Remove non-displayable chars from outline title (mozilla#14267)

 - it aims to fix mozilla#14267;
 - there is nothing about chars in range [0-1F] in the specs but acrobat doesn't display them in any way.

Author: calixteman

src/shared/util.js

@@ -566,16 +566,20 @@ class AbortException extends BaseException {
   }
 }
 
-const NullCharactersRegExp = /\x00/g;
+const NullCharactersRegExp = /\x00+/g;
+const InvisibleCharactersRegExp = /[\x01-\x1F]/g;
 
 /**
  * @param {string} str
  */
-function removeNullCharacters(str) {
+function removeNullCharacters(str, replaceInvisible = false) {
   if (typeof str !== "string") {
     warn("The argument for removeNullCharacters must be a string.");
     return str;
   }
+  if (replaceInvisible) {
+    str = str.replace(InvisibleCharactersRegExp, " ");
+  }
   return str.replace(NullCharactersRegExp, "");
 }
 

test/pdfs/.gitignore

@@ -486,3 +486,4 @@
 !pr12828.pdf
 !secHandler.pdf
 !rc_annotation.pdf
+!issue14267.pdf

test/pdfs/issue14267.pdf

@@ -1102,6 +1102,19 @@ describe("api", function () {
       await loadingTask.destroy();
     });
 
+    it("gets outline with non-displayable chars", async function () {
+      const loadingTask = getDocument(buildGetDocumentParams("issue14267.pdf"));
+      const pdfDoc = await loadingTask.promise;
+      const outline = await pdfDoc.getOutline();
+      expect(Array.isArray(outline)).toEqual(true);
+      expect(outline.length).toEqual(1);
+
+      const outlineItem = outline[0];
+      expect(outlineItem.title).toEqual("hello\x11world");
+
+      await loadingTask.destroy();
+    });
+
     it("gets non-existent permissions", async function () {
       const permissions = await pdfDocument.getPermissions();
       expect(permissions).toEqual(null);

test/unit/api_spec.js

@@ -185,6 +185,18 @@ describe("util", function () {
       const str = "string\x00With\x00Null\x00Chars";
       expect(removeNullCharacters(str)).toEqual("stringWithNullChars");
     });
+
+    it("should modify string with non-displayable characters", function () {
+      const str = Array.from(Array(32).keys())
+        .map(x => String.fromCharCode(x) + "a")
+        .join("");
+      // \x00 is replaced by an empty string.
+      const expected =
+        "a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a";
+      expect(removeNullCharacters(str, /* replaceInvisible */ true)).toEqual(
+        expected
+      );
+    });
   });
 
   describe("ReadableStream", function () {

test/unit/util_spec.js

@@ -59,7 +59,12 @@ class BaseTreeViewer {
    * @private
    */
   _normalizeTextContent(str) {
-    return removeNullCharacters(str) || /* en dash = */ "\u2013";
+    // Chars in range [0x01-0x1F] will be replaced with a white space
+    // and 0x00 by "".
+    return (
+      removeNullCharacters(str, /* replaceInvisible */ true) ||
+      /* en dash = */ "\u2013"
+    );
   }
 
   /**