feat(core): sourceMutation data-hf-id targeting (R1, T7) (#1272)

* feat(core): clip-model hf- ids minted at parse, emitted as data-hf-id (R1)

* docs(core): document legacy-id round-trip in clip-model readback (R1 review)

Addresses Rames' review on #1270: clarifies that a pre-R1 clip authored with
id="my-title" round-trips as data-hf-id="my-title" (non-hf-shaped but stable,
exact-match) by design — targeting uses exact [data-hf-id="…"] match and does
not require the hf- shape; legacy values re-mint only at the R7 write-back. Not
a bug. Comment-only.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* docs(core): fix misleading legacy-id migration comment in htmlParser.ts

The original comment said legacy data-hf-id values "are re-minted only
once the R7 write-back persists freshly-minted ids to source" — which is
incorrect. ensureHfIds skips elements that already carry data-hf-id, so
legacy values (e.g. data-hf-id="my-title") persist indefinitely and are
NOT automatically re-minted. Exact-match targeting still works correctly.
Update comment to reflect actual behaviour.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* feat(studio): sourcePatcher data-hf-id targeting (R1, T3)

* fix(studio): warn on duplicate match in execDataAttrPattern (R1, T3 review)

Addresses Rames' review on #1271: execDataAttrPattern returned the first regex
match without checking for a second. A duplicate id/data-hf-id in source (id
drift) would silently patch one element and leave the other stale. Now warns
when more than one element matches. By the mint contract it should never fire.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* test(studio): pin hfId-is-authoritative-over-selector contract (R1, T3 review)

Adds test: "hfId match is authoritative — selector is not used as a
narrowing filter". When hfId matches element A and selector points at
element B, findTagByTarget returns A without consulting selector as a
narrowing filter. Pins the intended behaviour so a future refactor
cannot silently start narrowing by selector.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* feat(core): sourceMutation data-hf-id targeting (R1, T7)

* test(core): update htmlParser baselines for R1 hf- id format

Elements now get data-hf-id minted by ensureHfIds; parser reads
data-hf-id as model id, so HTML id attrs are no longer the model id.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(core): data-hf-id survives id/selector patch (R1, T7)

Locks the preservation guarantee the write-back design depends on: a
Studio edit targeting by id or selector (it never sends hfId) must not strip
an existing data-hf-id, or the stable handle is destroyed by the next edit.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* fix(core): escape hfId in selector + warn on duplicate match (R1, T7 review)

Addresses review on #1272 (Miguel P3 + Rames): findTargetElement interpolated
target.hfId raw into a [data-hf-id="..."] selector. Escape it (CSS attr-value
injection guard) and warn when a hfId matches more than one element instead of
silently patching an arbitrary one. Adds an injection-guard test.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

packages/core/src/parsers/htmlParser.test.ts

@@ -26,13 +26,13 @@ describe("parseHtml", () => {
     const result = parseHtml(html);
 
     expect(result.elements).toHaveLength(2);
-    expect(result.elements[0].id).toBe("text1");
+    expect(result.elements[0].id).toMatch(/^hf-[a-z0-9]{4}$/);
     expect(result.elements[0].startTime).toBe(0);
     expect(result.elements[0].duration).toBe(5);
     expect(result.elements[0].name).toBe("Title");
     expect(result.elements[0].type).toBe("text");
 
-    expect(result.elements[1].id).toBe("text2");
+    expect(result.elements[1].id).toMatch(/^hf-[a-z0-9]{4}$/);
     expect(result.elements[1].startTime).toBe(2);
     expect(result.elements[1].duration).toBe(5);
   });
@@ -53,7 +53,7 @@ describe("parseHtml", () => {
 
     expect(result.elements).toHaveLength(1);
     expect(result.elements[0].type).toBe("composition");
-    expect(result.elements[0].id).toBe("comp1");
+    expect(result.elements[0].id).toMatch(/^hf-[a-z0-9]{4}$/);
     if (result.elements[0].type === "composition") {
       expect(result.elements[0].compositionId).toBe("abc123");
       expect(result.elements[0].src).toBe("/compositions/abc123");
@@ -76,20 +76,20 @@ describe("parseHtml", () => {
 
     expect(result.elements).toHaveLength(3);
 
-    const video = result.elements.find((e) => e.id === "vid1");
+    const video = result.elements.find((e) => e.type === "video");
     expect(video).toBeDefined();
-    expect(video?.type).toBe("video");
+    expect(video?.id).toMatch(/^hf-[a-z0-9]{4}$/);
     if (video?.type === "video") {
       expect(video.src).toBe("video.mp4");
     }
 
