Commit Graph
110 Commits
Author SHA1 Message Date
dotCooCoo f5bfe0c425 Effects: effect.apply refuses a wrongly typed parameter that has no default
effect.apply, appearance.addEffect and effect.setParams checked an effect
parameter's type only when the effect has a dialog default for the key. Keys
that only the params text names were stored as given: an Outline Stroke with
`width: "10 pt"` drew at the stroke's own weight, and a Gaussian Blur with
`linear: 1` blurred in linear light.

- The earlier 3D `perspective`, `lightAzimuth` and `lightElevation`,
  Scribble's `variation` and Outline Stroke's `width` must be numbers, and
  Gaussian Blur's `linear` must be true or false. A test walks every effect
  so a new key without a default needs a listed type or a check of its own.
- The `style` of Crop Marks must be `roman` or `japanese`, and `null` is
  refused. Any other value drew as Roman.
- An installed plug-in effect's parameters are checked against its manifest
  when the command runs, as plugin.run already checks the effect records a
  filter returns.
- effect.setParams with `warpStyle` checks the parameters against the new
  Warp style's effect, as it already did for a 3D effect's new
  `threeDType`. It checked them against the old style.
- effect.setParams refuses a `visible` that is not true or false, such as
  `1` or `"false"`, when called and when a step plays, and nothing changes.
  Before, the value was ignored and the rest of the call applied. `null`
  still leaves the visibility as it is.

A step of command.batch is refused the same way. A replayed journal or
action still plays a wrongly typed effect parameter, as it plays a key the
effect no longer takes.

Docs: docs/mcp.md, docs/target-app-parity.md.
2026-11-12 20:27:09 -08:00
dotCooCoo 3570efa0a9 Effects: a fill's 3D Revolve tells the plug-in effect before it which fill it sits on
The per-item Revolve loop passed a copy of each fill or stroke to the
geometry context, which finds the item by identity in the object's
appearance, so a plug-in effect before the Revolve was told it sat on
the object. Each copy now keeps its original item beside it.
2026-11-12 15:48:59 -08:00
BFlat c2f97d2efa Merge pull request #6961 from dotCooCoo/fix/gradient-repeat-runs
Gradients: a repeated gradient with many stops keeps every run
2026-11-13 05:22:59 +07:00
BFlat 10e87ff993 Merge pull request #6873 from dotCooCoo/fix/transform-effect-scales-raster-effects
Effects: a Transform effect transforms the raster effects listed before it
2026-11-13 04:25:45 +07:00
dotCooCoo c513114674 Gradients: Expand reports a capped gradient as unavailable
- object.expand.info reports no gradient or fill to expand for a
  gradient that repeats in more runs across its box than are written,
  matching object.expand, which leaves it a gradient. For a path that a
  painting live effect draws, it measures each piece the effect returns.
- CSS writes a repeating gradient's stop positions, and a repeating
  radial gradient's radius, with enough decimals that their rounding
  adds up to at most a hundredth of a period across the box (at least
  four for positions and three for the radius).
- docs/mcp.md says that sharp steps and swatch-linked stops are the last
  to be left out rather than always kept, and that the CSS conic limit
  counts stops before midpoints and Perceptual curves add more.
2026-11-12 12:55:54 -08:00
dotCooCoo 6da0e0f115 Tests: the shadow-on-each-fill test installs its own plug-in id
a_shadow_on_each_fill_a_painted_effect_returns_follows_the_transform_after_it
installed test.engine.twofills, which
a_second_painted_effect_on_a_fill_keeps_the_piece_s_other_fills also
installs with a different module. Installed plug-ins are kept for the
whole test process, so running the two tests in parallel could run one
test's effect with the other's module. The test now installs
test.engine.shadowedfills.
2026-11-12 11:29:37 -08:00
dotCooCoo 85b3740cca Effects: a fill's or stroke's own Transform effect leaves the effects listed after it unscaled
Effects apply in the order they are listed, so an effect listed after a
Transform effect applies to the art the Transform effect made, at its
own values, as an object's later effects do after the object's
Transform effect. With Scale Strokes & Effects on, a fill's or stroke's
own Transform effect scaled the effects listed after it in that fill's
or stroke's list by the mean scale: a 300% Transform turned a 5 pt
Offset Path after it into a 15 pt one, and a Drop Shadow after it took
three times its offset. With Transform Patterns or Scale Strokes &
Effects on, an SVG filter after it also had the origin of its user space
moved.

