* 🐛 Add regression test: main-side component edits break copy swap slots
Reproduces the :missing-slot referential-integrity failure ("Shape has been
swapped, should have swap slot") that crashes files with component copies.
Root cause: reordering or deleting a nested sub-head IN THE MAIN of a component,
while copies exist, does not propagate swap slots to the copies. find-near-match
matches a copy's sub-heads to the main's children by POSITION, so once the main's
order changes the copies' shape-refs no longer match their position and, lacking a
swap slot, fail referential-integrity validation.
- Copy-side edits are handled correctly (characterization tests, pass today).
- The two main-side tests fail today with :missing-slot and go green once the
sync assigns swap slots to copies on a main reorder/delete.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* 🐛 Fix integrity crashes from copy/main child-order divergence
Copy sub-heads were matched to their main's children purely by position
(find-near-match), while several code paths reorder or mutilate one side
only. Any of them made file validation fail with :missing-slot ("Shape
has been swapped, should have swap slot"), crashing the workspace on the
next validated commit, or persisting a corrupt file whose later edits
crash. Reproduced live: deleting a sub-head inside a copy (which only
hides it) and then reflowing the copy's grid moved the hidden (cell-less)
child to the front of :shapes, knocking every sibling out of its
positional slot.
Four fixes, one per divergence path:
- validate/check-required-swap-slot: a sub-head whose shape-ref is still
a child of the near main parent is a REORDER (the component sync
realigns it), not a swap; a swap slot is required only when the ref
points outside the near main parent (a real swap). This matches how
the sync engine itself pairs children (by shape-ref, not by position).
comp-processors/fix-missing-swap-slots (migration 0019) is aligned:
adding slots to merely-reordered sub-heads would freeze them out of
normal synchronization.
- changes/:reorder-children now refuses to alter the child structure of
component copies unless allow-altering-copies is set, mirroring the
is-valid-move? rule of :mov-objects; that structure is owned by the
component sync engine. Grid reflows emitted this change type with no
guard. pcb/reorder-grid-children also skips copy grids producer-side.
- layout/reorder-grid-children keeps children that participate in no
cell (hidden or absolute positioned) at their original index instead
of lumping them at the front: moving them gratuitously changed their
z-order and, in copies, broke the positional matching. Note :shapes
stays reversed relative to the sorted cell order for in-cell children.
- logic/generate-delete-shapes: deleting shapes from inside a main
(without deleting the main root, whose copies keep working against the
deleted component) now also deletes the copy shapes that reference
them, transitively (copies of copies) and across all pages of the
file, so no dangling shape-refs remain. Skipped for
allow-altering-copies flows (component swap replaces the shape and the
sync reconciles copies via swap slots). Cross-page removals build
redo/undo changes against that page's objects directly, since the
changes-builder mounts only the current page; their undo mov-objects
carry allow-altering-copies so restoring inside a copy is not rejected
by the new guard.
The regression tests assert the fixed semantics: main-side reorders and
deletes keep copies valid, the previously crashing full chain (unvalidated
main reorder + later copy edit) stays healthy, :reorder-children cannot
scramble copies, and reorder-grid-children keeps cell-less children in
place. The namespace is now also registered in the JS test runner.
AI-assisted-by: Claude Opus 4.8 (1M context)
* ✨ Extend composable slot cases to the grid-reflow crash sweep
Case D (CopySubheadDeletePreservesSlots) now sweeps which copy sub-head
is deleted (first or last) and whether the copy root is resized
afterwards, forcing a grid reflow: the reflow used to move the hidden
(cell-less) child to the front of the copy's children, shifting every
sibling out of its positional slot. Deleting the FIRST sub-head masked
the bug (moving it to the front is a no-op), which is why the case
passed before this sweep. The foundation layout is grid accordingly.
Case E (MainReorderKeepsCopySlots) no longer hangs the app now that a
main-side reorder leaves a valid file, so its warning docstring is
replaced: it runs as a routine test (verified headless, passing) and is
safe in a "run all".
SlotIntegrity's doc is updated to the new validator semantics (a slot is
required only for real swaps, not reorders); the positional alignment it
asserts remains the correct, stronger steady-state invariant for these
cases. The lockfile change materializes the playwright devDependency
already declared in package.json.
AI-assisted-by: Claude Opus 4.8 (1M context)
* 📚 Record copy/main order-divergence invariants in memories
Swap-slot semantics (membership, not positional; slots only for real
swaps), the copy-structure guards on :mov-objects/:reorder-children, the
grid reorder stability for cell-less children, and the main-side delete
propagation in generate-delete-shapes.
