Kaladaran/briar
0
0
Fork 0
mirror of https://code.briarproject.org/briar/briar.git synced 2026-02-16 04:39:54 +01:00
Code Issues Projects Releases Wiki Activity

Reject invalid UTF-8 instead of ignoring it.

Browse Source
This commit is contained in:
akwizgran
2023-02-18 17:10:57 +00:00
parent 10b1fe756d
commit 5a09530670
3 changed files with 52 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
bramble-core/src/test/java/org/briarproject/bramble/data/BdfReaderImplFuzzingTest.java
View File

@@ -1,5 +1,6 @@
package org.briarproject.bramble.data;
import org.briarproject.bramble.api.FormatException;
import org.briarproject.bramble.test.BrambleTestCase;
import org.junit.Before;
import org.junit.Test;
@@ -32,10 +33,13 @@ public class BdfReaderImplFuzzingTest extends BrambleTestCase {
in.reset();
BdfReaderImpl r = new BdfReaderImpl(in, DEFAULT_NESTED_LIMIT,
DEFAULT_MAX_BUFFER_SIZE);
int length = r.readString().length();
assertTrue(length >= 0);
assertTrue(length <= 20);
assertTrue(r.eof());
try {
int length = r.readString().length();
assertTrue(length <= 20);
assertTrue(r.eof());
} catch (FormatException e) {
// Expected when bytes are not valid UTF-8
}
}
}
}

41
bramble-core/src/test/java/org/briarproject/bramble/util/StringUtilsTest.java
View File

@@ -88,51 +88,58 @@ public class StringUtilsTest extends BrambleTestCase {
}
@Test
public void testFromUtf8AcceptsNullCharacterUsingStandardUtf8() {
public void testFromUtf8AcceptsNullCharacterUsingStandardUtf8()
throws Exception {
// The UTF-8 encoding of the null character is valid
assertEquals("\u0000", StringUtils.fromUtf8(new byte[1]));
byte[] utf8 = new byte[1];
String actual = StringUtils.fromUtf8(utf8);
assertEquals("\u0000", actual);
// When we convert back to UTF-8 we should get the original encoding
assertArrayEquals(utf8, StringUtils.toUtf8(actual));
}
@Test
public void testFromUtf8RemovesNullCharacterUsingModifiedUtf8() {
@Test(expected = FormatException.class)
public void testFromUtf8RejectsNullCharacterUsingModifiedUtf8()
throws Exception {
// The modified UTF-8 encoding of the null character is not valid
byte[] b = new byte[] {
(byte) 0xC0, (byte) 0x80, // Null character as modified UTF-8
(byte) 0xC8, (byte) 0x85 // U+0205
};
// Conversion should ignore the invalid character and return the rest
String expected = "\u0205";
assertEquals(expected, StringUtils.fromUtf8(b));
StringUtils.fromUtf8(b);
}
@Test
public void testFromUtf8AcceptsSupplementaryCharacterUsingStandardUtf8() {
public void testFromUtf8AcceptsSupplementaryCharacterUsingStandardUtf8()
throws Exception {
// The UTF-8 encoding of a supplementary character is valid and should
// be converted to a surrogate pair
byte[] b = new byte[] {
byte[] utf8 = new byte[] {
(byte) 0xF0, (byte) 0x90, (byte) 0x90, (byte) 0x80, // U+10400
(byte) 0xC8, (byte) 0x85 // U+0205
};
String expected = "\uD801\uDC00\u0205"; // Surrogate pair
assertEquals(expected, StringUtils.fromUtf8(b));
String actual = StringUtils.fromUtf8(utf8);
assertEquals(expected, actual);
// When we convert back to UTF-8 we should get the original encoding
assertArrayEquals(utf8, StringUtils.toUtf8(actual));
}
@Test
public void testFromUtf8RemovesSupplementaryCharacterUsingModifiedUtf8() {
@Test(expected = FormatException.class)
public void testFromUtf8RejectsSupplementaryCharacterUsingModifiedUtf8()
throws Exception {
// The CESU-8 or modified UTF-8 encoding of a supplementary character
// is not valid
byte[] b = new byte[] {
byte[] utf8 = new byte[] {
(byte) 0xED, (byte) 0xA0, (byte) 0x81, // U+10400 as CSEU-8
(byte) 0xED, (byte) 0xB0, (byte) 0x80,
(byte) 0xC8, (byte) 0x85 // U+0205
};
// Conversion should ignore the invalid character and return the rest
String expected = "\u0205";
assertEquals(expected, StringUtils.fromUtf8(b));
StringUtils.fromUtf8(utf8);
}
@Test
public void testFromUtf8EmptyInput() {
public void testFromUtf8EmptyInput() throws Exception {
assertEquals("", StringUtils.fromUtf8(new byte[0]));
}