Merge pull request #17331 from calixteman/lang_marker

Remove language codes from text strings.

Author: calixteman

src/shared/util.js

@@ -905,12 +905,21 @@ const PDFStringTranslateTable = [
 ];
 
 function stringToPDFString(str) {
+  // See section 7.9.2.2 Text String Type.
+  // The string can contain some language codes bracketed with 0x0b,
+  // so we must remove them.
   if (str[0] >= "\xEF") {
     let encoding;
     if (str[0] === "\xFE" && str[1] === "\xFF") {
       encoding = "utf-16be";
+      if (str.length % 2 === 1) {
+        str = str.slice(0, -1);
+      }
     } else if (str[0] === "\xFF" && str[1] === "\xFE") {
       encoding = "utf-16le";
+      if (str.length % 2 === 1) {
+        str = str.slice(0, -1);
+      }
     } else if (str[0] === "\xEF" && str[1] === "\xBB" && str[2] === "\xBF") {
       encoding = "utf-8";
     }
@@ -919,7 +928,11 @@ function stringToPDFString(str) {
       try {
         const decoder = new TextDecoder(encoding, { fatal: true });
         const buffer = stringToBytes(str);
-        return decoder.decode(buffer);
+        const decoded = decoder.decode(buffer);
+        if (!decoded.includes("\x1b")) {
+          return decoded;
+        }
+        return decoded.replaceAll(/\x1b[^\x1b]*(?:\x1b|$)/g, "");
       } catch (ex) {
         warn(`stringToPDFString: "${ex}".`);
       }
@@ -928,7 +941,13 @@ function stringToPDFString(str) {
   // ISO Latin 1
   const strBuf = [];
   for (let i = 0, ii = str.length; i < ii; i++) {
-    const code = PDFStringTranslateTable[str.charCodeAt(i)];
+    const charCode = str.charCodeAt(i);
+    if (charCode === 0x1b) {
+      // eslint-disable-next-line no-empty
+      while (++i < ii && str.charCodeAt(i) !== 0x1b) {}
+      continue;
+    }
+    const code = PDFStringTranslateTable[charCode];
     strBuf.push(code ? String.fromCharCode(code) : str.charAt(i));
   }
   return strBuf.join("");

test/unit/util_spec.js

@@ -99,11 +99,21 @@ describe("util", function () {
       expect(stringToPDFString(str)).toEqual("string");
     });
 
+    it("handles incomplete UTF-16 big-endian strings", function () {
+      const str = "\xFE\xFF\x00\x73\x00\x74\x00\x72\x00\x69\x00\x6E\x00";
+      expect(stringToPDFString(str)).toEqual("strin");
+    });
+
     it("handles UTF-16 little-endian strings", function () {
       const str = "\xFF\xFE\x73\x00\x74\x00\x72\x00\x69\x00\x6E\x00\x67\x00";
       expect(stringToPDFString(str)).toEqual("string");
     });
 
+    it("handles incomplete UTF-16 little-endian strings", function () {
+      const str = "\xFF\xFE\x73\x00\x74\x00\x72\x00\x69\x00\x6E\x00\x67";
+      expect(stringToPDFString(str)).toEqual("strin");
+    });
+
     it("handles UTF-8 strings", function () {
       const simpleStr = "\xEF\xBB\xBF\x73\x74\x72\x69\x6E\x67";
       expect(stringToPDFString(simpleStr)).toEqual("string");
@@ -134,6 +144,22 @@ describe("util", function () {
       const str4 = "\xEF\xBB\xBF";
       expect(stringToPDFString(str4)).toEqual("");
     });
+
+    it("handles strings with language code", function () {
+      // ISO Latin 1
+      const str1 = "hello \x1benUS\x1bworld";
+      expect(stringToPDFString(str1)).toEqual("hello world");
+
+      // UTF-16BE
+      const str2 =
+        "\xFE\xFF\x00h\x00e\x00l\x00l\x00o\x00 \x00\x1b\x00e\x00n\x00U\x00S\x00\x1b\x00w\x00o\x00r\x00l\x00d";
+      expect(stringToPDFString(str2)).toEqual("hello world");
+
+      // UTF-16LE
+      const str3 =
+        "\xFF\xFEh\x00e\x00l\x00l\x00o\x00 \x00\x1b\x00e\x00n\x00U\x00S\x00\x1b\x00w\x00o\x00r\x00l\x00d\x00";
+      expect(stringToPDFString(str3)).toEqual("hello world");
+    });
   });
 
   describe("ReadableStream", function () {