. Add Camera::snap_to_face, which
snaps to a canonical orientation — north (UCS +Y) up for top/bottom,
world (UCS +Z) up for the side elevations — ignoring the current
up-sense. Face clicks use it; edges/corners still preserve their up so
they spin smoothly around the clicked feature.
The Home button routes through the same path, giving a reliable reset
for a twisted view.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Build the ViewCube out into a full navigation widget:
- Compass ring painted in the cube's ground plane (z = base) so it
reads as a ground compass and foreshortens with the view.
- N/E/S/W cardinals painted flat onto the ring band; clicking one snaps
to that side elevation. Added S/W glyphs to the bitmap atlas.
- Home button (top-down view), roll arrows (90° roll about the view
axis), and four nudge triangles (90° tip/spin) as SVG iced controls
laid out around the cube. New home.svg.
- WCS / named-UCS selector under the cube, switching the active UCS.
The cube hit area now shares one stack with the control buttons, so a
click on a button is caught by the button while a click on the cube
falls through. Face clicks snap to the highlighted region rather than a
cursor position the viewport's move handler can overwrite.
Camera gains home_view / roll_by / nudge_90; floating viewports drive
the same ops through mutate_active_viewport_camera, which re-encodes the
result to the stored view direction + twist.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Lay each label's glyph quads out in the face's own local plane and
project the corners individually, instead of drawing screen-aligned
billboard text at the projected face centre. The letters now skew with
perspective and sit on the surface as the cube rotates.
Per-face (u, v) basis vectors are chosen so u x v equals the outward
normal, which keeps the text from ever reading mirrored.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A dimension's drawn geometry lives in a baked anonymous *D block. COPY
cloned the Dimension entity (translating its definition points) but left
its block_name pointing at the source block, whose sub-entities stay at
the original location — so the copy rendered on top of the original and
appeared not to copy at all.
Add `clone_transformed_block`: when copying a dimension that has a baked
block, duplicate that block under a fresh *D name with every sub-entity
transformed by the same offset, and repoint the copy at it. The copy now
lands at the drop point with its baked geometry and text preserved (no
synthesis, so diameter/radius values stay correct). Covered by a portable
regression test.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
ROTATE's reference-point step prompts "or skip: type angle directly", but
on_text_input only handled the later Angle step. So picking a centre and
typing the angle did nothing, and the next Enter cancelled the command —
the objects never rotated, which reads as the rotate "erasing" / losing
them.
Accept a typed angle at the reference-point step too, rotating directly
about the centre (the reference point stays optional). Added an automation
regression test that rotates a line 90° with centre + angle only.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The internal PDF export set each wire's colour and thickness but never a
dash pattern, so every dashed / centre / dash-dot line printed solid. It
also derived thickness from the on-screen pixel weight, which carries a 2×
screen-legibility boost — so print weights were inconsistent and barely
differentiated.
Emit a PDF dash pattern per wire from its linetype run lengths (mm→pt,
real length = index of the last non-zero element + 1, 1 pt floor so a dot
still marks), resetting to solid where there is no pattern. Convert
lineweight with a factor that undoes both the 96-dpi scaling and the
display boost, so print thickness is the true physical mm.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A new drawing's layouts carry paper limits but no sheet viewport; it is
created lazily on the first switch into the layout. That viewport kept
Viewport::new's default view (target 0,0 / height 210), so camera_from_view
aimed at the paper's bottom-left corner, and with no geometry yet the
fit_all fallback was a no-op — the layout opened off-centre.
Give the freshly created sheet viewport a view that frames the page:
view_target at the paper centre and view_height a little taller than the
sheet. The first switch now opens centred on the page, and the view persists.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Creating a viewport in a layout commits through document.add_entity_to_layout
+ Scene::auto_fit_viewport, both of which skip the geometry-cache bump that
Scene::add_entity performs for every other entity. So the new viewport's
border wasn't tessellated until the next zoom/pan happened to invalidate the
cache. Bump the wire tessellation (no-blocks: a viewport touches no block
definitions) right after the viewport is added.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Driving the annotation / viewport scale picker purely from the drawing's
ACAD_SCALELIST regressed every drawing that carries no Scale objects
(most fresh drawings and minimal DXF/DWG exports): scale_list() returned
empty, so the status-bar scale popup opened with nothing to pick and the
button looked dead.
Fall back to the standard ratio set (1:500 … 1:1 … 10:1) when the drawing
defines no scales of its own; file-defined scales still take precedence
whenever present. Also re-enables the viewport-scale picker in the
properties panel for those drawings.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The text / dimension / multileader / table style combos rendered their
open list inline inside the ribbon row, so the popup was clipped to the
fixed ribbon-row height and got cut off even with only two entries. The
Draw-tab dropdowns and the layer combo avoid this by rendering as
floating overlays.
Render the style list as an overlay too: the combo now draws only its
button inline, and a new `style_combo_overlay` builds the list as a
floating panel that grows to fit its entries, positioned with the
existing `compute_dropdown_left` (which already accounts for
StyleComboGroup widths) and mirroring `layer_combo_overlay`.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two issues in a file whose dimensions carry baked *D### blocks.
Missing diameter text: the baked MText values are "\U+220520" /
"\U+220510" — the \U+2205 (⌀) escape immediately followed by the literal
digits "20"/"10". The MText parser read up to six hex digits for a \U+
escape, so it greedily consumed "220520" = 0x220520, which is past the
maximum code point, char::from_u32 returned None, and the whole text
dropped. A \U+ escape is exactly four hex digits; read four so the symbol
decodes and the trailing digits stay literal. (Angular "90\U+00B0" only
worked because its escape sits at the end of the string.)
Node artifacts: PDMODE=0 points were drawn as a fixed 0.1-world-unit "+",
which in a small drawing turned the 31 dimension def-points on the
Defpoints layer into visible crosses. Size the dot from world-per-pixel
(~1 px, zoom-adaptive via the tessellation cache key) to match the near-
invisible dot other CAD apps draw.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The continued dimension line was positioned with an offset measured from
the base dimension's *first* extension origin, then re-applied to each new
chain point (the previous dim's second origin). When the two extension
origins sat at different perpendicular positions — i.e. the extension
lines had different lengths — that offset placed the new dim line at
`D + (p2 - p1)·perp` instead of the base line's `D`, so the chain drifted
off the base line. Equal-length extension lines have `(p2 - p1)·perp = 0`,
which is why it only misbehaved sometimes.
