* claim: #6646
* Control channel: document.revision, a design revision and a document id (#6646)
A program that follows the open design asks document.revision and exports
only when the answer moves: document_id names the open design (kept through
edits, saves and renames, another for each design started or opened, never
shared between two tabs or two runs) and revision moves whenever the design,
what is hidden or an uncaptured component position changes. It stays put on
reads, selections and view changes, unlike the session's view revision.
The revision is read lazily: the session remembers the design it last
reported (held weakly, so Arc::make_mut can never change it in place and its
address can't be reused) and takes a new number from one counter for the
whole run when the design is another one. The control method answers at once,
also while a slow command computes; engine.execute reaches the same command
for scripts and MCP.
* document.revision: hash the run's id prefix with the process's random keys
A clock set before 1970 gave every run the prefix 0, so a restarted app could
answer a new design with the id and revision of the old one. The docs no
longer promise an answer at once: a screenshot waiting for an opened design to
be meshed still holds the requests behind it.
* document.revision: its label in the complete catalogs (ja, zh)
* claim: #6985
* Control channel: control.subscribe and design.changed events (#6985)
A connection that sends control.subscribe gets the active design's id and
revision in the reply (as document.revision answers), then a
{"event":"design.changed","params":{document_id, revision}} line whenever
either changes, so a live link need not poll.
The window publishes the design at the end of each frame's logic and
drawing while a connection is subscribed (one atomic load otherwise). Each
subscribed connection waits on that from a thread of its own and writes
the latest after a 100 ms gap, so a burst of changes is one line. Replies
and events share one lock per connection, each written whole (#6390).
* control.subscribe: the reply before any event, a second one refused
The events thread started before the subscribe reply was written, so a
change right after the window's answer could reach the client before the
reply that named the older revision. The subscription now starts holding
the connection's line lock and writes the reply under it.
A second control.subscribe on a subscribed connection is refused on the
connection itself: its answer could arrive behind a newer event. When no
thread or socket is left for the events, the reply says it did not
subscribe instead of ok.
* control.subscribe: a second one does not lose a finished late reply
The already-subscribed answer came before the busy check, so a request that timed out and had finished since lost its reply: control.pending then said the app ended it without one. It now comes after the busy check and drops a finished late reply as every other request does.
* control_server: declare the subscribe guard at the top of the loop
Clear of the lines #6643 changes (its token check moves to where the declaration was).
* control_server: the subscription's items beside serve
Away from the lines next to send, where #6643 adds token_of.
* control test: New on a blank design reuses its tab, so redo the box first
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #3713
* Exploded views export as an animation (#3713)
motion.export_animation takes an exploded view (view: id|name) as well as a
study or a joint: the parts come apart at a steady pace from the assembled
placement to where the view shows them, and back together with back: true
(exploded at the middle frame). 41 frames by default, 81 with back. The home
view frames the parts both assembled and exploded. As with explode.show, the
design keeps its placements, its history and its saved state.
explode.show and the export share the moves; moves a design file gives with
no finite translation are left out.
* Mechanisms skill: an exploded view's default frame count
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #3330
* Linked components: joints keep their face after Update Links (#3330)
A linked design comes in as STEP text, which keeps no face names, so the
faces of its bodies were named by their place in the file (`face:<k>`).
An edit of the part that adds a face earlier in the file (a second hole
of the same size) gave that place to another face, and a joint on the
linked part moved there silently.
When a link is made or updated, each imported face is matched to the
part face it was written from (the same geometry), and the Import
feature keeps that face's name in the part (`Feature::link_faces`, a new
pending format change). The linked face is named
`F<import>🔗<digest of that name>`, pieces keeping their qualifiers,
so references follow the part's own history through Update Links as they
do inside the part, and a face the part no longer has is reported.
Linked faces still answer to `face:<k>`, so older references keep
resolving; Update Links warns about joints that use them and reports
joints that no longer solve.
* Linked faces: a solid's twin keeps its own names; renamed faces count as an update
Two solids of a part alike in one place each take the names of the solid
the file names them after. Update Links reports a link whose part named
its faces differently since as updated (a joint may follow that name to
another face), not as unchanged; only a link getting its first names
stays unchanged.
* Linked faces: Extract to File carries the design's face names too
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Co-authored-by: Claude Sonnet 5.5 <noreply@anthropic.com>
* claim: #3220
* Model cache: keep fewer states of a design too big for its file instead of none (#3220)
A model cache over the container's budget keeps the final state, then the
slowest features' states while the bodies they add fit (a body several
states hold counts once), and is left out only when not even the final
state fits. Opening evaluates just the features whose state the file left
out, each from the state before it, so the model is whole afterwards.
* Model cache: keep each encoded body apart, copy only what a cache stores
Encoding a model too big for its file no longer copies every distinct body
into one buffer before thinning: the payload is copied from the encodings
once its size fits, so saving holds no more than before.
