fix(ts-sdk): decode base64 data URLs with media-type parameters

packages/sie_ts_sdk/src/images.ts

@@ -79,19 +79,61 @@ export async function toImageBytes(input: ImageInput): Promise<Uint8Array> {
 
   // Base64 string or data URL
   if (typeof input === "string") {
-    // Check if it's a data URL
-    const dataUrlMatch = input.match(/^data:[^;]+;base64,(.+)$/);
-    if (dataUrlMatch?.[1]) {
-      return base64ToBytes(dataUrlMatch[1]);
-    }
-
-    // Assume it's raw base64
-    return base64ToBytes(input);
+    // A `data:` URL is parsed structurally by parseBase64DataUrl; a plain
+    // base64 string (which can never begin with "data:", since ":" is not a
+    // base64 character) is decoded as-is.
+    const payload = parseBase64DataUrl(input);
+    return base64ToBytes(payload ?? input);
   }
 
   throw new Error(`Unsupported image input type: ${typeof input}`);
 }
 
+/**
+ * Extract the base64 payload from a `data:` URL.
+ *
+ * Rather than encoding the RFC 2397 grammar in a regex, this walks the URL
+ * structurally: everything before the first comma is the header, and the
+ * `;`-delimited metadata's final segment must be the `base64` marker. The
+ * scheme and marker are matched case-insensitively, and the payload is
+ * percent-decoded (so escaped characters such as `%3D` padding are restored)
+ * before it reaches the base64 decoder.
+ *
+ * @returns the base64 payload (possibly empty), or `undefined` when `input` is
+ *   not a `data:` URL at all — in which case the caller treats it as a raw
+ *   base64 string.
+ * @throws if `input` is a `data:` URL that is malformed (no payload delimiter
+ *   or invalid percent-encoding) or is not base64-encoded, so such inputs fail
+ *   loudly instead of silently corrupting in the base64 decoder.
+ */
+function parseBase64DataUrl(input: string): string | undefined {
+  if (!/^data:/i.test(input)) {
+    return undefined;
+  }
+
+  const comma = input.indexOf(",");
+  if (comma === -1) {
+    throw new Error("Malformed data URL: missing ',' delimiter between metadata and payload");
+  }
+
+  const metadata = input.slice("data:".length, comma);
+  const params = metadata.split(";");
+  // split(";") always yields at least one element, but the compiler can't prove
+  // it under noUncheckedIndexedAccess; optional chaining keeps this type-safe and
+  // still treats a missing marker as "not base64".
+  const marker = params.at(-1);
+  if (marker?.toLowerCase() !== "base64") {
+    throw new Error("Unsupported data URL: only base64-encoded payloads are supported");
+  }
+
+  const payload = input.slice(comma + 1);
+  try {
+    return decodeURIComponent(payload);
+  } catch {
+    throw new Error("Malformed data URL: payload has invalid percent-encoding");
+  }
+}
+
 /**
  * Convert base64 string to Uint8Array.
  */

packages/sie_ts_sdk/tests/images.test.ts

@@ -51,6 +51,63 @@ describe("toImageBytes", () => {
     expect(new TextDecoder().decode(result)).toBe("test");
   });
 
+  it("decodes a data URL whose media type carries a parameter", async () => {
+    // Valid per RFC 2397: the media type may be followed by ";param=value"
+    // (e.g. charset) before ";base64,". "Hello" base64-encoded.
+    const dataUrl = "data:image/svg+xml;charset=utf-8;base64,SGVsbG8=";
+    const result = await toImageBytes(dataUrl);
+
+    expect(result).toBeInstanceOf(Uint8Array);
+    expect(new TextDecoder().decode(result)).toBe("Hello");
+  });
+
+  it("decodes a data URL with an omitted media type", async () => {
+    // RFC 2397 permits an empty media type (defaults to text/plain).
+    const dataUrl = "data:;base64,SGVsbG8=";
+    const result = await toImageBytes(dataUrl);
+
+    expect(result).toBeInstanceOf(Uint8Array);
+    expect(new TextDecoder().decode(result)).toBe("Hello");
+  });
+
+  it("decodes a data URL with an uppercase scheme and BASE64 marker", async () => {
+    // RFC 2397: the scheme and the "base64" marker are case-insensitive.
+    const dataUrl = "DATA:image/png;BASE64,SGVsbG8=";
+    const result = await toImageBytes(dataUrl);
+
+    expect(result).toBeInstanceOf(Uint8Array);
+    expect(new TextDecoder().decode(result)).toBe("Hello");
+  });
+
+  it("percent-decodes the payload before base64 decoding", async () => {
+    // The "=" padding can arrive percent-escaped as "%3D".
+    const dataUrl = "data:image/png;base64,SGVsbG8%3D";
+    const result = await toImageBytes(dataUrl);
+
+    expect(result).toBeInstanceOf(Uint8Array);
+    expect(new TextDecoder().decode(result)).toBe("Hello");
+  });
+
+  it("returns an empty array for an empty base64 data URL payload", async () => {
+    const dataUrl = "data:;base64,";
+    const result = await toImageBytes(dataUrl);
+
+    expect(result).toBeInstanceOf(Uint8Array);
+    expect(result.length).toBe(0);
+  });
+
+  it("throws a clear error for a data URL that is not base64-encoded", async () => {
+    // A non-base64 data URL cannot yield image bytes; it must fail loudly
+    // instead of being handed to the base64 decoder.
+    await expect(toImageBytes("data:text/plain,Hello")).rejects.toThrow(
+      "only base64-encoded payloads are supported",
+    );
+  });
+
+  it("throws a clear error for a data URL missing the payload delimiter", async () => {
+    await expect(toImageBytes("data:image/png;base64")).rejects.toThrow("missing ',' delimiter");
+  });
+
   it("throws for unsupported input type", async () => {
     await expect(toImageBytes(123 as unknown as Uint8Array)).rejects.toThrow(
       "Unsupported image input type",