Store the base dim line's absolute perpendicular coordinate and project
each extension origin onto it, so the whole chain stays collinear
regardless of extension-line lengths. The preview uses the same
projection.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
DIMALIGNED's drag preview never matched where the dimension landed. The
preview offset the dimension line in the XZ plane while the committed
entity's renderer offsets in the XY plane, and the commit path passed the
straight-line p2→cursor *distance* to `set_offset`, which re-applies it
along the perpendicular — wrong distance, always on the +perp side. Both
now store the cursor as the definition point and project it onto the
XY-plane perpendicular, matching the renderer and DIMLINEAR.
The same XZ-vs-XY plane confusion affected other annotate commands:
- DIMORDINATE worked entirely in the XZ plane (dropping the Y coordinate)
and left `definition_point` at the origin, so the leader kinked through
(0,0,0). Reworked onto the XY plane with a proper orthogonal elbow and a
live preview that matches the placed entity.
- DIMTEDIT / DIMJOGLINE cursor markers were built in the XZ plane, so in
the top-down view they collapsed to a flat line instead of a box /
zigzag. Rebuilt in the XY plane.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A dot in an LTYPE pattern is a zero-length element. The pattern resolver
encoded it as a tiny `0.01 * scale` world-length dash, which had two bad
effects: it was sub-pixel at normal zoom so the dash LOD saw the pattern's
smallest feature drop below one pixel and collapsed the whole line to
solid, and at larger LTSCALE it left only invisible sub-pixel dots between
big gaps. Either way dotted and dash-dot linetypes were never visible;
only plain dashed (no zero elements) worked.
Keep dots as literal 0.0 so they no longer drag `min_elem` down, and
render each as a fixed ~1.5 px mark in the wire shader (pixel-snapped via
world_per_pixel, mirroring the hatch shader's dot handling). The shader
tells a real dot from trailing array padding by element count — the index
of the last non-zero element plus one — so no producer needs a sentinel.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The Center object snap was a pre-baked point gated purely on the
cursor's screen distance to the centre point. A circle's centre sits
inside it and an arc's centre is usually off in empty space well away
from the curve, so hovering the curve — the natural gesture — never
landed within the snap radius and no centre marker appeared.
Add a curve-proximity pass: when Center is enabled and the cursor is
near a circle/arc/ellipse curve, offer that curve's centre, ranked by
the cursor's distance to the curve. It runs after the `try_pt` closure's
borrow ends so it can update the candidate state directly, mirroring the
Tangent pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The hatch boundary was flat-truncated to 1024 vertices before upload.
The active renderer is the batched path, which stores boundaries in a
GPU storage buffer with no length limit, so the cap only ever served
the per-hatch uniform path (now wipeout-only, and self-clamping).
On a real drawing a 36-loop, depth-3 nested hatch (1482 verts) was cut
mid-boundary: trailing island loops were dropped and the final partial
loop was left open, flipping the even-odd parity so the fill bled across
the rest of the shape and islands rendered incorrectly.
Drop the flat truncate. Keep every loop; only guard pathological vertex
counts (16384), and when trimming, cut at a NaN sentinel (whole-loop
boundary) so no sub-loop is ever left open.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Lineweights were rendered at their true physical size (mm × 96/25.4 px),
which is ~2 px for a 0.5 mm line — too timid to tell weights apart, and
every weight ≤ 0.25 mm floored to 1 px (no visible difference). Other DWG
editors display model-space lineweights larger than true size for screen
legibility; apply the same 2× boost so weights are pronounced and
distinguishable.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Pull in acadrust 16332e5, which reads/writes the DWG SORTENTSTABLE in the
correct stream layout (sort handles inline in the data section; owner block
+ entity handles in the handle stream). Draw order set in other CAD apps —
e.g. a line sent behind a hatch — is now honoured instead of falling back to
handle order.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The web build had bit-rotted: native-only features added since it last
compiled pulled crates that don't build for wasm32-unknown-unknown.
- printpdf (PDF export / printing) → native-only target; the web build gets
stub export_pdf / print_wires. It pulls a wasm-incompatible memchr 1.0.2
via lopdf → nom_locate, and the web has no filesystem.
- getrandom 0.3 (via ahash via acadrust) → enable its wasm_js feature and add
the matching --cfg getrandom_backend rustflag in .cargo/config.toml.
- ocs_plugin_api "host" feature (out-of-process plugin runtime: interprocess,
libloading, memmap2, rkyv) → native-only target; the web gets the
dependency-free manifest/ribbon contract. The host code was already
wasm-gated; only the Cargo wiring pulled it in unconditionally.
`trunk build --release` succeeds again.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The OSNAP settings tooltip read "Click ▾"; the ▾ (U+25BE) is absent from the
web font and showed as a tofu box. Tooltips are plain text and can't embed
SVG, so add a rich-body tooltip variant (tip_node) and use the existing SVG
caret (icons::arrow_down) — consistent with the SVG-icon approach the rest of
the chrome already uses for the web.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Bare PDMODE / PDSIZE / LTSCALE / CELTSCALE now enter a generic
ValuePromptCommand front-end that prompts for the value and delegates to the
existing inline `<name> <value>` handler via CmdResult::Dispatch, so they
prompt step-by-step like UCS. A bare Enter reports the current value;
`PDMODE 3` still works as one typed / headless line.
Only pure single-value setting commands are converted. Dialog-opening
commands (LAYER, PURGE, DIMSTYLE, STYLE …) intentionally keep their
bare-name behaviour (open the panel), and subcommand-tree commands keep
their inline handlers — all already typable with arguments since F1.
Adds headless test value_prompt_commands_inline_args. Fourth phase of the
unified step driver.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Space is now a literal character in the command line, so a whole command
line — `UCS Z 90`, `LINE 0,0 10,10`, `PDMODE 3` — is typed before Enter
instead of the first Space submitting the buffer. Enter tokenises the line
and runs it.
The command line and the headless automation feeder now share one runner,
`run_command_line` (moved to the non-wasm-gated cmd_result.rs): a single or
inline-argument command dispatches as-is; a multi-token line whose first
word starts an interactive tool feeds the rest as steps; plugin commands
get the whole line first (#162). The shared feeder is now UCS-aware —
typed coordinates are interpreted in the active UCS in both the GUI and
headless. `run_headless` is a thin wrapper over it.
A free-form text command (TEXT / MTEXT / a name with spaces) still receives
the whole line as one input, via the existing `wants_text_with_spaces`
guard. Unfocused Space still repeats the last command.
Closes the command-line half of #169 generically for every command, not
just UCS. Third phase of the unified step driver: every input source funnels
through feed_command.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Bare `UCS` now enters an interactive front-end (UcsCommand) that gathers the
option then its value as separate steps, instead of the command line
submitting `UCS` alone on the first Space. The front-end delegates execution
back to the existing inline `UCS <option> …` handler via a new
`CmdResult::Dispatch(String)`, so the coordinate-system math and the
save/persist path are untouched.