AI-assisted-by: Claude Opus 4.8 (1M context)
* 🐛 Add adjustements to the code
---------
Co-authored-by: Michael Panchenko <michael.panchenko@oraios-ai.de>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Combined fixes from PR #10656 (numeric-input redesign) and PR #10696
(global finite? guard) for issue #10638.
- Add (number? v) guard to cljs mth/finite? so strings are rejected
- Redesign numeric-input last-value* to store number, not formatted string
- Invalid-input fallback restores display without emitting on-change
- Esc now fully discards typed text (resets raw-value* + dirty flag)
- Token dedup by name instead of resolved value
- Defense-in-depth: d/parse-double at 4 padding/gap handlers
- Math tests (common), unit tests, Storybook play tests, Playwright E2E
AI-assisted-by: deepseek-v4-flash
Co-authored-by: Akshit Nassa <nassaakshit@gmail.com>
Co-authored-by: Ulises Millán <ulises.millanguerrero@gmail.com>
Repair text shapes with empty/broken content at all three levels
(root, paragraph-set, paragraph) in migration 0025 to prevent
workspace update failures. Handle all corner cases: nil/empty/non-
vector children, non-map items, wrong types. Remove geometry-only
validation skip in changes.cljc so all shapes are validated.
AI-assisted-by: qwen3.7-plus
Guard remaining text pipeline locations that accessed :selrect directly
against nil/zero-dimension selrects by using safe-size-rect, which
provides a 4-level fallback chain (selrect -> points -> shape fields ->
empty 0.01x0.01 rect).
Fixes:
- fix-position in viewport_texts_html.cljs: replaced dm/get-prop
:selrect with ctm/safe-size-rect for both old and new shape
- assoc-position-data in modifiers.cljs: replaced (:selrect ...)
with ctm/safe-size-rect for delta computation
- change-orientation-modifiers in modifiers.cljc: replaced raw
:selrect access with safe-size-rect for scale and origin computation
Closes#10617
AI-assisted-by: mimo-v2.5-pro
Guard the content-to-PathData coercion on whether
stp/convert-to-path actually produced a new value, so
SVG-raw shapes (whose :content is a hiccup map) pass through
unchanged instead of crashing.
Closes#10612
AI-assisted-by: deepseek-v4-pro
Preserve real size overrides during variant switches without copying stale absolute composite geometry from the source variant.
Signed-off-by: Codex <codex@openai.com>
* Revert "🐛 Detect duplicated token names in the whole library (#9034)"
This reverts commit 61cd7573553b1c5e9fc2d7300cf9b2c36b4dcbb6.
* 🔧 Preserve some enhancements and fixes that are still valid
* 🔧 Fix broken integration tests
* 🐛 Fix DTCG token import discriminator and group-level $type inheritance
Closes#8342.
The DTCG Community Group Final Report (W3C, 2025-10-28) specifies:
"The presence of a $value property definitively identifies an object
as a token."
"A token's type can be specified by the optional $type property [...]
Furthermore, the $type property on a group applies to all tokens
nested within that group."
Two bugs in `common/src/app/common/types/tokens_lib.cljc` violate the
spec and silently break import of any third-party DTCG file that uses
group-level type inheritance:
1. `flatten-nested-tokens-json` used `(not (contains? v "$type"))` as
the group-vs-token discriminator. A group node carrying only a
`$type` (to set a default for child tokens) was misidentified as a
token, then immediately discarded because it had no `$value`.
2. `schema:dtcg-node` declared both `$type` and `$value` as required,
so even after the discriminator was fixed any leaf token that
relied on group-level type inheritance failed `dtcg-node?`
validation and never reached the parser.
The combined effect: importing a spec-compliant DTCG file that
expressed types at the group level produced a TokensLib with no
tokens at all, because every leaf was discarded as "unknown type".
Penpot-exported files were unaffected because Penpot always emits
both `$type` and `$value` on every token and never attaches `$type`
to a group, so the existing tests covered only the inline-type
shape.
- `schema:dtcg-node`: mark `$type` optional.
- `flatten-nested-tokens-json`: use `$value` as the discriminator
(anything without `$value` is a group), accept an optional
`inherited-type` accumulator that carries the nearest enclosing
group `$type` down through recursion, and resolve a token's type
from its own `$type` first, falling back to the inherited type.
A token's own `$type` always wins over the inherited one (per
spec).
Added `parse-dtcg-group-type-inheritance` covering both cases:
- group `$type` is inherited by tokens that don't declare their own
(`colors.red`, `colors.blue`, `space.small`)
- token `$type` overrides the inherited group `$type`
(`colors.danger`, `space.large`)
Existing DTCG round-trip tests continue to pass because they all
declare `$type` at the token level, which the new code still honours.