- The effects after such a Transform effect now keep their values, on
  the transformed art and on each copy, on every fill and stroke a
  live-effect plug-in paints the piece with, and when the copies are
  counted against the effect budget.
- With Transform Objects off, the effects before it keep their values
  too.
- The Transform params text, docs/target-app-parity.md and docs/gaps.md
  describe the rule, and drop the note that listed an object's later
  effects keeping their distances as a shortfall.
2026-11-12 10:59:21 -08:00
dotCooCoo 70a4c44033 Effects: every fill a painted effect returns follows the Transform effect
A painted live effect on a fill can return a piece with several fills,
each a copy of that fill with its effects. A Transform effect after it
moved the piece, but only the first returned fill's raster effects
followed it: with a Drop Shadow before the painted effect and a 200%
Transform effect after it, the first fill's shadow offset became 20 pt
and the second fill's stayed 10 pt. Each returned fill and stroke now
takes the followed values.
2026-11-12 10:08:53 -08:00
BFlat cc6509827d Merge pull request #6878 from dotCooCoo/fix/opacity-mask-follows-reshape
Opacity masks: linked masks follow group effects, the Free Transform distort and envelopes
2026-11-12 23:10:24 +07:00
dotCooCoo 94e9839438 Opacity masks: linked masks follow group effects through type, plug-ins and perspective
Type and live objects in the art of a linked opacity mask are outlined
and evaluated before a group's or layer's geometry effects run over it,
so the mask moves and bends with the art it masks. Before, the effects
reached only the paths in a mask's art.

- With Scale Strokes & Effects on, a Transform effect scales the strokes
  of an image's or symbol instance's linked mask as it scales those of a
  path's mask.
- A mask's art takes Transform copies in the form it had before the
  first effect, as the art it masks does.
- A linked mask whose art is one image or symbol instance takes a
  Transform effect's move and copies with the art it masks. Under any
  other geometry effect, which images and symbol instances do not take,
  art whose masks hold one keeps its shape.
- A symbol instance in perspective projects the linked masks in its art
  with that art.
- A live-effect plug-in that paints its pieces leaves art with linked
  masks as it is, since a mask can't take the pieces.

Docs: the docs/target-app-parity.md line moves next to the Transparency
entry and lists the cases where existing documents draw differently.
2026-11-12 06:35:37 -08:00
dotCooCoo 2212739c54 MCP: plug-in, recovery and script report replies leave out paths outside the read roots
A client confined by --automation-read-root saw paths outside its read
roots that the person or an earlier request had put in the session:

- plugin.list: the plug-in folder last loaded and its failed files;
  plugin.list and plugin.info: a plug-in's source file; plugin.list,
  plugin.safeMode and plugin.setEnabled: the file of a plug-in turned
  off for the next launch.
- file.recovery.list and file.recovery.restore: a document's original
  file; file.recovery.list and file.recovery.save: the recovery folder.
- ui.inspect's scriptReport: the files a run wrote and the included
  file a failed run stopped in.

A folder outside the read roots is now null, and a file outside them is
named by its file name. plugin.setEnabled {file} takes that name. The
restored documents keep their files, and unconfined clients and the
app's own UI see the same replies as before.
2026-11-12 06:03:01 -08:00
BFlat 0ad57dc872 Merge pull request #6811 from dotCooCoo/fix/effect-refresh-deferred-nested
MCP: effect.refreshDeferred takes objects in symbol, pattern, brush and mask art
2026-11-12 15:13:13 +07:00
BFlat 053dc7ab8c Merge pull request #6746 from dotCooCoo/fix/plugin-ignored-grants-shared
Plug-ins: a grant one session stops counting stops counting for every session in the process on its folder
2026-11-12 14:15:46 +07:00
dotCooCoo f494578072 MCP: effect.refreshDeferred takes objects in symbol, pattern, brush and mask art
effect.deferredStatus lists a stage in the art of a symbol, pattern, brush
or opacity mask by the id of the object in that art. effect.refreshDeferred
took such an id only while the object still had a deferred stage; an
existing object there without one was refused as no such object, while a
layer object without a stage was accepted. An id is now valid when it names
an object in the layers or in that art, an object without a stage adds
nothing to the counts, and an id of no object is an error.

The params text and docs/mcp.md said `id` names one object, but the params
text did not list the key, so `{id: N}` was refused as an unknown
parameter. It is now `{ids?|id?}`, and the command checks only the form of
the ids before it looks them up itself.