* Model cache tests: canonical runs in the timeline fixture; rustfmt
* Model cache: count the records a thinned cache keeps; always try the final state
Thinning counted only the bodies, so a model of many copies of one part
(records, not bodies, filling the room) shrank its budget to nothing and
saved no cache although its final state fitted. The choice now counts the
records of what it keeps, always keeps the final state, and says what it
counted, so each retry keeps less and the final state alone is always tried.
* docs: a thinned model cache counts its records and always tries the final state
* Model cache test: the slowest feature's state holds bodies 16 to 23
* Model cache: thinned opens with fingerprints that read the state before a feature
Since #3397 a feature's fingerprint reads where the bodies before it lie.
After a state the file left out, adopt takes the saved fingerprint and the
evaluation that works the state out checks it. A result evaluated ahead
is never taken as one the file left out. The thinned round trips include
the placed components' design, whose cut reads those placements.
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #5227
* Web: leave Ctrl+Tab and Ctrl+Shift+Tab to the browser (#5227)
eframe cancels the default action of every Tab keydown on its canvas
(egui moves focus with Tab), so browsers that let a page cancel their
shortcuts, such as Vivaldi, could not switch tabs while SolidCraft had
the keyboard. A capture listener on the window now stops Ctrl+Tab and
Ctrl+Shift+Tab before they reach the canvas, without cancelling them,
so the browser acts on them as usual. Plain Tab and Shift+Tab still
move between the app's fields. Desktop builds are unchanged.
web-api-test checks it in headless Chrome: Ctrl+Tab and Ctrl+Shift+Tab
keydowns are not cancelled, Tab still is.
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #3654
* Control channel: control.running lists what the app is running, across connections (#3654)
control.pending answers only for the connection that asks, so a second client,
the MCP bridge after a reconnect or a script saw {"pending": null} while the app
was busy. control.running answers for the whole app, the same on every
connection and at once (the connection's own thread answers it, also while it
is busy with a timed-out request): every request the app has been sent and not
answered yet, oldest first, with its connection, id, method, command id (or a
batch's step count), seconds since it arrived and whether it still waits its
turn. A timed-out request stays listed until the app answers it, also after its
connection closed. Parameters are never shown.
The window marks a request started when it takes it (ControlRequest.started),
which tells queued from running. The MCP server gets a read-only `running` tool.
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
A face pinched to a point, where its boundary touches itself or another of its loops within
TOLERANCE (a sphere touching a box's edge leaves one on the box), had a constraint refused by the
CDT, and the fallback cut across the face: it came out with no triangles or without those at the
pinch. When a constraint is refused, the vendored truck-meshalgo now puts every boundary point
within TOLERANCE of a segment into it (at most 8192) and triangulates the face again. Faces whose
segments all go in mesh as before.
The regression tests live in crates/kernel/src/thin_face_tests.rs, and the #1648 property test in
boolean_props.rs is no longer ignored. Robustness seeds 20 and 55 are recorded with their Linux
values.
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
* claim: #1801
* Insert SVG: the file picker opens on SVG files (#1801)
Every Open went through one host picker whose filters start with designs, CAD
files and meshes, so Insert SVG (and Insert DXF) in a sketch opened on STEP,
IGES and meshes, and the SVG file had to be found under another filter.
The host's pick_open now takes the extensions the command reads and offers
them as the picker's first, default filter. Insert SVG and Insert DXF pass
their type; every other Open passes none and keeps the list it had. The
browser host ignores them (it shows its own file panel).
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
Motion links can carry a coupling (gear pair, rack and pinion, screw) whose ratio follows radii, picked round faces or pitch. Design format 23.
Co-authored-by: SychPL <SychPL@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A motion link's ratio is read in b's units per a's (mm per degree, degrees per mm) when an angle and a distance are linked; stored values stay per radian, so saved designs are unchanged. Follow-up to #1415.
Co-authored-by: SychPL <SychPL@users.noreply.github.com>
List menus open below-right of the cursor and flip to end at the cursor where they don't fit (shared context_menu::place / show for viewport, browser, toolbar and Select Other). Right-clicking a plane in the viewport, a construction plane row or an origin plane row (new menu with Create Sketch and Show/Hide) offers Create Sketch on that plane. Fixes#1354.
joint.motion_link offset is read in the linked value's units (degrees or mm, or an expression); links are still stored in radians, so saved designs open unchanged. A new joint whose values fall outside its limits starts at the nearer limit, stores it, and says so in warnings. Loop dof also counts at a nearby pose, so a part at a dead point of its loop keeps its mobility. The mechanisms skill is updated to match.
The Joint and As-built Joint dialogs explain why picks make no joint and offer Create Components from Bodies when every pick is on one component; joint.create and joint.as_built suggest component.from_bodies. Labels in all catalogs. Fixes#1239.