Now `UCS Z 90` works the same from the command line (Space-separated),
headless automation, and any source that feeds the active command — closing
the input half of #169. Inline `UCS Z 90` (one-shot / paste) still hits the
inline handler directly.
Second phase of routing input sources through the unified step driver: the
interactive command is driven by feed_command like any tool. Adds headless
test ucs_interactive_inline_args.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Introduce the single place every input source funnels through to drive the
active command's step machine: a `StepInput` vocabulary (one variant per
`CadCommand::on_*` method) and `OpenCADStudio::feed_command`, which routes
the input to the matching `on_*` and applies the result. Sources keep their
own classification (raw input → which StepInput); only the routing is
centralized.
Migrate the headless automation feeder onto it: feed_active_cmd builds a
StepInput, the terminator sends StepInput::Enter. No behavior change —
automation_ops_round_trip still passes.
Also fixes headless inline-argument loss for plugin commands (#162):
run_headless now offers the whole line to plugin dispatch first (plugins
parse their own inline args, e.g. `HC_PIPE 2B 2C 1.25 0.013`), falling back
to the first-word + fed-tokens path for built-in interactive tools.
First phase of routing all input sources (command line, dynamic input,
plugin API, viewport clicks) through one driver.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The PAN command/tool dispatched a command nobody handled, so it did
nothing. Combined with no middle mouse button on a trackpad/web client,
there was no way to pan at all (issue 164).
PAN (and the "P" alias) now enters a pan mode: a left-button drag pans the
view, reusing the existing middle-button pan path. Exit with Esc or by
starting another command. While active the crosshair is hidden and the
cursor becomes a hand (open when hovering, closed while dragging). PAN/P
are also added to command-line autocomplete.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
OFFSET derived the side from the first segment alone: which half-plane of
that edge's infinite line the pick fell on. For a closed loop that misreads
a pick placed beside the shape — outside the loop yet on the inner side of
the first edge's line — so a counter-clockwise rectangle (RECTANG's winding
when dragged bottom-left to top-right) offset outward by a side pick
collapsed inward instead.
Decide the side for a closed polyline with a point-in-polygon test mapped
through the loop winding (left normal points inward for a CCW loop): inside
→ inward, outside → outward, regardless of winding. Open paths keep the
first-segment heuristic. Adds a regression test covering both windings and
both directions.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Floating viewports on a survey-coordinate drawing rendered blank: the
model camera was correct, but content viewports whose saved view was
rejected auto-fitted onto empty space.
Two causes, both in model-space extents handling:
- compute_model_space_extents tested vertex finiteness in f64 but stored
f32. A long RAY/XLINE endpoint is a huge-but-finite f64 that overflows
to inf in f32, poisoning the extents to (inf, inf). Also, the wire/mesh
AABB paths checked only x while folding y in, so a vertical ray's
infinite y leaked through. Now every component is validated after the
f32 cast.
- camera_for_viewport framed its overlap test and auto-fit on the raw
extents centre. When a drawing has a second, far cluster (a small-
coordinate legend beside a UTM survey), that centre sits in the empty
gap, so a valid saved view fails the overlap test and the auto-fit aims
at blank space. cluster_extent_from_centers already computed a robust
centre (median of entity centroids) but discarded it; surface it as
Scene.local_center and frame on that instead.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Pull in acadrust 88ea9f0, which fixes the DWG VPort zoom-extents safety
net to be twist-aware. A twisted saved view on a survey-coordinate
drawing no longer has its camera flung to empty space on save/reopen
(the view_center was double-rotated about the origin).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Follow-ups after merging the out-of-process plugin isolation work:
- Convert the out-of-process interactive adapter's `on_point` /
`on_entity_pick` to `glam::DVec3`, matching the `CadCommand` trait,
which moved to f64 pick coordinates. The branch predated that change
and no longer compiled against the trait.
- Contain plugin constructor panics in `export_plugin!`: the constructor
runs across the C ABI boundary in `ocs_plugin_register`, where an
unwinding panic is undefined behavior. Catch it and return null, which
the loader already treats as a failed registration.
- Reap killed runner processes. The spawn-timeout and disconnect paths
called `child.kill()` without `wait()`, leaving a zombie on Unix until
the host exited. Route both — and `shutdown()` — through a shared
`reap()` that kills and waits on a detached thread.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
`clip_hatch_boundary` lifted each loop into the clip polygon's world
space (`+world_origin`) to run the Sutherland–Hodgman clip, but returned
the result in absolute world coordinates. The renderer adds the per-hatch
`world_origin` anchor back in the vertex shader, so clipped hatches with a
large origin (e.g. UTM drawings) were drawn double-shifted and lost f32
precision. Subtract the origin again on the way out so the output stays in
the same precision-preserving anchor space as the input.
Also drop a stale third argument from two xclip tests left over from the
world-offset removal, which had broken `cargo test` compilation.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Issue #170's fix added a "This PC" volumes root above the drive
letters on Windows. Generalize it so the Save As dialog can switch
volumes on every platform:
- parent_folder() now climbs the filesystem root into the volumes
list on all platforms (`/` -> Computer), not just Windows.
- available_drives() is platform-specific: Windows probes A:-Z:,
macOS lists `/` + /Volumes/*, Linux lists `/` + the removable
mounts under /run/media/$USER, /media/$USER and /mnt. Other Unix
and wasm fall back to `/`.
- drives_root_label() shows "This PC" on Windows, "Computer" elsewhere.
- read_dir_entries() lists the volumes at the drives root on every
platform.
Also gitignore the dist/ web build output so the wasm binary is never
committed by accident.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
ViewCube "already there → flip to opposite" tested the main (paper) camera,
which always looks straight down, so inside a floating viewport clicking
"top" landed on "bottom". Route the check through active_gaze_dir() — the
gaze of the camera the cube actually reflects.
Off-screen viewports:
- The active-viewport outline clamped only its origin while keeping full
width/height, shifting the whole border inward. Clip it to the visible
canvas (origin AND size) so it hugs the real viewport edge.