The requested ids are held in a set while the art is walked, and only those
with a stage are handed to the refresh.
2026-11-11 22:45:06 -08:00
BFlat 41e4126089 Merge pull request #6679 from dotCooCoo/fix/plugin-grants-delete-test-portable
Plug-ins: the failed delete of grants.json is tested on every platform
2026-11-12 13:18:04 +07:00
dotCooCoo 1eb02f2f7a Merge main into fix/plugin-ignored-grants-shared 2026-11-11 22:03:31 -08:00
BFlat 485d4e3fff Merge pull request #6747 from dotCooCoo/fix/plugin-grant-sealed-api
Plug-ins: only an approval records a grant, and installs need one
2026-11-12 12:41:15 +07:00
dotCooCoo 2846fb3f43 Merge main into fix/plugin-grants-delete-test-portable 2026-11-11 21:01:50 -08:00
BFlat d843a4f1ec Merge pull request #6670 from dotCooCoo/fix/plugin-put-back-result
Plug-ins: an update left in the store by a failed put-back turns a disabled plug-in on
2026-11-12 11:42:01 +07:00
dotCooCoo a447b78ac4 Merge main into fix/plugin-grants-delete-test-portable 2026-11-11 19:59:24 -08:00
dotCooCoo c6d9712b9e Merge main into fix/plugin-grants-delete-test-portable 2026-11-11 19:03:22 -08:00
dotCooCoo be4fd1c23e Merge main into fix/plugin-put-back-result 2026-11-11 18:55:34 -08:00
dotCooCoo 822efb5db9 Merge main into fix/plugin-grant-sealed-api
# Conflicts:
#	docs/target-app-parity.md
2026-11-11 18:49:27 -08:00
dotCooCoo fabc28370f Merge main into fix/plugin-call-reprepare-test 2026-11-11 18:29:37 -08:00
BFlat 4c64f6206b Merge pull request #6712 from dotCooCoo/fix/plugin-store-test-isolated
Tests: the empty-params command sweeps leave plug-in caches alone
2026-11-12 09:10:52 +07:00
dotCooCoo cde367b889 Plug-ins: only an approval records a grant, and installs need one
registry::grant takes a GrantToken, which the engine makes after a
person's approval and when it loads grants.json. registry::install,
install_bytes and folder loads refuse a module that lists permissions no
recorded grant gives this very module (Error::NeedsApproval), and leave
the plug-in under its id and the revision as they were. A folder load's
approve callback only narrows what installs. A module without
permissions, or with an empty list, installs as before.

A folder load in a session with a plug-in store records the store's
grant for each module it approves, so those modules still load.

docs/plugins.md says what the approval does not protect against: native
code linked into the app, and another program running as the same user,
which can edit the Plug-ins folder and its unsigned grants.json.
2026-11-11 17:35:06 -08:00
dotCooCoo 45987bc948 Plug-ins: a grant one session stops counting stops counting for every session on its folder
The grants that should be gone from grants.json but could not be removed
were kept per Session, so a second Session attached to the same Plug-ins
folder in the same process still counted them for plugin.inspect,
plugin.install and loading plug-ins. They are now kept for the process,
by the folder's canonical path and the plug-in id, and a successful
write of grants.json for the id in any session makes the grant count
again. Attaching another folder no longer clears them.

A failed removal does not override a write another session made for
the id in the meantime. The process keeps the latest 256 grant writes
for that check. Another process on the folder still reads grants.json
as it is written.
2026-11-11 17:25:19 -08:00
dotCooCoo ef7fbaac31 Plug-ins: a failed rename back of a removed plug-in's settings is reported and undone later
When plugin.remove could not remove the grant and then could not rename
the plug-in's settings folder back from <folder>.removed, the failure was
only logged: the plug-in got empty settings, and the next removal deleted
the .removed folder with the settings in it.

- The error now names the .removed folder, the plug-in folder and the
  reason, and the plug-in gets no storage until the settings are back.
- The next load, install or plugin.remove of the plug-in renames a kept
  folder back while the plug-in has no folder of its own. When it has one,
  that folder is used and the kept one stays until plugin.remove deletes
  both. A removal while a kept .removed folder exists renames the plug-in
  folder to <folder>.removing.
- Before deleting a renamed folder, the store writes an empty
  <folder>.removed.delete file and removes it once the folder is gone, so
  what a failed deletion leaves is not moved back. When that file can't
  be written, nothing is deleted.
- A .removed folder without a .delete file, including one an earlier
  version left, counts as kept settings: a new install of the id needs
  replace: true, and plugin.remove of the id deletes it.
2026-11-11 15:51:11 -08:00
dotCooCoo afe2a71e59 Tests: the empty-params command sweeps leave plug-in caches alone
every_command_survives_empty_params and
read_only_commands_read_the_instance_of_a_picked_object run every command
with {}. That includes plugin.clearCache, which without an id empties the
cache of every plug-in installed in the process. When either sweep ran
during plugin_storage_is_capped_and_survives_a_damaged_settings_file
(called from plugins_keep_settings_and_a_cache_of_their_own_in_the_store),
it deleted that test's cache entries Tile and tile. The test's
st.clear_cache() == MAX_CACHE_BYTES + 3 assertion after the
MAX_CACHE_SCAN folders then failed with 536870912 bytes freed.

