Fix text extraction when ToUnicode cmap contains not default codespace ranges

DEVSIX-8035

Author: vitali-pr

kernel/src/main/java/com/itextpdf/kernel/font/PdfType0Font.java

@@ -744,27 +744,32 @@ private static String getOrdering(PdfDictionary cidFont) {
     }
 
     private static boolean containsCodeInCodeSpaceRange(List<byte[]> codeSpaceRanges, int code, int length) {
+        long unsignedCode = code & 0xffffffff;
         for (int i = 0; i < codeSpaceRanges.size(); i += 2) {
             if (length == codeSpaceRanges.get(i).length) {
-                int mask = 0xff;
-                int totalShift = 0;
                 byte[] low = codeSpaceRanges.get(i);
                 byte[] high = codeSpaceRanges.get(i + 1);
-                boolean fitsIntoRange = true;
-                for (int ind = length - 1; ind >= 0; ind--, totalShift += 8, mask <<= 8) {
-                    int actualByteValue = (code & mask) >> totalShift;
-                    if (!(actualByteValue >= (0xff & low[ind]) && actualByteValue <= (0xff & high[ind]))) {
-                        fitsIntoRange = false;
-                    }
-                }
-                if (fitsIntoRange) {
+                long lowValue = bytesToLong(low);
+                long highValue = bytesToLong(high);
+                if (unsignedCode >= lowValue && unsignedCode <= highValue) {
                     return true;
                 }
             }
         }
         return false;
     }
 
+    private static long bytesToLong(byte[] bytes) {
+        long res = 0;
+        int shift = 0;
+        for (int i = bytes.length - 1; i >= 0; --i) {
+            res += (bytes[i] & 0xff) << shift;
+            shift += 8;
+        }
+
+        return res;
+    }
+
     private void flushFontData() {
         if (cidFontType == CID_FONT_TYPE_0) {
             getPdfObject().put(PdfName.Type, PdfName.Font);

kernel/src/test/java/com/itextpdf/kernel/pdf/canvas/parser/PdfTextExtractorTest.java

@@ -132,4 +132,23 @@ public void shortOctalDataAsTextTest() throws IOException {
             Assert.assertEquals("EC", PdfTextExtractor.getTextFromPage(pdfDocument.getPage(1)));
         }
     }
+
+    @Test
+    public void notDefaultCodespacesCyrillicTest() throws IOException {
+        String inFile = sourceFolder + "notDefaultCodespacesCyrillic.pdf";
+        try (PdfDocument pdfDocument = new PdfDocument(new PdfReader(inFile))) {
+            String extractedText = PdfTextExtractor.getTextFromPage(pdfDocument.getPage(1));
+            Assert.assertTrue(extractedText.contains("бронирование"));
+            Assert.assertTrue(extractedText.contains("From"));
+        }
+    }
+
+    @Test
+    public void notDefaultCodespacesChineseTest() throws IOException {
+        String inFile = sourceFolder + "notDefaultCodespacesChinese.pdf";
+        try (PdfDocument pdfDocument = new PdfDocument(new PdfReader(inFile))) {
+            String extractedText = PdfTextExtractor.getTextFromPage(pdfDocument.getPage(1));
+            Assert.assertTrue(extractedText.contains("L3B 廠： 新竹科學工業園區新竹市東區力行二路 1 號"));
+        }
+    }
 }