jhy
diff --git a/‎src/main/java/org/jsoup/parser/Tokeniser.java‎
Lines changed: 17 additions & 0 deletions b/‎src/main/java/org/jsoup/parser/Tokeniser.java‎
Lines changed: 17 additions & 0 deletions
diff --git a/‎src/test/java/org/jsoup/parser/TokeniserTest.java‎
Lines changed: 26 additions & 0 deletions b/‎src/test/java/org/jsoup/parser/TokeniserTest.java‎
Lines changed: 26 additions & 0 deletions
@@ -13,6 +13,17 @@ final class Tokeniser {
     static final char replacementChar = '\uFFFD'; // replaces null character
     private static final char[] notCharRefCharsSorted = new char[]{'\t', '\n', '\r', '\f', ' ', '<', '&'};
 
+    // Some illegal character escapes are parsed by browsers as windows-1252 instead. See issue #1034
+    static final int win1252ExtensionsStart = 0x80;
+    static final int[] win1252Extensions = new int[] {
+            // we could build this manually, but Windows-1252 is not a standard java charset so that could break on
+            // some platforms - this table is verified with a test
+            0x20AC, 0x0081, 0x201A, 0x0192, 0x201E, 0x2026, 0x2020, 0x2021,
+            0x02C6, 0x2030, 0x0160, 0x2039, 0x0152, 0x008D, 0x017D, 0x008F,
+            0x0090, 0x2018, 0x2019, 0x201C, 0x201D, 0x2022, 0x2013, 0x2014,
+            0x02DC, 0x2122, 0x0161, 0x203A, 0x0153, 0x009D, 0x017E, 0x0178,
+    };
+
     static {
         Arrays.sort(notCharRefCharsSorted);
     }
@@ -148,6 +159,12 @@ int[] consumeCharacterReference(Character additionalAllowedCharacter, boolean in
                 codeRef[0] = replacementChar;
                 return codeRef;
             } else {
+                // fix illegal unicode characters to match browser behavior
+                if (charval >= win1252ExtensionsStart && charval < win1252ExtensionsStart + win1252Extensions.length) {
+                    characterReferenceError("character is not a valid unicode code point");
+                    charval = win1252Extensions[charval - win1252ExtensionsStart];
+                }
+
                 // todo: implement number replacement table
                 // todo: check for extra illegal unicode points as parse errors
                 codeRef[0] = charval;
 
@@ -1,5 +1,6 @@
 package org.jsoup.parser;
 
+import java.io.UnsupportedEncodingException;
 import org.jsoup.Jsoup;
 import org.jsoup.nodes.Attribute;
 import org.jsoup.nodes.Comment;
@@ -154,4 +155,29 @@ public void bufferUpInAttributeVal() {
         assertEquals(title, child.getWholeText());
         assertEquals(title, doc.title());
     }
+
+    @Test public void cp1252Entities() {
+        assertEquals("\u20ac", Jsoup.parse("&#0128;").text());
+        assertEquals("\u201a", Jsoup.parse("&#0130;").text());
+        assertEquals("\u20ac", Jsoup.parse("&#x80;").text());
+    }
+
+    @Test public void cp1252EntitiesProduceError() {
+        Parser parser = new Parser(new HtmlTreeBuilder());
+        parser.setTrackErrors(10);
+        assertEquals("\u20ac", parser.parseInput("<html><body>&#0128;</body></html>", "").text());
+        assertEquals(1, parser.getErrors().size());
+    }
+
+    @Test public void cp1252SubstitutionTable() throws UnsupportedEncodingException {
+        for (int i = 0; i < Tokeniser.win1252Extensions.length; i++) {
+            String s = new String(new byte[]{ (byte) (i + Tokeniser.win1252ExtensionsStart) }, "Windows-1252");
+            assertEquals(1, s.length());
+
+            // some of these characters are illegal
+            if (s.charAt(0) == '\ufffd') { continue; }
+
+            assertEquals("At: " + i, s.charAt(0), Tokeniser.win1252Extensions[i]);
+        }
+    }
 }