fix: type-check the e2e specs clean

.lantern/map/test/e2e.lantern

@@ -47,7 +47,7 @@ test/e2e/graph-navigation.spec.ts . test 0910cf6edfbd
 #feature file-graph
   = test/e2e/support/graph-harness.ts
 
-test/e2e/graph-pixi.spec.ts . test b75dd27e683c
+test/e2e/graph-pixi.spec.ts . test 661b9b4dc1e3
 #feature file-graph
   = test/e2e/support/graph-harness.ts
 
@@ -123,12 +123,14 @@ test/e2e/text-select.spec.ts . test b09fe6828137
   = src/renderer/lib/lanternview.ts#mountLanternView
   = test/fixtures/graph/pr-review-demo.ts#prReviewDemo
 
-test/e2e/tsconfig.json . config b024315ec7d5
+test/e2e/tsconfig.json . config 20fb9354fd10
   < playwright.config.ts
   < src/main/filecontent.ts
   < src/main/lantern-prfile.ts
   < src/main/lantern-types.ts
   < src/main/lantern.ts
+  < src/main/skill-scripts.d.ts
   < src/renderer/global.d.ts
+  < src/renderer/svg.d.ts
   < test/helpers/lantern-map.ts
   < tsconfig.json

.lantern/prs/35.pr

@@ -0,0 +1,26 @@
+lantern-pr 1
+# 35
+#title fix: type-check the e2e specs clean
+#url https://github.com/MobileReality/lantern/pull/35
+#branch fix/e2e-typecheck-green
+#base main
+#opened 2026-11-12
+#head cf67f57134b4d52fcc4907cdcf253d7f4135e0da
+#describe what The e2e typecheck now passes: its tsconfig includes the existing svg and skill-script declarations, and one spec uses the typed toBitmap.
+#describe before pnpm run test:graph:typecheck failed with about 64 errors.
+#describe after It exits zero with no suppressions.
+#describe verified test:graph:typecheck, typecheck, lint, graph-pixi e2e (14 passed).
+~ test/e2e/graph-pixi.spec.ts
+~ test/e2e/tsconfig.json
+#order test/e2e/tsconfig.json
+#order test/e2e/graph-pixi.spec.ts
+#reason test/e2e/graph-pixi.spec.ts getBitmap is typed void in Electron's types; toBitmap returns the Buffer the checks need.
+#reason test/e2e/tsconfig.json Lets the e2e typecheck see the svg and skill-script declarations the main tsconfig already has.
+#story Type-checking the graph specs
+#step actor A developer
+#file test/e2e/tsconfig.json
+#step action runs the graph spec typecheck
+#step subject the e2e specs
+#file test/e2e/graph-pixi.spec.ts
+#below interaction Trust a green typecheck for the specs
+#below consequence The typecheck failed with many errors => The typecheck passes

test/e2e/graph-pixi.spec.ts

@@ -38,8 +38,8 @@ test('repaints retained cards when platform tabs change their column width', asy
     if (!freshCanvasBox) throw new Error('Pixi canvas is not visible');
     const fresh = await page.screenshot({ path: testInfo.outputPath('fresh.png'), clip: { ...comparisonClip, x: freshCanvasBox.x, y: freshCanvasBox.y } });
     const difference = await harness.app.evaluate(({ nativeImage }, images) => {
-      const navigatedBitmap = nativeImage.createFromBuffer(Buffer.from(images.navigated, 'base64')).getBitmap();
-      const freshBitmap = nativeImage.createFromBuffer(Buffer.from(images.fresh, 'base64')).getBitmap();
+      const navigatedBitmap = nativeImage.createFromBuffer(Buffer.from(images.navigated, 'base64')).toBitmap();
+      const freshBitmap = nativeImage.createFromBuffer(Buffer.from(images.fresh, 'base64')).toBitmap();
       let channelDelta = 0;
       for (let index = 0; index < freshBitmap.length; index += 1) channelDelta += Math.abs(navigatedBitmap[index] - freshBitmap[index]);
       return channelDelta / freshBitmap.length / 255;
@@ -433,7 +433,7 @@ test('matches all four relation edge styles and highlights the canvas node under
     await page.mouse.move(canvasBox.x + 4, canvasBox.y + 4);
     const stable = await canvas.screenshot();
     const pixels = await harness.app.evaluate(({ nativeImage }, images) => {
-      const bitmap = (image: string) => nativeImage.createFromBuffer(Buffer.from(image, 'base64')).getBitmap();
+      const bitmap = (image: string) => nativeImage.createFromBuffer(Buffer.from(image, 'base64')).toBitmap();
       return {
         highlighted: bitmap(images.highlighted).equals(bitmap(images.cleared)),
         stable: bitmap(images.stable).equals(bitmap(images.cleared)),
@@ -505,8 +505,8 @@ test('keeps cached tile text consistent with freshly rendered tiles after zoomin
     await expect(host).toHaveAttribute('style', zoomedWidth!);
     const fresh = await canvas.screenshot({ path: testInfo.outputPath('fresh.png') });
     const difference = await harness.app.evaluate(({ nativeImage }, images) => {
-      const before = nativeImage.createFromBuffer(Buffer.from(images.before, 'base64')).getBitmap();
-      const after = nativeImage.createFromBuffer(Buffer.from(images.after, 'base64')).getBitmap();
+      const before = nativeImage.createFromBuffer(Buffer.from(images.before, 'base64')).toBitmap();
+      const after = nativeImage.createFromBuffer(Buffer.from(images.after, 'base64')).toBitmap();
       if (before.length !== after.length) return 1;
       let changed = 0;
       for (let index = 0; index < before.length; index += 1) if (before[index] !== after[index]) changed += 1;

test/e2e/tsconfig.json

@@ -13,6 +13,8 @@
     "../../src/main/filecontent.ts",
     "../../src/main/lantern-prfile.ts",
     "../../src/renderer/global.d.ts",
+    "../../src/renderer/svg.d.ts",
+    "../../src/main/skill-scripts.d.ts",
     "../../src/renderer/lib/**/*.ts"
   ]
 }