- Double-click activation tested each viewport's full paper rect, so a
click on empty space beside a viewport that runs off-screen matched its
partly-off-canvas rect and switched to it. Activation now keys on the
visible on-screen rect (viewport_at_screen_point), picks the smallest
(topmost) on overlap, and accepts the border-wire pick only when the
click also lands inside that viewport's visible rect — so a far
viewport's edge passing near the cursor can't steal the activation.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Clicking the ViewCube inside a floating viewport set only view_direction,
so the rebuilt camera fell back to WCS-up and the view opened tilted
relative to the UCS-oriented cube the user clicked. Build the full
UCS-aligned orientation (snap_to_direction with the ViewCube UCS) and
decode it to the viewport's stored (view_direction, twist_angle).
twist can't come from -Camera::roll(): camera_from_view rebuilds rotation
with its own yaw convention (atan2(x,-y)) and roll = -twist, which differs
from the camera's internal sync_yaw_pitch. Instead reproduce the decoder's
zero-twist basis and store twist = -signed_angle(up0 -> desired_up).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The hover hit-tests used the bare camera rotation (no UCS) and the main
camera even inside a floating viewport, so the highlighted face diverged
from the cursor whenever a UCS was active or in a paper viewport — while
the snap path was correct. Route hover, click-consume and snap through
active_view_rotation_mat(), and compose the ViewCube UCS in its viewport
branch so all three match ViewportData's drawn cam_rotation exactly.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The whole pipeline is relative-to-eye now, so the f32/f64 split is
vestigial. Merge eye_f64 into eye() returning DVec3; direction-only
callers take .as_vec3().
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
With the relative-to-eye double-single render path, the per-defn
`local_offset` precision origin is redundant: block geometry can be stored
absolute and stays precise via the high/low residual. Remove it
end-to-end:
- Drop the Pass-1 centroid computation, the BlockDefn.local_offset field,
and the `defn_lo` parameter threaded through emit_wire /
emit_greeked_text / emit_text_baseline / transform_tangent and the AABB
helper (transform_offset_aabb_xy → transform_aabb_xy), plus the nested
parent/child offset re-basing.
- Store text_obb_local as absolute f64 (was offset-relative f32) so the
greek rectangles stay precise; emit splits to double-single. Give the
greek-baseline points a low residual too.
Geometry is now absolute + double-single throughout the block cache,
matching the top-level path.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The function computed a median-centroid offset that is now always
discarded (geometry is absolute + double-single). Rename it to
cluster_extent_from_centers, return only the cluster span used for camera
fit / cull, and drop the dead centroid computation. Rename the companion
WORLD_OFFSET_SANE_EXTENT const to CLUSTER_SANE_EXTENT.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The GPU draws exclusively via the rotation-only `view_rot` + eye double-
single; the full view-projection survived only to project AABBs in the
CPU frustum-cull / scissor / LOD passes. Migrate those to the same
relative-to-eye math (project `point - eye` through `view_rot`) — pixel-
equivalent but precise at UTM scale — then drop the dead uniform:
- pipeline cull/scissor/LOD helpers take (view_rot, eye) instead of the
full view_proj; render.rs reconstructs eye from the uniform's
eye_high/eye_low and passes the cropped view_rot.
- Remove the `view_proj` field from the shared Uniforms struct and all six
shader structs that only declared it for layout (none read it), and
delete Camera::view_proj(). The struct shrinks by one mat4 on both
sides, keeping the Rust / WGSL layouts in lock-step.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The world_offset removal stripped two precision mechanisms in the block
cache that were masked while geometry was offset-relative (small coords)
but broke once coordinates became absolute UTM:
- build_defn no longer subtracted the defn's `local_offset` from the
tessellated wire (it was passed as tessellate's world_offset). The defn
was stored absolute while expand still re-added local_offset → block
lines shifted. Restore the subtraction (with a double-single re-split)
locally in tessellate_sub_local.
- Block fill triangles (greek-text rectangles, block solid fills) were
batched with f32-only `fill_tris` and an empty `fill_tris_low`, so at
UTM scale the absolute f32 quantized to ~0.5 m and the greek rects
sheared. Plumb `fill_tris_low` through BatchEntry and emit the
double-single residual from emit_wire and emit_greeked_text.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Clean up the last zero-valued world_offset locals outside the scene
tessellation core: the UCS wire-affine / grid-basis origin shift
(document.rs), the coordinate readout (view.rs), grip absolute-apply
(update.rs), the dead properties-edit re-offset guard, the MTEXT preview
glyph-box shift (mtext_editor.rs), and the cross-drawing paste correction
(commands.rs / cmd_result.rs) — including the now-dead
`clipboard_world_offset` field. Coordinates are absolute throughout.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Clean up the remaining `let [ox,oy,oz] = [0,0,0]` bindings and their
zero-valued add/subtract arithmetic across scene/mod.rs (mesh LOD resplit,
model AABB, wide-polyline fills, wipeout boundary, hatch/solid models,
insert/entity AABB), table, multileader and xclip. Coordinates now read as
the plain absolute values they are. No world_offset reference remains.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Object-snap-tracking points now project through the relative-to-eye path
(view_proj_rte + eye) instead of the full view-projection, so they stay
precise at UTM-scale coordinates — the last consumer of the legacy full
view_proj outside the CPU frustum-cull / scissor path.
- Rename Camera::project_f64 → project and unproject_on_plane_f64 →
unproject_on_plane: the f32 twins are gone, so the disambiguating suffix
is vestigial. eye_f64 keeps its name — the f32 eye() still builds the
view matrices.
- Remove stale #[allow(dead_code)] from eye_f64 / project / unproject_on_plane
(all live now).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
In the files where ox/oy/oz came solely from the now-zero offset
destructure (tessellate, block_cache, dimension, leader, text_support),
remove the `let [ox,oy,oz] = [0,0,0]` binding and the `- ox` arithmetic so
the coordinate math reads as the plain absolute value it now is.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Now that geometry reaches the GPU/CPU as absolute coordinates via the
double-single relative-to-eye path, world_offset is always [0,0,0]. Strip
the parameter that was threaded through the whole tessellation / block-
expansion / fallback / camera-decode chain and the ExpandCtx field that
carried it — ~340 references across 20 files, all subtracting zero.
- truck_tess: to_local / to_local_low / tessellate_* drop the offset arg;
to_local is now a pure double-single split.
- tessellate: offset_to_ds → points_to_ds (pure DS split); tessellate(),
tessellate_entity(), fallback_geometry(), solid_wire_fallback(),
entity_aabb(), expand_insert(), expand_block_meshes(), the dimension
helpers (vec3_local, dimension_snap_pts, …), text_support, leader,
multileader, image_model, xclip and camera_from_view all lose the param.
- block_cache: ExpandCtx loses its world_offset field.
- Remove the now-dead offset_snap_pts and the unused set_grid_snap.
Behaviour is unchanged (the offset was zero everywhere); the compiler
verifies every call site.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The GPU scissors floating-viewport content to the viewport rectangle, but
the CPU hit-test/snap wire set runs past it (the frustum cull keeps a
margin and lines extend beyond the rect), so box selection, lasso, click
and object snap could reach geometry clipped out of the viewport.
Reject candidates whose projected screen point falls outside `bounds` (the
pane rectangle), mirroring the scissor. `bounds` is the full canvas in
model space, so every change is a no-op there.
