Files
next-ai-draw-io/packages/mcp-server/src/diagram-operations.ts
Siddhant Shekhar 5c884766a8 feat(mcp): add multi-page (mxfile) support to MCP server (#862)
* feat(mcp): add multi-page (mxfile) support

The MCP server's write path could only address a single drawio page even
though the underlying .drawio file format and the embedded editor both
natively support multi-page documents. A user asking for "a second page
with a CNN diagram" would hit the validator with the error
"Expected closing tag </root> but found </mxCell>" because the validator
assumed input was a bare <mxGraphModel> and could not walk past the
<mxfile><diagram>...</diagram></mxfile> wrapper.

This patch closes the gap end to end:

* New helper module `pages.ts` centralises page CRUD (normalize, parse,
  list, find, add, rename, delete) so every layer agrees that the
  canonical in-memory shape is always <mxfile>. normalizeToMxfile and
  addPageToDoc both strip any leading <?xml ?> declaration before
  embedding a fragment inside <diagram> (the declaration is only valid
  at document start). addPageToDoc explicitly rejects full <mxfile>
  inputs so a caller cannot accidentally nest a document inside a page.
* `xml-validation.ts` now detects an <mxfile> root and scopes the
  duplicate-id check per <diagram>. The legacy regex check would
  otherwise reject every multi-page doc, because cells "0" and "1"
  repeat in each page's <root> by design. The DOM-parse path is gated
  by a cheap regex pre-check so legacy bare <mxGraphModel> callers
  don't pay any extra cost. The autoFix duplicate-id rename step is
  also guarded against mxfile inputs — renaming those sentinel cells
  would silently break drawio's parent references.
* `diagram-operations.ts` accepts an optional PageSelector. For
  <mxfile> input it resolves the page first and scopes all
  querySelectorAll calls to that page's <root>, so a delete on page 2's
  cell "2" no longer touches page 1's cell "2".
* `create_new_diagram` accepts either a bare <mxGraphModel> (legacy,
  auto-wrapped into a single-page mxfile) or a full <mxfile> with N
  diagrams. All existing single-page callers keep working unchanged.
* `edit_diagram`, `get_diagram`, and `export_diagram` gain optional
  `page_id` / `page_name` / `page_index` parameters. When omitted they
  target the first page — the "active by convention" default. Tool
  handlers with all-optional input schemas coalesce missing arguments
  via `input ?? {}` so a no-args MCP invocation can't crash on
  destructure before reaching the session-existence check.
* New tools: `list_pages`, `add_page`, `rename_page`, `delete_page`.
* Page-targeted PNG/SVG export uses a "load + export + restore" dance:
  the server projects the target page into a single-page <mxfile>,
  pushes it into the transient state so the browser reloads the iframe
  with just that page, waits for drawio to render (~3s), triggers the
  export, captures the data, and then restores the original multi-page
  document. The dance is wrapped in `try/finally` so the restore runs
  unconditionally — even if an exception is thrown mid-dance, the
  user's multi-tab view is recovered before the function returns.
  The earlier attempt to use drawio's `selectPage` postMessage was a
  no-op because drawio's JSON embed protocol does not expose that
  action — silently exporting whatever tab happened to be active. The
  load-export-restore approach trades a brief visible tab-flicker for
  correctness: the exported image is guaranteed to match the requested
  page.
* Tool description strings reflect the multi-page semantics so the LLM
  client learns the new contract.
* Package version bumped 0.2.0 → 0.3.0 (additive surface — four new
  tools, three extended input schemas, canonical XML shape change).
* CI: `.github/workflows/test.yml` gains an explicit install + vitest
  run for the mcp-server package so the new multi-page invariants are
  covered by automation, not just local runs.

Backward compatibility: every existing single-page caller continues to
work without modification. The session.xml shape is normalised on every
write, removing the wrapper-injection hack from the .drawio download
path.

Tests: 43 unit tests under `packages/mcp-server/tests/multi-page.test.ts`
pin the validator's mxfile path, the page-scoped operations, the XML
declaration-prefix handling for both normalizeToMxfile and addPageToDoc,
addPageToDoc's rejection of full <mxfile> inputs, the single-page
projection used by export_diagram (a direct regression test for the
selectPage bug — two distinct page selectors must produce visually
different projections), and the Transformer + CNN motivating scenario.
A `tests/smoke.mjs` smoke test drives the built `dist/index.js` over
JSON-RPC and asserts all 9 tools register with the right input schemas.
Root vitest suite (107 tests) still green.

* fix(mcp): rewrite page-targeted export browser-side; harden edit/get

The page-targeted PNG/SVG export never worked: export_diagram swapped the
live session to a single-page projection, slept 3s, then wrote the export
flag onto a state object that setState() had already replaced in the store
Map — so the browser never saw the request and every such export timed out.
The swap+restore also clobbered concurrent edits.

