package org.jivesoftware.openfire.nio;
import static org.junit.Assert.assertFalse;
import static org.junit.Assert.assertTrue;
import org.junit.Test;
/**
* This unit tests verifies the correct detection of numeric character references that have invalid numeric values.
*
* From the XML 1.0 specificaton: <quote><b>Character Reference</b> <code><pre>CharRef ::= '&#' [0-9]+ ';'
* | '&#x' [0-9a-fA-F]+ ';' [WFC: Legal Character]</pre></code>
* <p/>
* <b>Well-formedness constraint: Legal Character<b> Characters referred to using character references must match the
* production for Char.
*
* (...)
* <code><pre>Char ::= #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD] | [#x10000-#x10FFFF]</pre></code> any
* Unicode character, excluding the surrogate blocks, FFFE, and FFFF. <quote>
*
* This test is based on three large sets of values.
* <ol>
* <li>A set containing only illegal numeric character values only;</li>
* <li>A set containing legal numeric character reference values only;</li>
* <li>A set containing values that are no character references at all (but do resemble them)</li>
* </ol>
*
* The first and second set consist of lines in which each line contains both the decimal as well as the hexidecimal
* representation of the same numeric character reference. The remainder of the line is filled with zero-padded copies
* of the same values. The various values (on each new line) are picked from the edges of each of the ranges that make
* up the complete set of valid characters.
*
* @author Guus der Kinderen, guus.der.kinderen@gmail.com
*
* @see http://www.w3.org/TR/2008/REC-xml-20081126/#dt-charref
*/
public class XmlNumericCharacterReferenceTest {
final String[] illegalNumericCharacterReferences = new String[] {
"�" , "�", "�" , "�", "�" , "�", "�" , "�", "�" , "�", "�" , "�",
"" , "", "" , "", "" , "", "" , "", "" , "", "" , "",
"" , "", "" , "", "" , "", "" , "", "" , "", "" , "",
"" , "", "" , "", "" , "", "" , "", "" , "", "" , "",
"" , "", "" , "", "" , "", "" , "", "" , "", "" , "",
"" , "", "" , "", "" , "", "" , "", "" , "", "" , "",
"" , "", "" , "", "" , "", "" , "", "" , "", "" , "",
"" , "", "" , "", "" , "", "" , "", "" , "", "" , "",
"" , "", "" , "", "" , "", "" , "", "" , "", "" , "",
" ", " ", " ", " ", " ", " ", " ", " ", " ", " ", " ", " ",
"", "", "", "", "", "", "", "", "", "", "", "",
"", "", "", "", "", "", "", "", "", "", "", "",
"", "", "", "", "", "", "", "", "", "", "", "",
" ", " ", " ", " ", " ", " ", " ", " ", " ", " ", " ", " ",
"�", "�", "�", "�", "�", "�", "�", "�", "�", "�",
"�", "�", "�", "�", "�", "�", "�", "�", "�", "�",
"", "", "", "", "", "", "", "", "", "",
"", "", "", "", "", "", "", "", "", "",
"�", "�", "�", "�", "�", "�", "�", "�"
};
final String[] legalNumericCharacterReferences = new String[] {
" " , " ", " " , " ", " " , " ", " " , " ", " " , " ", " " , " ",
"
", "
", "
", "
", "
", "
", "
", "
", "
", "
", "
", "
",
"
", "
", "
", "
", "
", "
", "
", "
", "
", "
", "
", "
",
" ", " ", " ", " ", " ", " ", " ", " ", " ", " ", " ", " ",
"!", "!", "!", "!", "!", "!", "!", "!", "!", "!", "!", "!",
"", "", "", "", "", "", "", "", "", "",
"", "", "", "", "", "", "", "", "", "",
"", "", "", "", "", "", "", "", "", "",
"", "", "", "", "", "", "", "", "", "",
"", "", "", "", "", "", "", "", "", "",
"�", "�", "�", "�", "�", "�", "�", "�", "�", "�",
"𐀀", "𐀀", "𐀀", "𐀀", "𐀀", "𐀀", "𐀀", "𐀀",
"𐀁", "𐀁", "𐀁", "𐀁", "𐀁", "𐀁", "𐀁", "𐀁",
"", "", "", "", "", "", "", "",
"", "", "", "", "", "", "", ""
};
final String[] notNumericCharacterRefrences = new String[] {
"&",
" ", " ", "#9;", "#x9;", "", "", "#1;", "#x1;",
"	", "	", "", ""
};
/**
* Iterates of the collection of legal numeric character references and asserts that
* {@link XMLLightweightParser#hasIllegalCharacterReferences(String)} does not find an illegal reference when each
* of these values are passed as an argument to this method.
*/
@Test
public void testLegalNumericCharacterReferences() throws Exception {
for(final String reference : legalNumericCharacterReferences) {
// do magic
final boolean result = XMLLightweightParser.hasIllegalCharacterReferences(reference);
// verify
assertFalse("Value \""+reference+"\" is reported to contain an illegal numeric character reference, even though this hard-coded test value should not contain one.", result);
}
}
/**
* Iterates of the collection of illegal numeric character references and asserts that
* {@link XMLLightweightParser#hasIllegalCharacterReferences(String)} finds an illegal reference for each of these
* values passed as an argument to this method.
*/
@Test
public void testIllegalNumericCharacterReferences() throws Exception {
for(final String reference : illegalNumericCharacterReferences) {
// do magic
final boolean result = XMLLightweightParser.hasIllegalCharacterReferences(reference);
// verify
assertTrue("No illegal numeric character reference was found in value \""+reference+"\", even though this hard-coded test value should contain one.", result);
}
}
/**
* Iterates of the collection of values that are not numeric character references and asserts that
* {@link XMLLightweightParser#hasIllegalCharacterReferences(String)} does not find an illegal reference when each
* of these values are passed as an argument to this method.
*/
@Test
public void testnotNumericCharacterReferences() throws Exception {
for(final String reference : legalNumericCharacterReferences) {
// do magic
final boolean result = XMLLightweightParser.hasIllegalCharacterReferences(reference);
// verify
assertFalse("Value \""+reference+"\" is reported to contain an illegal numeric character reference, even though this hard-coded test value should not contain a numeric character reference at all.", result);
}
}
/**
* Checks if {@link XMLLightweightParser#hasIllegalCharacterReferences(String)} correctly skips over a legal numeric
* reference if it is embedded in a snippet of text.
*/
@Test
public void testTextWithLegalNumericCharacterReferences() throws Exception {
// setup
final String text = "The value is a legal numeric character reference.";
// do magic
final boolean result = XMLLightweightParser.hasIllegalCharacterReferences(text);
// verify
assertFalse(result);
}
/**
* Checks if {@link XMLLightweightParser#hasIllegalCharacterReferences(String)} correctly identifies an illegal
* numeric reference if it is embedded in a snippet of text.
*/
@Test
public void testTextWithIllegalNumericCharacterReferences() throws Exception {
// setup
final String text = "The value is an illegal numeric character reference.";
// do magic
final boolean result = XMLLightweightParser.hasIllegalCharacterReferences(text);
// verify
assertTrue(result);
}
/**
* Checks if {@link XMLLightweightParser#hasIllegalCharacterReferences(String)} correctly identifies an illegal
* numeric reference if it is embedded in a snippet of text that also contains a legal reference.
*/
@Test
public void testTextWithBothLegalAndIllegalNumericCharacterReferences() throws Exception {
// setup
final String text = "The value is a legal numeric character reference, but the value is not.";
// do magic
final boolean result = XMLLightweightParser.hasIllegalCharacterReferences(text);
// verify
assertTrue(result);
}
}