64 lines
2.8 KiB
Markdown
64 lines
2.8 KiB
Markdown
# @edgeever/plugin-api
|
|
|
|
Public TypeScript contracts and runtime manifest helpers for EdgeEver client plugins and code-free themes.
|
|
|
|
```ts
|
|
import { definePlugin } from "@edgeever/plugin-api";
|
|
|
|
export default definePlugin({
|
|
activate(context) {
|
|
return context.commands.register({
|
|
id: "hello",
|
|
title: "Hello",
|
|
run: () => context.ui.showNotice("Hello from EdgeEver"),
|
|
});
|
|
},
|
|
});
|
|
```
|
|
|
|
The package ships ESM JavaScript and TypeScript declarations. Plugin bundles must produce a single `main.js` file without relative imports before distribution. See the EdgeEver plugin development guide for the Manifest format, permissions, settings Schema, release assets, and marketplace verification rules.
|
|
|
|
Desktop plugins can declare the `schedules` permission and idempotently schedule one of their own registered commands:
|
|
|
|
```ts
|
|
context.commands.register({ id: "refresh", title: "Refresh", run: refresh });
|
|
await context.schedules.upsert({
|
|
key: "hourly-refresh",
|
|
name: "Hourly refresh",
|
|
commandId: "refresh",
|
|
cronExpression: "0 * * * *",
|
|
});
|
|
```
|
|
|
|
Conflict-safe Markdown edits use the note baseline returned by `context.notes.get()`:
|
|
|
|
```ts
|
|
const note = await context.notes.get(noteId);
|
|
await context.notes.editMarkdown(noteId, {
|
|
expectedRevision: note.revision,
|
|
expectedContentHash: note.contentHash,
|
|
edits: [{ from: 0, to: 0, insert: "- [ ] New task\n" }],
|
|
});
|
|
await context.ui.openNote(noteId, { search: "New task" }); // requires ui:navigation
|
|
```
|
|
|
|
Bulk-indexing plugins can page through `context.notes.queryContent()`. The API also exposes workspace templates, full live-editor reads and range edits, template mutation events, and programmatic opening of a plugin's own registered panels.
|
|
|
|
Speech plugins can transcribe a note attachment with `context.ai.transcribeResource(noteId, resourceId)`, or pass an audio/video `Blob` or `File` to `context.ai.transcribeMedia(media)`. EdgeEver extracts audio on the client and calls the user's configured speech provider directly. Neither method accepts a media URL or returns the provider credential.
|
|
|
|
Dashboard and workflow panels can ask EdgeEver to render system chrome (header actions, search, tabs, selects, empty states) through `mount` context `shell.set()`, while the plugin keeps owning the body DOM.
|
|
|
|
Canvas-style plugins can read and conflict-safely replace resource bytes, open full-screen guarded panels with JSON state, and register constrained block embed renderers:
|
|
|
|
```ts
|
|
const [resource] = await context.resources.list(noteId);
|
|
const scene = await context.resources.read(resource.id);
|
|
await context.resources.update(resource.id, {
|
|
file: new File([scene], "drawing.excalidraw"),
|
|
expectedContentHash: resource.contentHash!,
|
|
});
|
|
|
|
context.editor.embeds.register({ type: "drawing", mount(container, embed) {} });
|
|
await context.editor.insertEmbed({ type: "drawing", resourceId: resource.id, title: "Drawing" });
|
|
```
|