summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--luni/src/test/java/libcore/xml/DomTest.java31
-rw-r--r--luni/src/test/java/libcore/xml/PullParserTest.java27
-rw-r--r--xml/src/main/java/org/kxml2/io/KXmlParser.java169
3 files changed, 146 insertions, 81 deletions
diff --git a/luni/src/test/java/libcore/xml/DomTest.java b/luni/src/test/java/libcore/xml/DomTest.java
index 4b97c06..f27a1a4 100644
--- a/luni/src/test/java/libcore/xml/DomTest.java
+++ b/luni/src/test/java/libcore/xml/DomTest.java
@@ -17,6 +17,7 @@
package libcore.xml;
import dalvik.annotation.KnownFailure;
+import java.io.ByteArrayInputStream;
import java.io.File;
import java.io.FileWriter;
import java.io.IOException;
@@ -1656,6 +1657,36 @@ public class DomTest extends TestCase {
assertNull(text.getNextSibling());
}
+ public void testBomAndByteInput() throws Exception {
+ byte[] xml = {
+ (byte) 0xef, (byte) 0xbb, (byte) 0xbf,
+ '<', 'i', 'n', 'p', 'u', 't', '/', '>'
+ };
+ document = builder.parse(new InputSource(new ByteArrayInputStream(xml)));
+ assertEquals("input", document.getDocumentElement().getNodeName());
+ }
+
+ public void testBomAndByteInputWithExplicitCharset() throws Exception {
+ byte[] xml = {
+ (byte) 0xef, (byte) 0xbb, (byte) 0xbf,
+ '<', 'i', 'n', 'p', 'u', 't', '/', '>'
+ };
+ InputSource inputSource = new InputSource(new ByteArrayInputStream(xml));
+ inputSource.setEncoding("UTF-8");
+ document = builder.parse(inputSource);
+ assertEquals("input", document.getDocumentElement().getNodeName());
+ }
+
+ public void testBomAndCharacterInput() throws Exception {
+ InputSource inputSource = new InputSource(new StringReader("\ufeff<input/>"));
+ inputSource.setEncoding("UTF-8");
+ try {
+ builder.parse(inputSource);
+ fail();
+ } catch (SAXException expected) {
+ }
+ }
+
private class RecordingHandler implements UserDataHandler {
final Set<String> calls = new HashSet<String>();
public void handle(short operation, String key, Object data, Node src, Node dst) {
diff --git a/luni/src/test/java/libcore/xml/PullParserTest.java b/luni/src/test/java/libcore/xml/PullParserTest.java
index 06a40fd..fa00fe5 100644
--- a/luni/src/test/java/libcore/xml/PullParserTest.java
+++ b/luni/src/test/java/libcore/xml/PullParserTest.java
@@ -17,7 +17,6 @@
package libcore.xml;
import java.io.ByteArrayInputStream;
-import java.io.IOException;
import java.io.StringReader;
import junit.framework.TestCase;
import org.xmlpull.v1.XmlPullParser;
@@ -695,6 +694,32 @@ public abstract class PullParserTest extends TestCase {
assertParseFailure("not xml");
}
+ public void testBomAndByteInput() throws Exception {
+ byte[] xml = "\ufeff<?xml version='1.0'?><input/>".getBytes("UTF-8");
+ XmlPullParser parser = newPullParser();
+ parser.setInput(new ByteArrayInputStream(xml), null);
+ assertEquals(XmlPullParser.START_TAG, parser.next());
+ assertEquals("input", parser.getName());
+ assertEquals(XmlPullParser.END_TAG, parser.next());
+ assertEquals("input", parser.getName());
+ assertEquals(XmlPullParser.END_DOCUMENT, parser.next());
+ }
+
+ public void testBomAndByteInputWithExplicitCharset() throws Exception {
+ byte[] xml = "\ufeff<?xml version='1.0'?><input/>".getBytes("UTF-8");
+ XmlPullParser parser = newPullParser();
+ parser.setInput(new ByteArrayInputStream(xml), "UTF-8");
+ assertEquals(XmlPullParser.START_TAG, parser.next());
+ assertEquals("input", parser.getName());
+ assertEquals(XmlPullParser.END_TAG, parser.next());
+ assertEquals("input", parser.getName());
+ assertEquals(XmlPullParser.END_DOCUMENT, parser.next());
+ }
+
+ public void testBomAndCharacterInput() throws Exception {
+ assertParseFailure("\ufeff<?xml version='1.0'?><input/>");
+ }
+
private void assertParseFailure(String xml) throws Exception {
XmlPullParser parser = newPullParser();
parser.setInput(new StringReader(xml));
diff --git a/xml/src/main/java/org/kxml2/io/KXmlParser.java b/xml/src/main/java/org/kxml2/io/KXmlParser.java
index 7a2d052..2be8d36 100644
--- a/xml/src/main/java/org/kxml2/io/KXmlParser.java
+++ b/xml/src/main/java/org/kxml2/io/KXmlParser.java
@@ -1606,17 +1606,17 @@ public class KXmlParser implements XmlPullParser, Closeable {
documentEntities = null;
}
- public void setInput(InputStream is, String _enc) throws XmlPullParserException {
+ public void setInput(InputStream is, String charset) throws XmlPullParserException {
position = 0;
limit = 0;
- String enc = _enc;
+ boolean detectCharset = (charset == null);
if (is == null) {
throw new IllegalArgumentException();
}
try {
- if (enc == null) {
+ if (detectCharset) {
// read the four bytes looking for an indication of the encoding in use
int firstFourBytes = 0;
while (limit < 4) {
@@ -1630,93 +1630,102 @@ public class KXmlParser implements XmlPullParser, Closeable {
if (limit == 4) {
switch (firstFourBytes) {
- case 0x00000FEFF: // UTF-32BE BOM
- enc = "UTF-32BE";
- limit = 0;
- break;
-
- case 0x0FFFE0000: // UTF-32LE BOM
- enc = "UTF-32LE";
- limit = 0;
- break;
-
- case 0x0000003c: // '>' in UTF-32BE
- enc = "UTF-32BE";
- buffer[0] = '<';
- limit = 1;
- break;
+ case 0x00000FEFF: // UTF-32BE BOM
+ charset = "UTF-32BE";
+ limit = 0;
+ break;
- case 0x03c000000: // '<' in UTF-32LE
- enc = "UTF-32LE";
- buffer[0] = '<';
- limit = 1;
- break;
-
- case 0x0003c003f: // "<?" in UTF-16BE
- enc = "UTF-16BE";
- buffer[0] = '<';
- buffer[1] = '?';
- limit = 2;
- break;
-
- case 0x03c003f00: // "<?" in UTF-16LE
- enc = "UTF-16LE";
- buffer[0] = '<';
- buffer[1] = '?';
- limit = 2;
- break;
-
- case 0x03c3f786d: // "<?xm" in ASCII etc.