-    const audio = result.elements.find((e) => e.id === "aud1");
+    const audio = result.elements.find((e) => e.type === "audio");
     expect(audio).toBeDefined();
-    expect(audio?.type).toBe("audio");
+    expect(audio?.id).toMatch(/^hf-[a-z0-9]{4}$/);
 
-    const img = result.elements.find((e) => e.id === "img1");
+    const img = result.elements.find((e) => e.type === "image");
     expect(img).toBeDefined();
-    expect(img?.type).toBe("image");
+    expect(img?.id).toMatch(/^hf-[a-z0-9]{4}$/);
   });
 
   it("handles missing attributes gracefully", () => {
@@ -398,9 +398,12 @@ describe("parseHtml", () => {
     `;
     const result = parseHtml(html);
 
-    expect(result.keyframes["text1"]).toBeDefined();
-    expect(result.keyframes["text1"]).toHaveLength(2);
-    expect(result.keyframes["text1"][0].id).toBe("kf1");
+    const elId = result.elements[0]?.id ?? "";
+    expect(elId).toMatch(/^hf-[a-z0-9]{4}$/);
+    const kfs = result.keyframes[elId];
+    expect(kfs).toBeDefined();
+    expect(kfs).toHaveLength(2);
+    expect(kfs[0].id).toBe("kf1");
   });
 
   it("parses stage zoom keyframes", () => {

packages/core/src/studio-api/helpers/sourceMutation.test.ts

@@ -368,13 +368,90 @@ describe("probeElementInSource", () => {
 // Covers the same surface as T3 (Studio sourcePatcher) — Core sourceMutation supports
 // all patch types (inline-style, attribute, text-content) via patchElementInHtml.
 describe("T7 — data-hf-id targeting (spec for R1)", () => {
-  it.todo("updates inline style by data-hf-id when no HTML id attribute is present");
+  it("updates inline style by data-hf-id when no HTML id attribute is present", () => {
+    const source = `<h1 data-hf-id="hf-x7k2" style="color: red">Hello</h1>`;
+    const { html, matched } = patchElementInHtml(source, { hfId: "hf-x7k2" }, [
+      { type: "inline-style", property: "color", value: "blue" },
+    ]);
+    expect(matched).toBe(true);
+    expect(html).toMatch(/color:\s*blue/);
+    expect(html).toContain('data-hf-id="hf-x7k2"');
+  });
 
-  it.todo("updates text content by data-hf-id");
+  it("updates text content by data-hf-id", () => {
+    const source = `<p data-hf-id="hf-a1b2">Old text</p>`;
+    const { html, matched } = patchElementInHtml(source, { hfId: "hf-a1b2" }, [
+      { type: "text-content", property: "", value: "New text" },
+    ]);
+    expect(matched).toBe(true);
+    expect(html).toContain("New text");
+  });
 
-  it.todo("updates attribute by data-hf-id");
+  it("updates attribute by data-hf-id", () => {
+    const source = `<div data-hf-id="hf-c3d4" data-start="0"></div>`;
+    const { html, matched } = patchElementInHtml(source, { hfId: "hf-c3d4" }, [
+      { type: "attribute", property: "start", value: "2.5" },
+    ]);
+    expect(matched).toBe(true);
+    expect(html).toContain('data-start="2.5"');
+  });
 
-  it.todo("data-hf-id attribute survives the patch (can be targeted again)");
+  it("data-hf-id attribute survives the patch (can be targeted again)", () => {
+    const source = `<h1 data-hf-id="hf-x7k2" style="color: red">Hello</h1>`;
+    const { html } = patchElementInHtml(source, { hfId: "hf-x7k2" }, [
+      { type: "inline-style", property: "color", value: "blue" },
+    ]);
+    expect(html).toContain('data-hf-id="hf-x7k2"');
+  });
 