- snap: reject out-of-bounds points in try_pt, the grid block and tangent.
- click_hit / click_hits_all: no hit when the cursor is off-pane.
- box_hit: clamp the selection box to the pane rectangle.
- poly_hit: drop wire points projected off-pane.
- hatch click: no hit when the cursor is off-pane.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
camera_from_view computed the camera target in f32 (view_target cast to
f32 before folding view_center), quantizing it to ~0.5 m at UTM scale.
Panning/zooming inside a floating viewport nudges view_target by sub-metre
f64 steps, so the camera snapped along the f32 grid and the drawn content
jittered — while the model view (f64 target) stayed smooth.
Compute the target in f64 (axis directions stay f32 — orientation only),
matching the model camera's precision.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Extend UCS support to in-viewport (MSPACE) editing through the existing
single-converter structure: one active_ucs drives every UCS-aware system,
now in the Model tab AND inside a floating viewport.
- Single predicate DocumentTab::editing_model_space() (Model tab or active
viewport) replaces the model-only gate in ucs_wire_affine,
ucs_rotation_angle, ucs_grid_basis, viewcube_ucs_mat, the coordinate
readout and the cursor-plane pick.
- Per-viewport UCS: refresh_active_ucs() sets active_ucs to the pane's UCS
on every transition (the entered viewport's own UCS from its
ucs_origin/x_axis/y_axis/ucs_per_viewport fields, the header model UCS in
the Model tab, none in plain paper); persist_active_ucs() writes changes
back to the viewport entity or the header so they round-trip. Wired into
EnterViewport/ExitViewport, layout switch, load and the UCS command.
- Visuals: grid_views now returns each pane's handle so every content
viewport's grid aligns to its own UCS; the UCS icon draws inside the
active viewport projected through the viewport camera.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Replace the leaky paper-projection approach to MSPACE editing with one
coordinate adapter that routes all interaction through the viewport's own
camera — the exact camera+rect the GPU draws the content with.
- Add Scene::viewport_edit_frame(canvas) -> (camera_for_viewport,
viewport_screen_rect): the unified adapter. Inside a floating viewport
the viewport is the active pane, so snap/pick/grip/preview/DYN run in
model space and land on the pixels the GPU draws. Tracks pan/zoom/twist
exactly (twist handled by camera_from_view); crop is pixel-preserving so
full-rect projection + cursor offset matches.
- Route snap-move, click-pick, grip-drag, grip hit-test, grip render,
hover (HoverDwellTick), selection (single/box/poly/cycle) and command
entity-pick through it via pick_view / cursor_model_point helpers and an
edit_cam-aware "active pane = viewport rect" override, so the existing
tile offset plumbing just works. Fixes wrong-object hover/selection and
stale snap after pan/zoom.
- DYN: project anchors with the viewport camera and shift to canvas space
so the guide/labels sit correctly (also fixes a latent tiled-model bug).
- Separate grid snap from object snap entirely: new Snapper.grid_snap_on
flag, Grid removed from the osnap mode set, evaluated independently of
the osnap master. Object snap never catches grid; grid corners are
picked only when grid snap is on, at lowest priority. Default off.
- grip.rs: project_rte + grips_to_screen_rte / find_hit_grip_rte taking
(view_rot, eye) so grips project through the composed view.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Checkpoint on the way to unified in-viewport editing.
- Drop the _f64 transition suffixes: pick_on_plane / pick_on_target_plane /
paper_to_model are single f64 functions (wrappers removed).
- Paper-space hit-test/snap projects viewport content with points_low, and
projects its aabb / key_vertices / snap_pts into paper coords so click
picking and snap candidates land on the visible sheet geometry.
- camera_for_viewport + paper_to_model + EnterViewport agree on an absolute
drawing-centre view (normalize_active_viewport_view) so UTM viewports
render instead of aiming at the origin.
- model_to_paper inverse; dynamic-input anchors map model→paper before the
paper camera projects them, so the guide no longer tessellates a 512 MB
buffer (clicking the first in-viewport LINE point) — crash fixed.
Known-incomplete (next: unify on the model-space path via one coordinate
adapter): in-viewport snaps stale after pan/zoom (paper_projected_cache not
keyed on the viewport view), grips don't show, command preview misplaced.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Floating viewports rendered blank on UTM drawings. camera_for_viewport
centred both the saved-view overlap test and the auto-fit target on the
ORIGIN — correct only while world_offset re-centred the model there. With
world_offset removed the model sits at absolute (~5.7e6) coordinates, so a
stale (0,0,0) saved view failed the overlap test AND the auto-fit aimed at
empty origin, leaving the viewport empty.
Both now use the absolute drawing centre from model_space_extents, so saved
views that frame the model are kept and the auto-fit falls back onto the
real content.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
entities_centroid summed the f32 points (double-single HIGH half) alone, so
at UTM scale (~5.7e6) it both quantized each term ~0.5 m and lost low bits
across the running total — drifting the paste base and PASTEBLOCK's block
base metres off, which placed the wrapped block at the wrong spot/scale.
It now reconstructs each vertex's absolute f64 from points + points_low and
accumulates in f64. clipboard_centroid is DVec3; the PASTECLIP commit delta
is fully f64; define_block_from_owned_entities takes an f64 base so the
block-local fold (Translate(-base)) stays precise. The interactive ghost
anchors (PasteCommand, InsertBlockCommand) still take the f32 cast — display
only; the committed geometry is f64.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Selection projected wire vertices from points (the double-single HIGH half)
alone, ignoring points_low, so at UTM scale every selection test ran against
coordinates ~0.5 m off — box/crossing/window edges and click hit-tests missed
or grabbed the wrong entities, worsening with zoom.
hit_test now reconstructs each vertex's absolute f64 from points + points_low
(wp64) before projecting, across click_hit, click_hits_all, box_hit and
poly_hit (lasso). Hatch hit-tests use the f64 world_origin instead of an
f32-cast, and mesh hit-tests (click/box/lasso face picking) reconstruct from
verts + verts_low. world_to_screen now takes the f64 point directly.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Zooming in close made the drawing vanish. The orthographic near/far were
near = distance*0.001, far = distance*1000, which parks the target plane at
ndc-z ~= 0.001 — flush against the near plane. The shared draw-order depth
bias shifts clip-z by up to DRAW_ORDER_BIAS (0.001), so a front-biased
entity landed exactly at z = 0 and dropped out as soon as f32 rounding (which
worsens at high zoom) tipped it past the plane.
ortho_depth_range() now returns a symmetric near/far (distance ± r) that
centres the target plane at ndc-z ~= 0.5, giving the bias half the depth
buffer of headroom on each side. Perspective is unchanged. Applied to both
view_proj and view_proj_rte.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two UTM-scale rendering defects.