Drive Joints opens on the joint picked in the browser, else the first joint that moves, and won't start on a rigid joint. ui.start takes params: joint.drive opens on a joint (id or name) at given values, driving nothing until OK; a command or param it doesn't take is refused.
A solidcraft-mechanisms agent skill with a worked example that a test replays through the MCP server (the actuator follows the hinge with the stated stroke and the drive stops at the rod's limit); the skills index and docs/mcp.md list it.
The bridged MCP server starts when no app is listening, says so in its instructions and in each tool error, and connects on the first call after the app starts. A batch that could not be sent says the app is not running. docs/mcp.md describes it.
snapshot starts from --in (a design or STEP file) with the script on top, as run does, and refuses options it doesn't take; an unknown command id names the closest registered ids.
A control-channel method that reads the orbit camera, sets any of target/yaw/pitch/roll/distance, and fits listed occurrences (or everything) keeping the direction. Its reply passed back sets the same view again; a refused field changes nothing.
joint.drive answers moved, and a reason (with unconnected) when the joint can move nothing; motion.play answers moved and unmoved; joint.list answers floating chains. joint.drive values come back in joint.list's shape (deg/mm). quiet: true trims joint answers and motion.play poses. A single value keeps a joint's other values. Drive Joints shows the floating case (drive.unconnected, all locales).
component.list {compact: true} gives counts of bodies, features, sketches and construction planes instead of their names, lists occurrences without identity transforms and children as occurrence ids; bodies: true keeps body names. The default form is unchanged.
doc.open and the STEP/mesh imports run on the file loader's worker with the request's path scope, and a font named by file path in a design is checked against the scope before it is read.
* fix: a modelled thread that would cut nothing is refused (#995)
An M5 modelled thread in a 5 mm hole cut nothing into the nut: the
internal groove runs from the minor diameter out to the major one, so
with the hole as wide as the thread it lay entirely in the hole, and the
faceted cut only left the wall faceted with slivers ("the thread goes the
wrong way"). The fit check takes a 5 mm hole as an M5 hole.
- thread_groove measures how deep the groove cuts into the material
(hole wall out to the major radius, or shaft surface in to the root);
under a quarter of the basic depth 5H/8 the modelled thread is refused
with the measured depth, the tap drill (as solid.hole drills it) or
shaft size to use, and the size that fits the face, e.g. "this 5.00 mm
hole is too wide for a modelled M5 thread: it would cut 0.00 mm of the
thread's 0.43 mm depth; drill 4.2 mm for M5, or thread this hole M6".
- Cosmetic threads and their fit check are unchanged.
* test: pin the cut-nothing limit for coarse and fine pitch; show the hole size with show_mm
---------
Co-authored-by: MateuszSychXpl <151755914+MateuszSychXpl@users.noreply.github.com>
Co-authored-by: bherbruck <brennen.herbruck@gmail.com>
Fixes#928. Extrude To / From Object and Revolve To Object now resolve their target in the feature's own component frame, so a face in another component whose occurrence is moved is found where it is shown instead of the extrude stopping at a flat plane through the picked point. The point is named after the nearest face, a named point keeps its named face when another component touches it there, and such features re-evaluate when a placement changes, so they follow the target when it moves later.
Co-authored-by: SychPL <SychPL@users.noreply.github.com>
PathScope stores its roots canonicalised (\\?\ on Windows, /private/var on macOS), so the default-folders test now canonicalises scratch_dir() before looking it up, like the recovery folder on the line above.
The same fix came in from @Karl-VonRichter in #969, found on macOS.
Co-authored-by: Karl-VonRichter <20079846+Karl-VonRichter@users.noreply.github.com>
A design with components and joints exports as a URDF robot description (file.export to .urdf, the MCP export tool, solidcraft-cli run --out, File > Export URDF...). It writes one binary STL per component in metres, links with mass, centre of mass and inertia from material density, and maps rigid, revolute, slider, cylindrical, pin-slot, planar and ball joints to URDF joints or short chains. Loops, broken joints and loose parts are reported as warnings. Options: mesh_prefix, effort, velocity. Docs in docs/urdf.md.
The app's control protocol gets engine.batch: the steps run in one request (so one frame), each as engine.execute runs it, stopping at the first failure and rolling back to the checkpoint by default. The MCP batch tool sends it to a running app (with the control token), is never resent after a lost reply, and falls back to step-by-step when the app refuses (agent cursor shown, or a step loads a file). A batch may take up to an hour to answer.
STL (binary and ASCII) now welds each body's export mesh with the same tolerance as 3MF, so filleted and drilled bodies are closed. The weld grid's neighbour lookup no longer overflows for bodies far from the origin.
Fixes#784
The MCP server instructions now point to `body_topology` for finding edge points (there is no `list_edges` tool), with a test that every tool the instructions name exists.