-  it.todo("hfId lookup falls through to selector when hfId is not found in the document");
+  it("hfId lookup falls through to selector when hfId is not found in the document", () => {
+    const source = `<h1 class="headline" style="color: red">Hello</h1>`;
+    const { html, matched } = patchElementInHtml(
+      source,
+      { hfId: "hf-missing", selector: ".headline" },
+      [{ type: "inline-style", property: "color", value: "blue" }],
+    );
+    expect(matched).toBe(true);
+    expect(html).toMatch(/color:\s*blue/);
+  });
+
+  it("does not break out of the selector on a crafted hfId (CSS injection guard)", () => {
+    // A value with a quote/bracket must be escaped, not injected — it should
+    // simply match nothing and leave the source untouched, never throw.
+    const source = `<h1 class="safe">A</h1><h1 class="victim">B</h1>`;
+    const evil = `x"] , [class="victim`;
+    const run = () =>
+      patchElementInHtml(source, { hfId: evil }, [
+        { type: "text-content", property: "textContent", value: "HACKED" },
+      ]);
+    expect(run).not.toThrow();
+    const { html, matched } = run();
+    expect(matched).toBe(false);
+    expect(html).toBe(source);
+    expect(html).not.toContain("HACKED");
+  });
+
+  // The Studio edit path targets by id/selector (it never sends hfId). Once a
+  // persisted data-hf-id exists in source, those edits must NOT strip it — else
+  // the stable handle is destroyed by the next edit. This is the preservation
+  // guarantee the write-back design depends on.
+  it("preserves an existing data-hf-id when the element is patched by id", () => {
+    const source = `<h1 id="hero" data-hf-id="hf-x7k2" style="color: red">Hello</h1>`;
+    const { html, matched } = patchElementInHtml(source, { id: "hero" }, [
+      { type: "inline-style", property: "color", value: "blue" },
+    ]);
+    expect(matched).toBe(true);
+    expect(html).toMatch(/color:\s*blue/);
+    expect(html).toContain('data-hf-id="hf-x7k2"');
+  });
+
+  it("preserves an existing data-hf-id when the element is patched by selector", () => {
+    const source = `<p class="body" data-hf-id="hf-a1b2">Old</p>`;
+    const { html, matched } = patchElementInHtml(source, { selector: ".body" }, [
+      { type: "text-content", property: "textContent", value: "New" },
+    ]);
+    expect(matched).toBe(true);
+    expect(html).toContain("New");
+    expect(html).toContain('data-hf-id="hf-a1b2"');
+  });
 });

packages/core/src/studio-api/helpers/sourceMutation.ts

@@ -2,6 +2,7 @@ import { parseHTML } from "linkedom";
 
 export interface SourceMutationTarget {
   id?: string | null;
+  hfId?: string;
   selector?: string;
   selectorIndex?: number;
 }
@@ -31,7 +32,41 @@ function querySelectorAllWithTemplates(root: Document | Element, selector: strin
   return [];
 }
 
+// Prevent CSS attribute-selector injection via a crafted hfId: escape
+// backslashes first, then double-quotes. Keeps a malformed/hostile value from
+// breaking out of the `[data-hf-id="…"]` selector once callers beyond the
+// internal mint contract (R2+ user flows) pass values here.
+function escapeCssAttrValue(value: string): string {
+  return value.replace(/\\/g, "\\\\").replace(/"/g, '\\"');
+}
+
+function findByHfId(document: Document, hfId: string): Element | null {
+  try {
+    const matches = querySelectorAllWithTemplates(
+      document,
+      `[data-hf-id="${escapeCssAttrValue(hfId)}"]`,
+    );
+    if (matches.length > 1) {
+      // The mint contract guarantees uniqueness; a duplicate means upstream
+      // id drift. Don't silently patch an arbitrary one — surface it.
+      // eslint-disable-next-line no-console
+      console.warn(
+        `sourceMutation: data-hf-id "${hfId}" matched ${matches.length} elements; using the first. ids must be unique per document.`,
+      );
+    }
+    return matches[0] ?? null;
+  } catch {
+    // Malformed selector despite escaping — let the caller fall back.
+    return null;
+  }
+}
+
 function findTargetElement(document: Document, target: SourceMutationTarget): Element | null {
+  if (target.hfId) {
+    const el = findByHfId(document, target.hfId);
+    if (el) return el;
+  }
+
   if (target.id) {
     const byId = document.getElementById(target.id);
     if (byId) return byId;
@@ -207,7 +242,7 @@ export function patchElementInHtml(
 }
 
 export function probeElementInSource(source: string, target: SourceMutationTarget): boolean {
-  if (!target.id && !target.selector) return false;
+  if (!target.id && !target.hfId && !target.selector) return false;
   const { document } = parseSourceDocument(source);
   const el = findTargetElement(document, target);
   return el != null && isHTMLElement(el);