- UCS icon jittered: it projects only axis DIRECTIONS but went through the
full view_proj, whose ~5.7e6 translation cancels catastrophically in f32.
Feed it the rotation-only view_proj_rte instead (translation dropped).
- Fallback outline geometry (hatch boundary, viewport/insert/ole2frame
frames, Solid3D edge wires) built absolute f32 points with an empty
points_low, so at UTM the outline sat ~0.5 m off its own fill (the hatch
fill renders precisely via the world_origin-relative HatchModel pipeline).
FallbackGeometry now carries absolute f64 points + key vertices; the
tessellate dispatcher splits them into the double-single high/low pair the
relative-to-eye wire pass needs, so outlines land exactly on their fills.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Consolidate the duplicated paste logic and extend XCLIP-carry to every path.
- finalize_paste(): single shared step for PASTECLIP and PASTEORIG —
merge deps + block defs + xdictionary graphs, add clones (optionally
transformed), tessellate solids. A new cross-drawing concern is now wired
once instead of re-implemented per command (the gap that lost the clip on
PASTEORIG in the first place).
- recreate_clipboard_ext_roots() + recreate_ext_subtree(): PASTEBLOCK now
recreates each entity's xdictionary graph and stamps the new root onto the
wrapped entity before defining the block, so the block's nested insert
keeps its spatial filter.
- block_cache: a nested insert's clip is baked into NestedRef.clip_poly at
build time (the spatial filter isn't reachable at expand time) and applied
on expansion — the clipped wires are finalized in an isolated batch set,
clipped, and carried out via Batches.extra_wires. Single-instance only;
arrays of clipped nested inserts fall through unclipped.
All three paste paths (PASTECLIP / PASTEORIG / PASTEBLOCK) now preserve an
XCLIP clip across a cross-drawing paste.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Clipped blocks were broken at UTM and lost their clip when pasted into
another drawing.
- Clip geometry: world_clip_polygon_f64 keeps the boundary in absolute f64;
clip_wires now clips in a frame relative to the boundary's first vertex
(small coords → exact f32 Sutherland–Hodgman) and re-splits the result
into the double-single high/low pair the relative-to-eye renderer needs,
so the clipped block draws correctly and no longer poisons ZOOM Extents.
- ZOOM Extents (fit_all): the outlier reject compared absolute UTM
coordinates against a centre-RELATIVE span (local_extent_max), rejecting
the whole drawing → no-op. Made the reject distance-from-median-centre.
- Cross-drawing paste: clipboard now snapshots each copied entity's whole
xdictionary object graph (ClipboardDeps.ext_objects) and recreates it on
paste with fresh handles + remapped references, re-pointing the pasted
entity's xdictionary at the new root. Wired into PASTECLIP and PASTEORIG.
Fixed handle allocation (allocate_handle, not the non-advancing
next_handle) so the dictionary chain doesn't collapse onto one handle.
- Updated hit_test / plugin_host tests for the eye-relative + DVec3 signatures.
Known gaps: PASTEBLOCK does not yet carry the clip (clip would need to live
inside the new wrapper block definition); hatch-clip path still f32.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Computed snaps (midpoint/nearest/perpendicular/intersection/extension/
apparent/grid) reconstructed the snap point from f32 `wire.points`
(absolute high half), so at UTM-scale they landed ~0.5 m off the geometry
while endpoint/center (sourced from f64 key_vertices/snap_pts) stayed
correct. Now they rebuild each vertex from points + points_low (the
double-single low residual) and run the geometry in f64:
- snap() cursor + geometry helpers (nearest_on_segment, perp_foot,
seg_intersect_xy, extension_snap) are DVec3; wp_f64() reconstructs each
vertex; midpoint uses f64 key_vertices directly (was downcast to f32).
- snap callers pass the precise DVec3 cursor.
Snap capture at deep zoom: the wire-AABB cull compared a sub-millimetre
snap radius against f32 bounds quantized by ~1 ulp (~0.7 m at 5.7e6),
wrongly excluding wires under the cursor. Pad the cull by the bound's own
f32 quantization so it never rejects an in-range wire.
Rubber-band previews (line/circle/arc/ellipse/rect/polygon) built f32
absolute points with an empty points_low, so the wire pass (relative-to-
eye) drew them ~0.5 m off the cursor at UTM. Add WireModel::solid_f64,
which splits f64 points into the double-single high/low pair; previews now
track the cursor exactly.
Known: snapping at extreme zoom on large drawings spikes CPU (per-move
snap cost) — deferred.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Whole command/snap/grip/cursor chain now builds on f64 DVec3 end to end,
so picked points, snaps and grip drags stay precise at UTM-scale
coordinates instead of quantizing to ~0.5 m through f32.
- All ~80 command files migrated: draw (line/circle/arc/ellipse/shapes
in full f64; others via point methods), modify, annotate, insert, view,
inquiry, layers, groups, clipboard, layout.
- Entity apply_grip impls updated for GripApply::{Absolute,Translate}(DVec3).
- Construction sites (EntityTransform/CmdResult/DynAnchor) emit DVec3;
wire-preview boundaries cast to f32 only at the GPU edge.
- Modify/annotate commands keep internal Vec3 geometry where precision is
non-critical, casting at the DVec3 boundary; key draw commands store f64.
