* ✨ Add additional logging and validation for image upload
* 🎉 Add chunked upload support for font variants
Extend the font variant upload flow across frontend, backend, and common
to support the standardized chunked upload protocol.
**Backend:**
- Add \`:font-max-file-size\` config default (30 MiB) and schema entry
- Add \`validate-font-size!\` in \`media.clj\` (mirrors
\`validate-media-size!\`, raises \`:font-max-file-size-reached\`)
- Extend \`schema:create-font-variant\` to accept either \`:data\`
(legacy bytes or chunk-vector) or \`:uploads\` (new chunked session
map), with a validator requiring exactly one
- Add \`prepare-font-data-from-uploads\`: assembles each chunked
session via \`cmedia/assemble-chunks\`, validates type+size
- Add \`prepare-font-data-from-legacy\`: normalises legacy byte/chunk
entries, writing to a tempfile (joining via SequenceInputStream),
validates type+size
- Add structured logging ("init"/"end") with \`:size\`, \`:mtypes\`,
and \`:elapsed\` in \`create-font-variant\`
**Frontend:**
- \`upload-blob-chunked\` accepts a per-caller \`:chunk-size\` option
- Add \`font-upload-chunk-size\` (10 MiB) and \`upload-font-variant\`
fn that uploads each mtype as a separate chunked session
- \`on-upload*\` in dashboard fonts now calls \`upload-font-variant\`
instead of issuing \`create-font-variant\` RPC directly
- \`process-upload\` stores raw ArrayBuffer instead of chunking
client-side
**Common:**
- Replace \`"font/opentype"\` with \`"font/woff2"\` in \`font-types\`
**Tests:**
- 25 tests / 224 assertions covering all three upload paths (direct
bytes, legacy chunk-vector, new chunked sessions), size validation,
and media type validation
Signed-off-by: Andrey Antukh <niwi@niwi.nz>
* 📎 Add a script for check the commit format locally
---------
Signed-off-by: Andrey Antukh <niwi@niwi.nz>
Serena provides useful tools for the agentic workflow for penpot.
The following additional extensions are added:
1. uv and Serena installation, including a suitable serena_config.yml, are added to the devenv docker image
2. Serena configuration options are set via env vars and flags in manage.sh
3. run-devenv can now take -e flags which it forwards to docker exec
GitHub #9315
Adds a new MCP tool (devenv-only) that imports .penpot files into the
running Penpot instance. The tool downloads the file from a given URL,
stages it in the frontend's static directory, and triggers the import
via the ClojureScript REPL using the frontend's web worker infrastructure.
The temporary file is cleaned up after the import completes or fails.
Registered alongside CljsReplTool, sharing the same NreplClient instance.
Github #9217
Co-authored-by: Claude <noreply@anthropic.com>
Documents how to detect, diagnose, and recover from frontend crashes
(the Internal Error page) when working through the Penpot Plugin API:
- Detect via (some? (:exception @app.main.store/state)) in the cljs REPL
- Read cause from the same map (:type, :code, :hint, :details, :uri, ...)
- Reload by listing/selecting the workspace tab in playwright and
re-navigating to its URL, then re-checking the exception is gone
Co-authored-by: Claude <noreply@anthropic.com>
Add `critical-info` memory as an entrypoint (bootstrap memory) for the LLM, which
points to critical tools and memories, allowing the LLM to dynamically build up
relevant context
Exclude files like CONTRIBUTING.md or README.md from being ignored by /*.md pattern,
as this can influence agent behaviour (configurations that disallow ignored files
from being edited)
New tool to evaluate ClojureScript expressions by connecting to the
nREPL service already provided in devenv.
Add dependency 'nrepl-client' and a corresponding client class
as well as types to support this.
Add a new environment variable for 'devenv mode', which enables
the new tool (PENPOT_MCP_DEVENV).
* 🐛 Revert blend-mode hover preview when dismissing dropdown
When the blend-mode dropdown was dismissed by clicking outside instead
of selecting an option, the canvas kept rendering the last hovered
blend mode even though the inspector and data state had reverted. The
visible state of the shape no longer matched its stored state. Reset
the canvas render back to the shape's saved blend mode on dropdown
close so the preview never outlives the dropdown.
Signed-off-by: jack-stormentswe <crazycoder131@gmail.com>
* Fix blend-mode dropdown and various bugs
Fix blend-mode dropdown behavior and revert WASM render on pointer leave. Also, address multiple bugs related to user interactions and data handling.
Signed-off-by: Alejandro Alonso <alejandro.alonso@kaleidos.net>
---------
Signed-off-by: jack-stormentswe <crazycoder131@gmail.com>
Signed-off-by: Jack Storment <88656337+jack-stormentswe@users.noreply.github.com>
Co-authored-by: Alejandro Alonso <alejandro.alonso@kaleidos.net>
* 📎 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?
