4fc95131bf* feat(pdf-server): get_viewer_state interact action
New interact action that returns a JSON snapshot of the live viewer:
{currentPage, pageCount, zoom, displayMode, selectedAnnotationIds,
selection: {text, contextBefore, contextAfter, boundingRect} | null}.
The viewer already pushes selection passively via setModelContext as
<pdf-selection> tags, but not all hosts surface model-context. This gives
the model an explicit pull.
selection.boundingRect is a single bbox in PDF points (top-left origin,
y-down) so it can be fed straight back into add_annotations. selection is
null when nothing is selected or the selection is outside the text-layer.
Wiring: new PdfCommand variant -> processCommands case ->
handleGetViewerState -> submit_viewer_state (new app-only tool, mirrors
submit_save_data) -> waitForViewerState -> text content block.
Also fills a gap in the display_pdf description: it listed interact
actions but was missing save_as; added that and get_viewer_state.
e2e: two tests covering selection:null and a programmatic text-layer
selection.
* test(pdf-server): fix get_viewer_state e2e race + assertions
readLastToolResult clicked .last() before the interact result panel
existed (callInteract doesn't block), so it expanded the display_pdf
panel instead. Wait for the expected panel count first.
Also: basic-host renders the full CallToolResult JSON, with the state
double-escaped inside content[0].text. Parse instead of regex-matching.
playwright.config.ts: honor PW_CHANNEL env to use system Chrome locally
when the bundled chromium_headless_shell is broken.
70 lines
2.2 KiB
TypeScript
70 lines
2.2 KiB
TypeScript
/**
|
|
* PdfCommand — the wire protocol between server and viewer.
|
|
*
|
|
* The server enqueues these via the `interact` tool; the viewer polls
|
|
* `poll_pdf_commands` and receives them as `structuredContent.commands`.
|
|
*
|
|
* This file is the single source of truth for the command shape. Both
|
|
* `server.ts` (which enqueues) and `mcp-app.ts` (which consumes) import
|
|
* from here — a new command variant must be added exactly once.
|
|
*
|
|
* Uses `import type` so neither side pulls pdf-lib into its bundle.
|
|
*/
|
|
|
|
import type { PdfAnnotationDef } from "./pdf-annotations.js";
|
|
|
|
/** Single form field assignment, as sent by `fill_form`. */
|
|
export interface FormFieldFill {
|
|
name: string;
|
|
value: string | boolean;
|
|
}
|
|
|
|
/**
|
|
* Partial annotation update: `id` + `type` pin the target, everything
|
|
* else is optional. Server validates shape; viewer merges into existing.
|
|
*/
|
|
export type PdfAnnotationPatch = Partial<PdfAnnotationDef> & {
|
|
id: string;
|
|
type: PdfAnnotationDef["type"];
|
|
};
|
|
|
|
/** Page range for text/screenshot extraction. Omitted bound = open-ended. */
|
|
export interface PageInterval {
|
|
start?: number;
|
|
end?: number;
|
|
}
|
|
|
|
/**
|
|
* Commands the server can send to the viewer via the poll queue.
|
|
* Adding a variant here means adding a `case` in both the server's
|
|
* `interact` handler (to enqueue) and the viewer's `processCommands`
|
|
* (to execute).
|
|
*/
|
|
export type PdfCommand =
|
|
| { type: "navigate"; page: number }
|
|
| { type: "search"; query: string }
|
|
| { type: "find"; query: string }
|
|
| { type: "search_navigate"; matchIndex: number }
|
|
| { type: "zoom"; scale: number }
|
|
| { type: "add_annotations"; annotations: PdfAnnotationDef[] }
|
|
| { type: "update_annotations"; annotations: PdfAnnotationPatch[] }
|
|
| { type: "remove_annotations"; ids: string[] }
|
|
| {
|
|
type: "highlight_text";
|
|
id: string;
|
|
query: string;
|
|
page?: number;
|
|
color?: string;
|
|
content?: string;
|
|
}
|
|
| { type: "fill_form"; fields: FormFieldFill[] }
|
|
| {
|
|
type: "get_pages";
|
|
requestId: string;
|
|
intervals: PageInterval[];
|
|
getText: boolean;
|
|
getScreenshots: boolean;
|
|
}
|
|
| { type: "save_as"; requestId: string }
|
|
| { type: "get_viewer_state"; requestId: string }
|
|
| { type: "file_changed"; mtimeMs: number };
|