Move the projection entirely browser-side: requestExport() hands a single
-page <mxfile> to the bridge via state.exportXml; the bridge loads it,
lets draw.io render, exports, then reloads the user's real document. The
canonical session state is never mutated, so there is no restore race and
no fixed-delay guessing. The export poll now re-reads the live store entry
each tick instead of a captured reference. autosave is suppressed and the
version-bump reload is skipped while a projection is on screen; if no real
document was captured, restore forces a server reload rather than leaving
the iframe stuck on the projection.

Also:
- edit_diagram now returns isError on a page-level failure (selector matched
  no page / page has no <root>) instead of reporting success-with-warnings
  and persisting a no-op; the pre-edit history snapshot is taken only after
  that gate so a failed edit leaves no phantom undo entry.
- edit_diagram/get_diagram re-normalise browser-pushed xml to mxfile so a
  bare <mxGraphModel> can't silently strip a multi-page document.
- get_diagram now errors (instead of silently returning the full doc) when a
  selector is given but the session isn't a parseable mxfile.
- page_id / page_name / add_page.id get .min(1) so empty strings can't
  silently target the first page.
- Extract pages.ts:projectPage(), collapsing three copies of the
  parse→find→serialise projection logic in index.ts.
- Replace the never-in-CI tests/smoke.mjs with tests/server-wiring.test.ts,
  which boots the server from source via tsx and runs under the existing
  vitest CI step.

