Merge pull request #20596 from Snuffleupagus/FileSpec-fixes

Simplify the `FileSpec` class, and remove no longer needed polyfills

Author: timvandermeij

src/core/annotation.js

@@ -5290,8 +5290,8 @@ class FileAttachmentAnnotation extends MarkupAnnotation {
   constructor(params) {
     super(params);
 
-    const { dict, xref } = params;
-    const file = new FileSpec(dict.get("FS"), xref);
+    const { dict } = params;
+    const file = new FileSpec(dict.get("FS"));
 
     this.data.annotationType = AnnotationType.FILEATTACHMENT;
     this.data.hasOwnCanvas = this.data.noRotate;

src/core/catalog.js

@@ -1057,7 +1057,7 @@ class Catalog {
     if (obj instanceof Dict && obj.has("EmbeddedFiles")) {
       const nameTree = new NameTree(obj.getRaw("EmbeddedFiles"), this.xref);
       for (const [key, value] of nameTree.getAll()) {
-        const fs = new FileSpec(value, this.xref);
+        const fs = new FileSpec(value);
         attachments ??= Object.create(null);
         attachments[stringToPDFString(key, /* keepEscapeSequence = */ true)] =
           fs.serializable;
@@ -1623,23 +1623,21 @@ class Catalog {
         case "GoToR":
           const urlDict = action.get("F");
           if (urlDict instanceof Dict) {
-            const fs = new FileSpec(
-              urlDict,
-              /* xref = */ null,
-              /* skipContent = */ true
-            );
-            const { rawFilename } = fs.serializable;
-            url = rawFilename;
+            const fs = new FileSpec(urlDict, /* skipContent = */ true);
+            ({ rawFilename: url } = fs.serializable);
           } else if (typeof urlDict === "string") {
             url = urlDict;
+          } else {
+            break;
           }
 
           // NOTE: the destination is relative to the *remote* document.
           const remoteDest = fetchRemoteDest(action);
-          if (remoteDest && typeof url === "string") {
+          if (remoteDest) {
             // NOTE: We don't use the `updateUrlHash` function here, since
-            // the `createValidAbsoluteUrl` function (see below) already
-            // handles parsing and validation of the final URL.
+            // the `createValidAbsoluteUrl` function (see below) already handles
+            // parsing/validation of the final URL and manual splitting also
+            // ensures that the `unsafeUrl` property will be available/correct.
             url = /* baseUrl = */ url.split("#", 1)[0] + "#" + remoteDest;
           }
           // The 'NewWindow' property, equal to `LinkTarget.BLANK`.

src/core/document.js

@@ -27,7 +27,6 @@ import {
   stringToBytes,
   stringToPDFString,
   stringToUTF8String,
-  toHexUtil,
   unreachable,
   Util,
   warn,
@@ -1628,8 +1627,8 @@ class PDFDocument {
     }
 
     return shadow(this, "fingerprints", [
-      toHexUtil(hashOriginal),
-      hashModified ? toHexUtil(hashModified) : null,
+      hashOriginal.toHex(),
+      hashModified?.toHex() ?? null,
     ]);
   }
 

src/core/file_spec.js

@@ -13,26 +13,18 @@
  * limitations under the License.
  */
 
-import { shadow, stringToPDFString, warn } from "../shared/util.js";
+import { stringToPDFString, warn } from "../shared/util.js";
 import { BaseStream } from "./base_stream.js";
 import { Dict } from "./primitives.js";
 