Both sweeps now leave plugin.clearCache out.
tests/plugin_clear_cache.rs already runs it with {} in a test binary of
its own. A new test runs the other plug-in commands with {} in the
selection states that every_command_survives_empty_params uses, and checks
that an installed plug-in's cache keeps its 3 bytes.
2026-11-11 15:26:46 -08:00
dotCooCoo f3cff033ca Edit: a command that ends a drag keeps its unapplied preview
With Real-time Drawing off, a tool stores its latest preview and applies
it only when the drag is released. A command that ends the drag first
(Undo, Redo, a filter that calls commands, a plug-in command that calls
commands, and the other callers of settle_interaction) committed the
interaction without that preview, so the drag was dropped: Undo then
took back the step before it, and a calling filter was given the
objects where they were before the drag.

settle_interaction now applies such a preview first, as the tool's own
commit does, and cancels the drag when the preview fails. The re-prepare
of a calling filter's input after settling is now reached, and tests
cover it at both plugin.run and plugin.<plug-in>.<command>.
2026-11-11 14:15:22 -08:00
dotCooCoo 08d4a5b486 Plug-ins: a put-back that fails after writing the old module keeps a disabled plug-in off
When plugin.install can't write or remove a grant and the write that puts
the old module back reports a failure, install now reads the stored module
again and turns the plug-in on only when the store holds the new module. A
put-back that wrote the old bytes and then failed (for example at its last
flush) leaves a disabled plug-in disabled. This applies to modules that
list permissions and to modules that don't.
2026-11-11 14:03:36 -08:00
dotCooCoo fd7ff70941 Plug-ins: an update left in the store by a failed put-back turns a disabled plug-in on
When plugin.install writes a module that lists permissions, its grant
can't be written, and the old module can't be put back (or, on a first
install, the new module can't be removed), the store holds the new
module. install now takes put_back_plugin's result, so the plug-in's
disabled entries (by id and by stored file) are cleared as for any
install the store holds. The old grant, or none, still doesn't cover the
new module, so the next start skips it until it is approved.

Docs: docs/plugins.md, docs/mcp.md, docs/target-app-parity.md.
2026-11-11 13:03:38 -08:00
dotCooCoo 42ba91fb6d Plug-ins: the failed delete of grants.json is tested on every platform
A test-only switch makes removing grants.json fail on the test's thread while the file still reads. The new test replaces an approved plug-in with one that asks for no permissions while the switch is on, and checks that the store gets the approved module back and keeps the grant unchanged. The Windows test that holds the file open without delete sharing stays.
2026-11-11 12:49:41 -08:00
dotCooCoo 949cbe7bea Tests: plug-in folder tests take turns on the folder last loaded
The folder the Additional Plug-ins Folder preference or plugin.reload last loaded is kept for the
whole process. The six engine tests that set the preference or run plugin.reload now hold one lock,
so another test loading its own folder between two steps cannot change what the next step loads or
lists.
2026-11-11 09:37:01 -08:00
dotCooCoo 8eb3788eba Plug-ins: setting the plug-ins folder again after clearing it installs its plug-ins
Clearing Preferences > Plug-ins & Scratch Disks > Additional Plug-ins Folder
left the folder recorded as the one last loaded. Setting the same folder
again then matched that record, and nothing in it was installed, so a
plug-in removed in the meantime stayed removed. Clearing the preference now
forgets the folder when it is the one last loaded.
2026-11-11 08:07:27 -08:00
dotCooCoo feda950bc8 Plug-ins: live effects of a document's first view share a 2,000 ms budget
While the app draws a document on screen, the live-effect plug-in runs of
its first view share a 2,000 ms budget (VectorCraft's own value); each run
takes the lesser of its 5,000 ms limit and what is left. A stage left out
for time, or one that times out, fails, has no plug-in or meets Safe Mode,
draws its input art unchanged and is deferred, and the status bar says how
many. Later drawings leave a deferred stage as it is until its input or
parameters change.