Builds clean (no warnings). Visual verification of crosshair/snap/grip at
extreme UTM zoom pending.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Route the remaining absolute-coordinate CPU projection through the
relative-to-eye path so it no longer multiplies by a view_proj matrix
carrying a ~1e7 translation (which cancels catastrophically in f32 after
world_offset removal):
- camera pick_on_plane / pick_on_target_plane delegate to
unproject_on_plane_f64 (rotation-only inverse + f64 eye)
- snapper (world_to_screen), hit_test (click_hit / mesh / hatch), the
scene mesh/solid hit methods, grid + UCS axes, and polar_constrain_near
all take view_rot + eye and project relative-to-eye
- compute_grid_step measures the per-unit scale via the rotation-only
matrix (no eye term needed)
view_proj remains only for conservative CPU frustum-cull / scissor and the
screen-space viewcube / UCS-icon directions. NOTE: SnapResult.world and
the command-point chain are still Vec3 (f32) — a ~0.5 m quantization at
extreme UTM zoom remains; the command layer goes f64 next.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Route the on-screen chrome that projects absolute world coordinates
through the f64 relative-to-eye helpers so it stays glued/jitter-free at
UTM scale after the world_offset removal: grips (grips_to_screen) and
grip hit-testing (find_hit_grip) now project from the f64 GripDef.world
via Camera::project_f64; the model grid and its UCS axes carry the camera
eye (DVec3) + rotation-only matrix and a f64 origin, projecting each line
relative to the eye. Snap-marker / OST-point / compute_grid_step still use
the f32 path (single points / spacing — negligible) and can follow.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Delete the world_offset field from Scene and DerivedCaches and the
load-time assignment; every reader now uses a zero origin. Geometry is
absolute and the double-single relative-to-eye pipeline keeps it precise,
so the coarse origin is gone. world_offset_from_centers still derives
local_extent_max (camera/cull span) but returns no offset. The remaining
`world_offset: [f64;3]` function parameters are now always zero — dead
plumbing left for a later mechanical sweep; behaviour is unchanged.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
world_offset_from_centers always returns [0,0,0]: geometry reaches the
GPU as absolute world coordinates and the double-single relative-to-eye
pipeline keeps it precise at UTM scale, so the coarse origin is no longer
needed. Camera target, snap candidates (f64) and the coordinate readout
all operate in absolute coordinates now. The `- world_offset` subtractions
remain as no-ops and are removed in the field-deletion step. local_extent_max
(camera/cull span) is still derived from content.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
WireModel/LocalWire/BatchEntry key_vertices are now Vec<[f64; 3]>;
tessellation keeps them in f64 (no f32 cast at offset subtraction),
producers (dimension, leader, multileader) build them from the acadrust
f64 source, the block-cache transform keeps f64, and the face3d Face3D
quad path splits them into the double-single (high, low) pair. Consumers
(snap endpoint/midpoint, xclip) read via f32 at the leaf. Still
offset-relative (inert); endpoint/midpoint snap is no longer f32-capped.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
TruckEntity.snap_pts is now DVec3 and every entity producer builds snap
candidates straight from the acadrust f64 coordinates instead of casting
to f32 first (text, mtext, arc, circle, ellipse, point, shape, solid,
mline, mtext, attribute, tolerance, table, underlay, ole2frame, mesh,
multileader, dimension, block-cache). offset_snap_pts stays f64
throughout. Still offset-relative (inert) — but snap precision is no
longer capped at ~0.5 m by an early f32 cast, which fixes a latent
UTM-scale snap error and sets up the absolute-coordinate switch.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
WireModel/LocalWire/TruckEntity snap buffers and the producers/consumers
now carry snap candidates as DVec3 (offset_snap_pts, dimension/multileader
producers, block-cache transform, pick/xclip, snap module). Still
offset-relative, so behaviour is unchanged — this just lets the snap
transport survive the upcoming switch to absolute coordinates without an
extra f32 quantization. The snap module still resolves in f32 (as_vec3 at
entry), matching today's precision.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Add camera-relative-f64 world↔screen on Camera (project_f64,
unproject_on_plane_f64): the point is made eye-relative in f64 before the
rotation-only projection, so CPU picking / overlay stay exact at large
absolute coordinates — the CPU twin of the GPU's relative-to-eye path.
Additive and inert for now; the upcoming phases route overlay, pick/snap
and coordinate readout through these so world_offset can be dropped.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
zoom_about_point picked the world point under the cursor twice via
view_proj.inverse() (the full f32 matrix) and shifted target by their
difference. At UTM coordinates that inverse carries ~0.5 m of f32 error
that didn't cancel between the two picks, so target drifted every zoom
step and the whole scene jumped as a block.
Compute the pivot as the cursor's offset *relative to target* in the
camera frame (small numbers, rotated to world) instead — it never touches
the large absolute target, so the before/after difference is exact and
the scene stays put while zooming.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
No GPU pipeline reads camera_pos any more — lighting and all transforms
moved to the relative-to-eye path (view_rot + eye_high/eye_low). Remove
the field from the shared Uniforms and every shader struct, plus the now
unused Camera::position_vec4. view_proj stays: the CPU frustum-cull /
scissor / LOD passes still project offset-relative AABBs with it (it is
no longer part of any GPU draw).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
3D solid / surface meshes were the last single-f32 path. truck's mesh
tessellation now returns a paired verts_low residual (TruckTessResult::Mesh,
polygon_to_result), MeshModel/MeshLodSet carry it, the model-space offset
and block-instance transforms reconstruct the f64 position and re-split it
after subtracting world_offset, MeshVertex gains position_low (location 3),
and mesh.wgsl reconstructs the eye-relative position. Native SAT / ACIS /
OBJ mesh paths default to a zero residual (no regression — they were never
sub-f32 precise).
With this every world-space pipeline (wire, hatch, image, face3d, mesh)
is full double-single relative-to-eye; large-coordinate jitter is gone.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Table / PolyfaceMesh / greek-text / mleader-background fills go through
the face3d pipeline; their vertices were single f32 and quantized at UTM
scale. WireModel now carries a fill_tris_low residual (filled by the
tessellate Lines path), Face3DVertex gains position_low (location 3), and
face3d.wgsl reconstructs the eye-relative position from the pair. 3DFACE
quads built from key_vertices use a zero residual (no low source).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Raster-image quad corners derive from a large f64 insertion point plus
small u/v spans; the single-f32 cast quantized them at UTM scale. The
ImageModel now keeps a corners_low residual, ImageVertex carries pos_low
(location 2), and image.wgsl reconstructs the eye-relative position from
the high/low pair.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The hatch anchor (world_origin / pattern origin) was cast to a single
f32, quantizing UTM-scale fills to ~0.5 m and reintroducing jitter the
wire path no longer had. Both the batched and per-hatch pipelines now
carry the anchor as a high/low f32 pair (HatchInstance.world_origin_low,
HatchUniformData.origin_low) and reconstruct the eye-relative position in
the shader as (origin_high - eye_high) + (local + origin_low - eye_low).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Vertices now carry the low residual alongside the high half all the way
to the GPU: tessellation splits the offset-relative f64 coordinate into
(high, low) f32 (TruckTessResult::Lines/Point, the Text path, Lines /
SegmentedLines, and the block-cache instance transform), WireModel and
the block batch carry a paired points_low buffer, WireInstance gains
pos_a_low/pos_b_low (locations 10/11), and wire.wgsl reconstructs
rel = (pos_high - eye_high) + (pos_low - eye_low).