- while (true) {
- int i = is.read();
- if (i == -1) {
- break;
- }
- buffer[limit++] = (char) i;
- if (i == '>') {
- String s = new String(buffer, 0, limit);
- int i0 = s.indexOf("encoding");
- if (i0 != -1) {
- while (s.charAt(i0) != '"'
- && s.charAt(i0) != '\'') {
- i0++;
- }
- char deli = s.charAt(i0++);
- int i1 = s.indexOf(deli, i0);
- enc = s.substring(i0, i1);
+ case 0x0FFFE0000: // UTF-32LE BOM
+ charset = "UTF-32LE";
+ limit = 0;
+ break;
+
+ case 0x0000003c: // '<' in UTF-32BE
+ charset = "UTF-32BE";
+ buffer[0] = '<';
+ limit = 1;
+ break;
+
+ case 0x03c000000: // '<' in UTF-32LE
+ charset = "UTF-32LE";
+ buffer[0] = '<';
+ limit = 1;
+ break;
+
+ case 0x0003c003f: // "<?" in UTF-16BE
+ charset = "UTF-16BE";
+ buffer[0] = '<';
+ buffer[1] = '?';
+ limit = 2;
+ break;
+
+ case 0x03c003f00: // "<?" in UTF-16LE
+ charset = "UTF-16LE";
+ buffer[0] = '<';
+ buffer[1] = '?';
+ limit = 2;
+ break;
+
+ case 0x03c3f786d: // "<?xm" in ASCII etc.
+ while (true) {
+ int i = is.read();
+ if (i == -1) {
+ break;
+ }
+ buffer[limit++] = (char) i;
+ if (i == '>') {
+ String s = new String(buffer, 0, limit);
+ int i0 = s.indexOf("encoding");
+ if (i0 != -1) {
+ while (s.charAt(i0) != '"' && s.charAt(i0) != '\'') {
+ i0++;
}
- break;
+ char deli = s.charAt(i0++);
+ int i1 = s.indexOf(deli, i0);
+ charset = s.substring(i0, i1);
}
+ break;
}
- break;
-
- default:
- // handle a byte order mark followed by something other than <?
- if ((firstFourBytes & 0x0ffff0000) == 0x0FEFF0000) {
- enc = "UTF-16BE";
- buffer[0] = (char) ((buffer[2] << 8) | buffer[3]);
- limit = 1;
- } else if ((firstFourBytes & 0x0ffff0000) == 0x0fffe0000) {
- enc = "UTF-16LE";
- buffer[0] = (char) ((buffer[3] << 8) | buffer[2]);
- limit = 1;
- } else if ((firstFourBytes & 0x0ffffff00) == 0x0EFBBBF00) {
- enc = "UTF-8";
- buffer[0] = buffer[3];
- limit = 1;
- }
+ }
+ break;
+
+ default:
+ // handle a byte order mark followed by something other than <?
+ if ((firstFourBytes & 0x0ffff0000) == 0x0feff0000) {
+ charset = "UTF-16BE";
+ buffer[0] = (char) ((buffer[2] << 8) | buffer[3]);
+ limit = 1;
+ } else if ((firstFourBytes & 0x0ffff0000) == 0x0fffe0000) {
+ charset = "UTF-16LE";
+ buffer[0] = (char) ((buffer[3] << 8) | buffer[2]);
+ limit = 1;
+ } else if ((firstFourBytes & 0x0ffffff00) == 0x0efbbbf00) {
+ charset = "UTF-8";
+ buffer[0] = buffer[3];
+ limit = 1;
+ }
}
}
}
- if (enc == null) {
- enc = "UTF-8";
+ if (charset == null) {
+ charset = "UTF-8";
}
- int sc = limit;
- setInput(new InputStreamReader(is, enc));
- encoding = _enc;
- limit = sc;
+ int savedLimit = limit;
+ setInput(new InputStreamReader(is, charset));
+ encoding = charset;
+ limit = savedLimit;
+
+ /*
+ * Skip the optional BOM if we didn't above. This decrements limit
+ * rather than incrementing position so that <?xml version='1.0'?>
+ * is still at character 0.
+ */
+ if (!detectCharset && peekCharacter() == 0xfeff) {
+ limit--;
+ System.arraycopy(buffer, 1, buffer, 0, limit);
+ }
} catch (Exception e) {
throw new XmlPullParserException("Invalid stream or encoding: " + e, this, e);
}
generated by cgit v1.1 at 2025-02-09 06:43:42 +0000