- effect.deferredStatus lists the deferred stages with object, effect
  index, fill or stroke, plug-in and reason, objects in symbol, pattern,
  brush and opacity-mask art included.
- Effect › Refresh Deferred Effects (effect.refreshDeferred {ids?}) runs
  them again with a fresh budget and returns {refreshed, stillDeferred,
  failed}.
- Each document has its own budget and deferred stages; exports, Expand
  Appearance and commands run every stage under the per-call limit.
- The web build has no clock for the budget: its first view runs no
  plug-in effect, and a refresh runs them.

Docs: docs/mcp.md, docs/plugins.md, docs/target-app-parity.md.
2026-11-10 22:16:01 -08:00
dotCooCoo 811c21e7c2 Plug-ins: calls need doc.read, and an edit in progress is kept before a calling run
A command's result describes the document, so a plug-in without doc.read
gets -6 for every call, whether or not the id names a command. A run that
calls commands commits an edit still in progress first and prepares its
input again when that changed the document or the selection.
2026-11-10 11:26:10 -08:00
dotCooCoo eebbe496d4 Merge main into fix/plugin-grant-app-approval 2026-11-08 23:53:57 -08:00
BFlat 0f2857e1a1 Merge pull request #6286 from dotCooCoo/fix/plugin-stale-grant
Plug-ins: replacing a granted plug-in with one that lists no permissions removes the grant
2026-11-09 15:41:04 +08:00
dotCooCoo 0efd9a9ccc Merge main into fix/plugin-grant-app-approval 2026-11-08 23:26:53 -08:00
dotCooCoo b71b88b6b0 Merge main into fix/plugin-read-only-input-held-once
# Conflicts:
#	docs/plugins.md
2026-11-08 22:10:42 -08:00
dotCooCoo 65ff9df849 Plug-ins: a grant left behind stops counting when its module can't go back
plugin.install and plugin.remove count an entry of grants.json that this
version can't use when it can't be removed. plugin.install stops counting
a grant it can't remove when the old module can't be written back to the
Plug-ins folder.
2026-11-08 22:00:26 -08:00
dotCooCoo e7c3307f07 Merge main into fix/plugin-grant-app-approval 2026-11-08 20:58:47 -08:00
dotCooCoo 0e1aa5cc2e Plug-ins: a filter's read-only objects stay in its input and failed decodes count
A filter run kept a second copy of every read-only object as JSON values.
The run now keeps where each object is in the input, and checks an output
entry against it by parsing that one object, unless the entry is `{"id": …}`
alone.

An image that reaches the decoder now counts toward the 16 MiB pixel total
even when decoding fails, so broken images use up the same total as images
that decode.

The plugin.run text, docs/mcp.md and the progress line state the paint
order and that rule, and docs/plugins.md states that a CMYK image's pixels
are sRGB. New tests cover the per-image 1 MiB, the 16 MiB pixel total and
the 16 MiB text total at and one past their values, memory-budget failures
before the plug-in starts, images that are not decoded, CMYK images
included, and leading per run.
2026-11-08 20:54:02 -08:00
dotCooCoo 9e655ae2e6 Merge main into fix/plugin-stale-grant 2026-11-08 20:38:46 -08:00
dotCooCoo 6cce39dd7f Plug-ins: a grant that can't be removed stops counting for the session
When loading the plug-in store can't remove the grant of an id whose stored
module lists no permissions, or can't read grants.json, the entry stays on
disk and still approved the bytes it covers. plugin.install has the same case
when the grant can't be removed and no earlier module can be put back.

The session now keeps those ids and does not count their grants in
plugin.inspect, plugin.install, plugin.reload or a later load of the store. A
successful update of the grant for the id ends that, and so does attaching a
different store.
2026-11-08 20:38:03 -08:00
BFlat 5402f17f15 Merge pull request #6276 from dotCooCoo/fix/plugin-last-error-origin
Plug-ins: lastError keeps a failed run's error, and hit tests follow swatch changes
2026-11-09 12:26:56 +08:00
dotCooCoo bff7f0ad54 Merge main into fix/plugin-grant-app-approval 2026-11-08 20:16:56 -08:00
dotCooCoo 623d8fdd83 Plug-ins: the settings file test passes remove_plugin_storage its callback 2026-11-08 19:37:01 -08:00
dotCooCoo 5d715a5051 Plug-ins: an effect that paints nothing clears a refused reference's error
Only a painted effect's references are checked, so a refusal recorded for a painted effect stayed in plugin.info after an effect that paints nothing replaced it under the same id.
2026-11-08 19:35:50 -08:00