* chore(mcp): set version to 0.2.1 for release

---------

Co-authored-by: dayuan.jiang <jdy.toh@gmail.com>
2026-06-16 09:15:50 +09:00

329 lines
11 KiB
TypeScript

/**
* ID-based diagram operations
*
* The xmlContent argument may be either a bare <mxGraphModel> (legacy) or a
* full <mxfile> with one or more <diagram> pages. For mxfile inputs, an
* optional pageSelector identifies which page to edit; when omitted, the
* first page is targeted (the "active page by convention" — see pages.ts).
*/
import { findPageElement, hasPageSelector, type PageSelector } from "./pages.js"
export interface DiagramOperation {
operation: "update" | "add" | "delete"
cell_id: string
new_xml?: string
}
export interface OperationError {
type: "update" | "add" | "delete"
cellId: string
message: string
}
export interface ApplyOperationsResult {
result: string
errors: OperationError[]
}
/**
* Apply diagram operations (update/add/delete) using ID-based lookup.
*
* @param xmlContent - The diagram XML. May be either a bare <mxGraphModel> or
* a full <mxfile> with one or more <diagram> children.
* @param operations - Array of operations to apply.
* @param pageSelector - Optional page selector for multi-page docs. Defaults
* to the first page.
* @returns Object with result XML (same shape as input) and any per-op errors.
*/
export function applyDiagramOperations(
xmlContent: string,
operations: DiagramOperation[],
pageSelector?: PageSelector,
): ApplyOperationsResult {
const errors: OperationError[] = []
// Parse the XML
const parser = new DOMParser()
const doc = parser.parseFromString(xmlContent, "text/xml")
// Check for parse errors
const parseError = doc.querySelector("parsererror")
if (parseError) {
return {
result: xmlContent,
errors: [
{
type: "update",
cellId: "",
message: `XML parse error: ${parseError.textContent}`,
},
],
}
}
// Locate the <root> element to operate on.
//
// - For <mxfile> input: resolve the page via pageSelector, then dive into
// its <root>. This scopes querySelectorAll calls below to one page so
// cells on other pages aren't accidentally matched.
// - For bare <mxGraphModel> input: use the document's only <root>.
let root: Element | null
if (doc.documentElement?.tagName === "mxfile") {
const found = findPageElement(doc as unknown as Document, pageSelector)
if (!found) {
const selDesc = hasPageSelector(pageSelector)
? ` matching selector ${JSON.stringify(pageSelector)}`
: ""
return {
result: xmlContent,
errors: [
{
type: "update",
cellId: "",
message: `Page${selDesc} not found in <mxfile>`,
},
],
}
}
root = found.element.querySelector("root")
if (!root) {
const pageId =
found.element.getAttribute("id") || `(index ${found.index})`
return {
result: xmlContent,
errors: [
{
type: "update",
cellId: "",
message: `Page "${pageId}" has no <root> element`,
},
],
}
}
} else {
if (hasPageSelector(pageSelector)) {
return {
result: xmlContent,
errors: [
{
type: "update",
cellId: "",
message:
"Page selector provided but document is not multi-page (no <mxfile> wrapper). Use create_new_diagram with a full <mxfile> first, or omit the page selector.",
},
],
}
}
root = doc.querySelector("root")
if (!root) {
return {
result: xmlContent,
errors: [
{
type: "update",
cellId: "",
message: "Could not find <root> element in XML",
},
],
}
}
}
// Build a map of cell IDs to elements (scoped to the resolved page).
const cellMap = new Map<string, Element>()
root.querySelectorAll("mxCell").forEach((cell) => {
const id = cell.getAttribute("id")
if (id) cellMap.set(id, cell)
})
// Process each operation
for (const op of operations) {
if (op.operation === "update") {
const existingCell = cellMap.get(op.cell_id)
if (!existingCell) {
errors.push({
type: "update",
cellId: op.cell_id,
message: `Cell with id="${op.cell_id}" not found`,
})
continue
}
if (!op.new_xml) {
errors.push({
type: "update",
cellId: op.cell_id,
message: "new_xml is required for update operation",
})
continue
}
// Parse the new XML
const newDoc = parser.parseFromString(
`<wrapper>${op.new_xml}</wrapper>`,
"text/xml",
)
const newCell = newDoc.querySelector("mxCell")
if (!newCell) {
errors.push({
type: "update",
cellId: op.cell_id,
message: "new_xml must contain an mxCell element",
})
continue
}
// Validate ID matches
const newCellId = newCell.getAttribute("id")
if (newCellId !== op.cell_id) {
errors.push({
type: "update",
cellId: op.cell_id,
message: `ID mismatch: cell_id is "${op.cell_id}" but new_xml has id="${newCellId}"`,
})
continue
}
// Import and replace the node
const importedNode = doc.importNode(newCell, true)
existingCell.parentNode?.replaceChild(importedNode, existingCell)
// Update the map with the new element
cellMap.set(op.cell_id, importedNode)
} else if (op.operation === "add") {
// Check if ID already exists
if (cellMap.has(op.cell_id)) {
errors.push({
type: "add",
cellId: op.cell_id,
message: `Cell with id="${op.cell_id}" already exists`,
})
continue
}
if (!op.new_xml) {
errors.push({
type: "add",
cellId: op.cell_id,
message: "new_xml is required for add operation",
})
continue
}
// Parse the new XML
const newDoc = parser.parseFromString(
`<wrapper>${op.new_xml}</wrapper>`,
"text/xml",
)
const newCell = newDoc.querySelector("mxCell")
if (!newCell) {
errors.push({
type: "add",
cellId: op.cell_id,
message: "new_xml must contain an mxCell element",
})
continue
}
// Validate ID matches
const newCellId = newCell.getAttribute("id")
if (newCellId !== op.cell_id) {
errors.push({
type: "add",
cellId: op.cell_id,
message: `ID mismatch: cell_id is "${op.cell_id}" but new_xml has id="${newCellId}"`,
})
continue
}
// Import and append the node
const importedNode = doc.importNode(newCell, true)
root.appendChild(importedNode)
// Add to map
cellMap.set(op.cell_id, importedNode)
} else if (op.operation === "delete") {
// Protect root cells from deletion
if (op.cell_id === "0" || op.cell_id === "1") {
errors.push({
type: "delete",
cellId: op.cell_id,
message: `Cannot delete root cell "${op.cell_id}"`,
})
continue
}
const existingCell = cellMap.get(op.cell_id)
if (!existingCell) {
// Cell not found - might have been cascade-deleted by a previous operation
// Skip silently instead of erroring (AI may redundantly list children/edges)
continue
}
// Cascade delete: collect all cells to delete (children + edges + self)
const cellsToDelete = new Set<string>()
// Recursive function to find all descendants
const collectDescendants = (cellId: string) => {
if (cellsToDelete.has(cellId)) return
cellsToDelete.add(cellId)
// Find children (cells where parent === cellId)
// Scoped to `root` so other pages' cells with the same parent id
// (notably "1") are never touched.
const children = root!.querySelectorAll(
`mxCell[parent="${cellId}"]`,
)
children.forEach((child) => {
const childId = child.getAttribute("id")
if (childId && childId !== "0" && childId !== "1") {
collectDescendants(childId)
}
})
}
// Collect the target cell and all its descendants
collectDescendants(op.cell_id)
// Find edges referencing any of the cells to be deleted
// Also recursively collect children of those edges (e.g., edge labels)
for (const cellId of cellsToDelete) {
const referencingEdges = root.querySelectorAll(
`mxCell[source="${cellId}"], mxCell[target="${cellId}"]`,
)
referencingEdges.forEach((edge) => {
const edgeId = edge.getAttribute("id")
// Protect root cells from being added via edge references
if (edgeId && edgeId !== "0" && edgeId !== "1") {
// Recurse to collect edge's children (like labels)
collectDescendants(edgeId)
}
})
}
// Log what will be deleted
if (cellsToDelete.size > 1) {
console.log(
`[applyDiagramOperations] Cascade delete "${op.cell_id}" → deleting ${cellsToDelete.size} cells: ${Array.from(cellsToDelete).join(", ")}`,
)
}
// Delete all collected cells
for (const cellId of cellsToDelete) {
const cell = cellMap.get(cellId)
if (cell) {
cell.parentNode?.removeChild(cell)
cellMap.delete(cellId)
}
}
}
}
// Serialize back to string
const serializer = new XMLSerializer()
const result = serializer.serializeToString(doc)
return { result, errors }
}