Add browser upload and viewer HTML resolution for HTML/Markdown

ingestHtmlFromFile and ingestMarkdownFromFile store session bytes with
correct content types. resolveDomViewerHtml renders sanitized HTML or
markdown paragraphs for HtmlViewerAdapter (EANCH-WP-0004).

src/browser/byte-store.ts

@@ -25,7 +25,11 @@ export interface PdfByteRecord {
 }
 
 export interface PdfByteStore {
-  put(documentId: DocumentId, bytes: Uint8Array): PdfByteRecord;
+  put(
+    documentId: DocumentId,
+    bytes: Uint8Array,
+    contentType?: string,
+  ): PdfByteRecord;
   get(documentId: DocumentId): PdfByteRecord | null;
   has(documentId: DocumentId): boolean;
   delete(documentId: DocumentId): boolean;
@@ -70,7 +74,7 @@ export function createPdfByteStore(
   const records = new Map<DocumentId, PdfByteRecord>();
 
   return {
-    put(documentId, bytes) {
+    put(documentId, bytes, contentType = "application/pdf") {
       // Replace previous record (revoking the prior URL) if any.
       const prior = records.get(documentId);
       if (prior) revokeUrl(prior.blobUrl);
@@ -79,7 +83,7 @@ export function createPdfByteStore(
       // refuses without help. The bytes here always come from a fresh
       // arrayBuffer() call, so a regular ArrayBuffer is guaranteed.
       const blob = new Blob([bytes as unknown as ArrayBuffer], {
-        type: "application/pdf",
+        type: contentType,
       });
       const blobUrl = createUrl(blob);
       const record: PdfByteRecord = { bytes, blobUrl };

src/browser/index.ts

@@ -15,6 +15,12 @@ export {
   type PdfByteStore,
 } from "./byte-store";
 export {
+  ingestHtmlFromFile,
+  ingestMarkdownFromFile,
   ingestPdfFromFile,
   type IngestPdfFromFileOptions,
 } from "./upload";
+export {
+  markdownSourceToViewerHtml,
+  resolveDomViewerHtml,
+} from "./view-html";

src/browser/upload.ts

@@ -18,6 +18,8 @@
  * place that revokes them.
  */
 
+import { ingestHtml, type IngestHtmlResult } from "../html/ingest";
+import { ingestMarkdown, type IngestMarkdownResult } from "../markdown/ingest";
 import { ingestPdf, type IngestPdfResult } from "../pdf/ingest";
 import type { PdfByteStore } from "./byte-store";
 
@@ -37,7 +39,39 @@ export async function ingestPdfFromFile(
     ...(filename !== undefined ? { filename } : {}),
     ...(options.title !== undefined ? { title: options.title } : {}),
   });
-  const record = store.put(ingested.document.id, bytes);
+  const record = store.put(ingested.document.id, bytes, "application/pdf");
+  const document = { ...ingested.document, uri: record.blobUrl };
+  return { document, representation: ingested.representation };
+}
+
+export async function ingestHtmlFromFile(
+  file: File | Blob,
+  store: PdfByteStore,
+  options: IngestPdfFromFileOptions = {},
+): Promise<IngestHtmlResult> {
+  const bytes = new Uint8Array(await file.arrayBuffer());
+  const filename = "name" in file && typeof file.name === "string" ? file.name : undefined;
+  const ingested = await ingestHtml(bytes, {
+    ...(filename !== undefined ? { filename } : {}),
+    ...(options.title !== undefined ? { title: options.title } : {}),
+  });
+  const record = store.put(ingested.document.id, bytes, "text/html");
+  const document = { ...ingested.document, uri: record.blobUrl };
+  return { document, representation: ingested.representation };
+}
+
+export async function ingestMarkdownFromFile(
+  file: File | Blob,
+  store: PdfByteStore,
+  options: IngestPdfFromFileOptions = {},
+): Promise<IngestMarkdownResult> {
+  const bytes = new Uint8Array(await file.arrayBuffer());
+  const filename = "name" in file && typeof file.name === "string" ? file.name : undefined;
+  const ingested = await ingestMarkdown(bytes, {
+    ...(filename !== undefined ? { filename } : {}),
+    ...(options.title !== undefined ? { title: options.title } : {}),
+  });
+  const record = store.put(ingested.document.id, bytes, "text/markdown");
   const document = { ...ingested.document, uri: record.blobUrl };
   return { document, representation: ingested.representation };
 }

src/browser/view-html.test.ts

@@ -0,0 +1,23 @@
+import { describe, expect, it } from "vitest";
+import { createPdfByteStore } from "./byte-store";
+import { markdownSourceToViewerHtml, resolveDomViewerHtml } from "./view-html";
+import { ingestHtmlFromFile } from "./upload";
+
+const HTML = `<!DOCTYPE html><html><body><p>Hello HTML viewer</p></body></html>`;
+
+describe("resolveDomViewerHtml", () => {
+  it("returns sanitized HTML for html-dom representations", async () => {
+    const store = createPdfByteStore();
+    const file = new File([HTML], "sample.html", { type: "text/html" });
+    const { document, representation } = await ingestHtmlFromFile(file, store);
+    const html = resolveDomViewerHtml(document, representation, store);
+    expect(html).toContain("Hello HTML viewer");
+    expect(html).not.toMatch(/<script/i);
+  });
+
+  it("renders markdown as paragraph HTML", () => {
+    const html = markdownSourceToViewerHtml("# Title\n\nBody paragraph.");
+    expect(html).toContain("<p>");
+    expect(html).toContain("Body paragraph.");
+  });
+});

src/browser/view-html.ts

@@ -0,0 +1,41 @@
+import type { Document, DocumentRepresentation } from "@citation-evidence/engine/shared";
+import { extractHtml } from "../html/extract";
+import { extractMarkdown } from "../markdown/extract";
+import type { PdfByteStore } from "./byte-store";
+
+function escapeHtml(value: string): string {
+  return value
+    .replace(/&/g, "&")
+    .replace(/</g, "&lt;")
+    .replace(/>/g, "&gt;")
+    .replace(/"/g, "&quot;");
+}
+
+/** Render markdown canonical text as selectable HTML paragraphs. */
+export function markdownSourceToViewerHtml(source: string): string {
+  const { canonicalText } = extractMarkdown(source);
+  const paragraphs = canonicalText.split(/\n\n+/).map((p) => p.trim()).filter(Boolean);
+  const body = paragraphs.map((p) => `<p>${escapeHtml(p)}</p>`).join("");
+  return `<article class="ea-markdown-body">${body}</article>`;
+}
+
+/**
+ * Resolve viewer HTML for a non-paginated document from session byte storage.
+ * Returns `null` when bytes are missing or the representation is unsupported.
+ */
+export function resolveDomViewerHtml(
+  document: Document,
+  representation: DocumentRepresentation,
+  store: PdfByteStore,
+): string | null {
+  const record = store.get(document.id);
+  if (!record) return null;
+  const source = new TextDecoder("utf-8", { fatal: false }).decode(record.bytes);
+  if (representation.representationType === "html-dom") {
+    return extractHtml(source).sanitizedHtml;
+  }
+  if (representation.representationType === "markdown-rendered") {
+    return markdownSourceToViewerHtml(source);
+  }
+  return null;
+}