 function pickPlatformItem(dict) {
-  if (!(dict instanceof Dict)) {
-    return null;
-  }
-  // Look for the filename in this order:
-  // UF, F, Unix, Mac, DOS
-  if (dict.has("UF")) {
-    return dict.get("UF");
-  } else if (dict.has("F")) {
-    return dict.get("F");
-  } else if (dict.has("Unix")) {
-    return dict.get("Unix");
-  } else if (dict.has("Mac")) {
-    return dict.get("Mac");
-  } else if (dict.has("DOS")) {
-    return dict.get("DOS");
+  if (dict instanceof Dict) {
+    // Look for the filename in this order: UF, F, Unix, Mac, DOS
+    for (const key of ["UF", "F", "Unix", "Mac", "DOS"]) {
+      if (dict.has(key)) {
+        return dict.get(key);
+      }
+    }
   }
   return null;
 }
@@ -51,11 +43,10 @@ function stripPath(str) {
 class FileSpec {
   #contentAvailable = false;
 
-  constructor(root, xref, skipContent = false) {
+  constructor(root, skipContent = false) {
     if (!(root instanceof Dict)) {
       return;
     }
-    this.xref = xref;
     this.root = root;
     if (root.has("FS")) {
       this.fs = root.get("FS");
@@ -73,56 +64,46 @@ class FileSpec {
   }
 
   get filename() {
-    let filename = "";
-
     const item = pickPlatformItem(this.root);
     if (item && typeof item === "string") {
-      filename = stringToPDFString(item, /* keepEscapeSequence = */ true)
+      // NOTE: The following replacement order is INTENTIONAL, regardless of
+      //       what some static code analysers (e.g. CodeQL) may claim.
+      return stringToPDFString(item, /* keepEscapeSequence = */ true)
         .replaceAll("\\\\", "\\")
         .replaceAll("\\/", "/")
         .replaceAll("\\", "/");
     }
-    return shadow(this, "filename", filename || "unnamed");
+    return "";
   }
 
   get content() {
     if (!this.#contentAvailable) {
       return null;
     }
-    this._contentRef ||= pickPlatformItem(this.root?.get("EF"));
+    const ef = pickPlatformItem(this.root?.get("EF"));
 
-    let content = null;
-    if (this._contentRef) {
-      const fileObj = this.xref.fetchIfRef(this._contentRef);
-      if (fileObj instanceof BaseStream) {
-        content = fileObj.getBytes();
-      } else {
-        warn(
-          "Embedded file specification points to non-existing/invalid content"
-        );
-      }
-    } else {
-      warn("Embedded file specification does not have any content");
+    if (ef instanceof BaseStream) {
+      return ef.getBytes();
     }
-    return content;
+    warn("Embedded file specification points to non-existing/invalid content");
+    return null;
   }
 
   get description() {
-    let description = "";
-
     const desc = this.root?.get("Desc");
     if (desc && typeof desc === "string") {
-      description = stringToPDFString(desc);
+      return stringToPDFString(desc);
     }
-    return shadow(this, "description", description);
+    return "";
   }
 
   get serializable() {
+    const { filename, content, description } = this;
     return {
-      rawFilename: this.filename,
-      filename: stripPath(this.filename),
-      content: this.content,
-      description: this.description,
+      rawFilename: filename,
+      filename: stripPath(filename) || "unnamed",
+      content,
+      description,
     };
   }
 }

src/core/xfa/template.js

@@ -90,7 +90,6 @@ import {
   XFAObject,
   XFAObjectArray,
 } from "./xfa_object.js";
-import { fromBase64Util, Util, warn } from "../../shared/util.js";
 import {
   getBBox,
   getColor,
@@ -103,6 +102,7 @@ import {
   getStringOption,
   HTMLResult,
 } from "./utils.js";
+import { Util, warn } from "../../shared/util.js";
 import { getMetrics } from "./fonts.js";
 import { recoverJsURL } from "../core_utils.js";
 import { searchNode } from "./som.js";
@@ -3420,7 +3420,7 @@ class Image extends StringObject {
     }
 
     if (!buffer && this.transferEncoding === "base64") {
-      buffer = fromBase64Util(this[$content]);
+      buffer = Uint8Array.fromBase64(this[$content]);
     }
 