CHANGES.md entry added under the 2.17.0 Bugs-fixed section.
* 📚 Do not update CHANGES.md
We are changing the procedures to not update the changelog on each PR. Instead, we use github tracking to check what issues come in a release, and update the changelog automatically in a batch.
Signed-off-by: Andrés Moya <hirunatan@hammo.org>
---------
Signed-off-by: Andrés Moya <hirunatan@hammo.org>
Co-authored-by: MilosM348 <milos.milic001@outlook.com>
* ⚡ Improve performance and fix orphan detection in validate-file
- Add `*ref-shape-cache*` dynamic var to memoize `find-ref-shape`
lookups per page, avoiding repeated O(depth) ancestor walks.
- Add `*children-sets*` pre-computed maps for O(1) parent-child
containment checks, replacing linear `some` scans.
- Short-circuit `inside-component-main?` when the shape context
already implies a main component.
- Use single-pass reduce with early exit for duplicate detection
(children, swap slots) instead of count/distinct or frequencies.
- Guard `check-missing-slot` to skip expensive `find-near-match`
when the shape already has a swap slot.
- Refactor variant-set validation to use `run!` with direct `get`.
- Refactor `check-ref-cycles` to use a single `reduce-kv` pass.
- Fix `get-orphan-shapes`: the original `map` pipeline produced
nils so orphan shapes were never validated; rewrite with
`reduce-kv` for correct results.
- Add `validate-file-affected!` for change-scoped validation,
replacing full file validation in `process-changes-and-validate`
to only validate pages and components touched by the changes.
Signed-off-by: Andrey Antukh <niwi@niwi.nz>
* ✨ Improved validation
---------
Signed-off-by: Andrey Antukh <niwi@niwi.nz>
Co-authored-by: alonso.torres <alonso.torres@kaleidos.net>
Do not show the library sync popup when the only differences are global x/y changes on library components. We now generate the actual sync changes and only notify if there are real redo-changes to apply.
Run cll/generate-sync-file-changes for candidate libraries and filter out those with empty :redo-changes. The expensive check is deferred via rx/timer 0 so it runs asynchronously and does not block the UI.
Why: Position-only changes are normalized during sync (via reposition-shape) and never propagate to copies; showing the popup in that case was a false positive.
Performance: The check is deferred to the next tick to avoid UI stutter on large files with many libraries.
Add focused common JavaScript test execution with log-level control and a quiet test wrapper matching the frontend workflow.
Update developer docs and testing memories to reflect the common/frontend test split, and document why runner helper extraction is deferred.
Signed-off-by: Codex <codex@openai.com>
Add a shared `schema:font-family` whitelist validator in
app.common.types.font that only allows letters, digits, spaces,
hyphens, underscores, and dots in font family names. Apply the schema
to create-font-variant and update-font RPC endpoints on the
backend, and add client-side validation in the dashboard fonts UI.
Include unit tests for the schema and integration tests for the RPC
handlers.
Signed-off-by: Andrey Antukh <niwi@niwi.nz>
* 📎 Add test that surfaces the bug described in #14070
The bug: alt-drag-duplicating a variant master into the variant container
auto-creates a new variant component cloned from the source
via duplicate-component, which preserves :touched on every cloned shape.
The resulting copy's children inherit those :geometry-group touched flags
through add-touched-from-ref-chain on switch.
variants-switch -> component-swap (keep-touched? true) -> generate-keep-touched
then runs update-attrs-on-switch for each touched child. The new shape's
:y is correctly skipped by the per-attr "different masters" guard, but
:selrect/:points fall through to a width/height-based safety check that
uses exact equality. In practice, the alt-drag modifier path leaves a
sub-pixel drift in :width on the copy, so equal-geometry? returns false,
the safety check is bypassed, and the :else branch copies the source
variant's :selrect verbatim onto the freshly instantiated target shape.
The shape ends up with :y from the target master but :selrect.y from the
source — the renderer reads :selrect, so the child appears at the source
position inside a parent that has resized to the target's dimensions:
the visible "button cut off" symptom.
The new test sets up a variant container whose children are themselves
component copies (matching production: variant masters' children carry
:touched), introduces the same kind of width drift on the copy that the
alt-drag path produces, and runs the swap directly via the existing
test harness. It asserts (a) :y matches the target, (b) :selrect.y
matches the target, and (c) :y and :selrect.y agree. With the current
code (a) passes and (b)/(c) fail — capturing both the wrong value and
the internal inconsistency that causes the visible regression.
* 🐛Fix#14070 by no longer comparing floats for exact equality in equal-geometry?