This is the piece the camera-only RTE was missing: at UTM-scale
coordinates the single-f32 vertex was quantized to ~0.5 m, so it snapped
each time the eye crossed an f32 boundary on pan. With the low half the
position is exact and the cross-drawing-paste jitter is gone.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Foundation for double-single positions: WireModel now carries an
index-paired f32 residual buffer (`points_low`) next to the existing
`points` high half. Empty means "all-zero residual", so every existing
constructor (interactive draw, previews) is unchanged in behaviour;
tessellation from CAD f64 will fill it in the next steps so vertex
precision survives at UTM-scale coordinates.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Switch the remaining world-space shaders (mesh, image, face3d, hatch,
hatch_batched) to the same relative-to-eye transform the wire pipeline
uses: subtract the eye (high+low f32) per vertex and multiply by the
rotation-only view-projection. Each shader's Uniforms struct is widened
to the full shared layout so the appended RTE fields land at the correct
offsets. Math is unchanged (proj·R·(pos−eye)); large coordinates no
longer jitter on camera movement.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The wire shader now subtracts the eye (high+low f32) per endpoint and
transforms by the rotation-only view-projection instead of multiplying
absolute positions by a view-projection that bakes in a large eye
translation. The math is identical (proj·R·(pos−eye)) but the large eye
no longer enters the f32 matrix, so lines stop jittering on
pan/zoom/rotate at large (UTM) coordinates and far from world_offset.
Geometry low bits are intentionally dropped: (pos − eye_high) is exact in
f32 for same-magnitude operands (Sterbenz) and eye_low corrects the eye,
so the only residual is a static sub-f32 position error, not motion
jitter — no per-vertex low buffer needed. view_rot is cropped in lock-step
with view_proj for tiled panes.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Append a rotation-only view-projection plus the eye split into high/low
f32 (double-single) to the shared Uniforms buffer, and the matching
Camera helpers (view_proj_rte, eye_high_low). Fields are added at the end
so existing field offsets — and every shader still reading only the
legacy view_proj — are unaffected. No shader consumes them yet; the wire
pipeline migrates first in the next step.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Foundation for relative-to-eye rendering: the pivot the camera orbits
must stay precise at large offset-relative coordinates so the derived eye
(which the RTE path subtracts per vertex) is exact. `Camera::target` is
now DVec3 and `eye_f64()` exposes the full-precision eye; `view_proj`
still casts to f32 at the matrix, so behaviour is unchanged for now.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
PASTECLIP recreated missing block definitions, tessellated pasted solids,
and refreshed the layer panel after a paste; PASTEORIG and PASTEBLOCK did
not. A block pasted through either of those rendered empty because its
definition was never recreated in the target drawing.
Both paths now call merge_clipboard_blocks (so referenced/nested block
definitions exist) and populate_meshes_from_document (so pasted ACIS
solids tessellate); PASTEORIG also refreshes the layer panel.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Single-line TEXT and ATTRIB/ATTDEF cast the insertion/alignment point to
f32 before computing the stroke origin. At large coordinates (UTM ~4.5M)
the f32 ulp is ~0.5, snapping text baselines onto a coarse grid so
adjacent rows collapse onto each other and overlap. Keep the anchor in
f64 throughout.
Horizontal alignment now uses the pen advance box [0, advance] instead of
the inked min_x/max_x, so leading/trailing spaces keep their width and
space-aligned columns no longer shift. text_local_bounds returns
TextLocalBounds { ink_min, ink_max, advance }.
TABLE laid out its whole grid in an f32 world frame; switch to a local
f32 frame plus an f64 base added at the widening step, fixing cell-text
baseline snap and the i32 overflow in the border-dedup keys.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
New annotations now place their text and reading geometry along the
active UCS X axis (identity = world), so they read square to the user's
coordinate system instead of world-horizontal:
- TEXT / MTEXT: rotation defaults to the UCS X angle
(DocumentTab::ucs_rotation_angle, model space only).
- Linear dimensions: measure along the UCS axis the span is nearest, and
bake the UCS X angle as the text rotation so the value reads square
even under a style that forces horizontal text. DIMCONTINUE / DIMBASELINE
derive their axis from the base dim's rotation (any angle), not a
world H/V test.
- LEADER: the MText annotation rotates to the UCS, and the hookline runs
along the leader's horizontal_direction (set to UCS X), used by the
renderer instead of world horizontal.
- MLEADER: context.text_direction = UCS X drives both the text rotation
and, in the renderer, the text side and dogleg/landing direction.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Commands that construct geometry with implicit axis directions ignored
the UCS, so a rectangle drawn under a rotated UCS came out world-aligned.
Give commands the UCS and build square to it:
- CadCommand::set_ucs(Mat4) (UCS→wire affine, default no-op); the host
pushes the current UCS via App::push_ucs_to_cmd before each point /
preview / text dispatch (DocumentTab::ucs_wire_affine, identity outside
model space). New axis-aligned commands just override set_ucs.
- RECT (two-corner) and RECT_CEN (center) place their derived corners on
the UCS axes (ucs_box_corners / ucs_box_around_center).
- ARRAYRECT row/column offsets run along the UCS axes.
RECT_ROT, POLYGON, and ELLIPSE already take their orientation from picked
points, so they need no change.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The dynamic-input boxes displayed deltas/angle in the UCS, but
dyn_resolve_point still resolved the typed (and live) values on the world
axes, so committing a typed coordinate ignored the UCS. It now rotates
the cursor delta into the UCS, interprets the typed cartesian/polar
values in that frame, and maps the relative result back to world via
UcsXform::vec_to_wcs — display and commit now agree. Absolute first-point
fields (no base) stay world; identity UCS is unchanged.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two leftovers from the UCS grid work:
- The coloured X/Y/Z axis lines (draw_axes) were drawn at the world
origin along world axes; they now run through the UCS origin along the
UCS axes, matching the grid lines.
- Grid snap rounded the cursor on the world XYZ lattice; it now rounds in
the UCS grid frame (to-UCS, round, back-to-world) so it lands on the
grid the user sees. The UCS origin (wire space) + rotation reach the
snapper via DocumentTab::ucs_grid_basis; identity outside model space.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
OTRACK alignment rays (0°/90° and polar steps, plus the polar pencil from
the last point) were built along world X/Y. They now rotate with the
active UCS via a Mat4 passed into otrack_snap (scene.viewcube_ucs_mat() —
the UCS in model space, identity in paper), so tracking lines and the
intersection lock align to the user's coordinate system like ortho,
polar, and the grid. sync_ucs_to_scene now reuses UcsXform::rotation_mat.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Clicking a ViewCube face rotated the gaze into the UCS frame but derived
the up vector from world X/Y/Z, so the snapped view landed world-aligned
(most visibly Top, where the UCS shares world Z and only the in-plane
roll differs). snap_to_direction now takes the UCS basis and picks its up
from the UCS axes, so a face snap lands square to the user's coordinate
system. Identity UCS reproduces the world-aligned snap.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The on-canvas dynamic-input boxes measured relative X/Y deltas and the
polar angle in world axes. They now read in the active UCS plane (via
UcsXform::vec_to_ucs on the offset-invariant delta), matching the UCS
ortho/polar constraints and coordinate readout. Identity UCS reproduces
the world-frame values; absolute first-point fields are unchanged.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