     if (!buffer) {

src/display/editor/drawers/signaturedraw.js

@@ -13,10 +13,10 @@
  * limitations under the License.
  */
 
-import { fromBase64Util, toBase64Util, warn } from "../../../shared/util.js";
 import { ContourDrawOutline } from "./contour.js";
 import { InkDrawOutline } from "./inkdraw.js";
 import { Outline } from "./outline.js";
+import { warn } from "../../../shared/util.js";
 
 const BASE_HEADER_LENGTH = 8;
 const POINTS_PROPERTIES_NUMBER = 3;
@@ -749,12 +749,12 @@ class SignatureExtractor {
     const buf = await new Response(cs.readable).arrayBuffer();
     const bytes = new Uint8Array(buf);
 
-    return toBase64Util(bytes);
+    return bytes.toBase64();
   }
 
   static async decompressSignature(signatureData) {
     try {
-      const bytes = fromBase64Util(signatureData);
+      const bytes = Uint8Array.fromBase64(signatureData);
       const { readable, writable } = new DecompressionStream("deflate-raw");
       const writer = writable.getWriter();
       await writer.ready;

src/display/font_loader.js

@@ -19,7 +19,6 @@ import {
   isNodeJS,
   shadow,
   string32,
-  toBase64Util,
   unreachable,
   warn,
 } from "../shared/util.js";
@@ -408,7 +407,7 @@ class FontFaceObject {
       return null;
     }
     // Add the @font-face rule to the document.
-    const url = `url(data:${this.mimetype};base64,${toBase64Util(this.data)});`;
+    const url = `url(data:${this.mimetype};base64,${this.data.toBase64()});`;
     let rule;
     if (!this.cssFontInfo) {
       rule = `@font-face {font-family:"${this.loadedName}";src:${url}}`;

src/shared/util.js

@@ -1235,44 +1235,6 @@ function MathClamp(v, min, max) {
   return Math.min(Math.max(v, min), max);
 }
 
-// TODO: Remove this once `Uint8Array.prototype.toHex` is generally available.
-function toHexUtil(arr) {
-  if (Uint8Array.prototype.toHex) {
-    return arr.toHex();
-  }
-  return Array.from(arr, num => hexNumbers[num]).join("");
-}
-
-// TODO: Remove this once `Uint8Array.prototype.toBase64` is generally
-//       available.
-function toBase64Util(arr) {
-  if (Uint8Array.prototype.toBase64) {
-    return arr.toBase64();
-  }
-  return btoa(bytesToString(arr));
-}
-
-// TODO: Remove this once `Uint8Array.fromBase64` is generally available.
-function fromBase64Util(str) {
-  if (Uint8Array.fromBase64) {
-    return Uint8Array.fromBase64(str);
-  }
-  return stringToBytes(atob(str));
-}
-
-// TODO: Remove this once https://bugzilla.mozilla.org/show_bug.cgi?id=1928493
-//       is fixed.
-if (
-  (typeof PDFJSDev === "undefined" || PDFJSDev.test("SKIP_BABEL")) &&
-  typeof Promise.try !== "function"
-) {
-  Promise.try = function (fn, ...args) {
-    return new Promise(resolve => {
-      resolve(fn(...args));
-    });
-  };
-}
-
 // TODO: Remove this once the `javascript.options.experimental.math_sumprecise`
 //       preference is removed from Firefox.
 if (typeof Math.sumPrecise !== "function") {
@@ -1338,7 +1300,6 @@ export {
   FeatureTest,
   FONT_IDENTITY_MATRIX,
   FormatError,
-  fromBase64Util,
   getModificationDate,
   getUuid,
   getVerbosityLevel,
@@ -1368,8 +1329,6 @@ export {
   stringToPDFString,
   stringToUTF8String,
   TextRenderingMode,
-  toBase64Util,
-  toHexUtil,
   UnknownErrorException,
   unreachable,
   updateUrlHash,