提交 53be92c6 编写于 作者: S Skylot

core: fix decoding UTF-8 strings in xml resources

上级 5f8f454b
......@@ -36,13 +36,13 @@ public enum ResourceType {
case CODE:
case ARSC:
case LIB:
case XML:
case FONT:
case IMG:
case UNKNOWN:
return false;
case MANIFEST:
case XML:
return true;
}
return false;
......
......@@ -33,10 +33,33 @@ public class BinaryXMLParser {
private static final Logger LOG = LoggerFactory.getLogger(BinaryXMLParser.class);
private static final Charset STRING_CHARSET = Charset.forName("UTF-16LE");
private static final Charset STRING_CHARSET_UTF16 = Charset.forName("UTF-16LE");
private static final Charset STRING_CHARSET_UTF8 = Charset.forName("UTF-8");
private static final int RES_NULL_TYPE = 0x0000;
private static final int RES_STRING_POOL_TYPE = 0x0001;
private static final int RES_TABLE_TYPE = 0x0002;
private static final int RES_XML_TYPE = 0x0003;
private static final int RES_XML_FIRST_CHUNK_TYPE = 0x0100;
private static final int RES_XML_START_NAMESPACE_TYPE = 0x0100;
private static final int RES_XML_END_NAMESPACE_TYPE = 0x0101;
private static final int RES_XML_START_ELEMENT_TYPE = 0x0102;
private static final int RES_XML_END_ELEMENT_TYPE = 0x0103;
private static final int RES_XML_CDATA_TYPE = 0x0104;
private static final int RES_XML_LAST_CHUNK_TYPE = 0x017f;
private static final int RES_XML_RESOURCE_MAP_TYPE = 0x0180;
private static final int RES_TABLE_PACKAGE_TYPE = 0x0200;
private static final int RES_TABLE_TYPE_TYPE = 0x0201;
private static final int RES_TABLE_TYPE_SPEC_TYPE = 0x0202;
// string pool flags
private static final int SORTED_FLAG = 1;
private static final int UTF8_FLAG = 1 << 8;
private CodeWriter writer;
private InputStream input;
private ParserStream is;
private String[] strings;
private String nsPrefix = "ERROR";
......@@ -76,7 +99,7 @@ public class BinaryXMLParser {
public synchronized CodeWriter parse(InputStream inputStream) {
writer = new CodeWriter();
writer.add("<?xml version=\"1.0\" encoding=\"utf-8\"?>");
input = inputStream;
is = new ParserStream(inputStream);
firstElement = true;
try {
decode();
......@@ -92,37 +115,37 @@ public class BinaryXMLParser {
}
void decode() throws IOException {
if (cInt16() != 0x0003) {
if (is.readInt16() != 0x0003) {
die("Version is not 3");
}
if (cInt16() != 0x0008) {
if (is.readInt16() != 0x0008) {
die("Size of header is not 8");
}
cInt32();
while (input.available() != 0) {
int type = cInt16();
int size = is.readInt32();
while (is.getPos() < size) {
int type = is.readInt16();
switch (type) {
case 0x0001:
case RES_NULL_TYPE:
// NullType is just doing nothing
break;
case RES_STRING_POOL_TYPE:
parseStringPool();
break;
case 0x0180:
case RES_XML_RESOURCE_MAP_TYPE:
parseResourceMap();
break;
case 0x0100:
case RES_XML_START_NAMESPACE_TYPE:
parseNameSpace();
break;
case 0x0101:
case RES_XML_END_NAMESPACE_TYPE:
parseNameSpaceEnd();
break;
case 0x0102:
case RES_XML_START_ELEMENT_TYPE:
parseElement();
break;
case 0x0103:
case RES_XML_END_ELEMENT_TYPE:
parseElementEnd();
break;
case 0x0000:
// NullType is just doing nothing
break;
default:
die("Type: " + Integer.toHexString(type) + " not yet implemented");
......@@ -132,68 +155,88 @@ public class BinaryXMLParser {
}
private void parseStringPool() throws IOException {
if (cInt16() != 0x001c) {
if (is.readInt16() != 0x001c) {
die("Header header size not 28");
}
int hsize = cInt32();
int stringCount = cInt32();
int styleCount = cInt32();
int flags = cInt32();
int stringsStart = cInt32();
int stylesStart = cInt32();
int[] stringsOffsets = new int[stringCount];
for (int i = 0; i < stringCount; i++) {
stringsOffsets[i] = cInt32();
}
int hsize = is.readInt32();
int stringCount = is.readInt32();
int styleCount = is.readInt32();
int flags = is.readInt32();
int stringsStart = is.readInt32();
int stylesStart = is.readInt32();
// skip string offsets
is.skip(stringCount * 4);
strings = new String[stringCount];
for (int i = 0; i < stringCount; i++) {
int off = 8 + stringsStart + stringsOffsets[i];
int strlen = cInt16();
byte[] str = new byte[strlen * 2];
readToArray(str);
strings[i] = new String(str, STRING_CHARSET);
cInt16();
if ((flags & UTF8_FLAG) != 0) {
// UTF-8
long start = is.getPos();
for (int i = 0; i < stringCount; i++) {
int charsCount = is.decodeLength8();
int len = is.decodeLength8();
strings[i] = new String(is.readArray(len), STRING_CHARSET_UTF8);
int zero = is.readInt8();
if (zero != 0) {
die("Not a trailing zero at string end: " + zero + ", " + strings[i]);
}
}
long shift = is.getPos() - start;
if (shift % 2 != 0) {
is.skip(1);
}
} else {
// UTF-16
for (int i = 0; i < stringCount; i++) {
int len = is.decodeLength16();
strings[i] = new String(is.readArray(len * 2), STRING_CHARSET_UTF16);
int zero = is.readInt16();
if (zero != 0) {
die("Not a trailing zero at string end: " + zero + ", " + strings[i]);
}
}
}
if (styleCount != 0) {
die("Styles parsing in string pool not yet implemented");
}
}
private void parseResourceMap() throws IOException {
if (cInt16() != 0x8) {
if (is.readInt16() != 0x8) {
die("Header size of resmap is not 8!");
}
int rhsize = cInt32();
int rhsize = is.readInt32();
int[] ids = new int[(rhsize - 8) / 4];
for (int i = 0; i < ids.length; i++) {
ids[i] = cInt32();
ids[i] = is.readInt32();
}
}
private void parseNameSpace() throws IOException {
if (cInt16() != 0x10) {
if (is.readInt16() != 0x10) {
die("NAMESPACE header is not 0x0010");
}
if (cInt32() != 0x18) {
if (is.readInt32() != 0x18) {
die("NAMESPACE header chunk is not 0x18 big");
}
int beginLineNumber = cInt32();
int comment = cInt32();
int beginPrefix = cInt32();
int beginLineNumber = is.readInt32();
int comment = is.readInt32();
int beginPrefix = is.readInt32();
nsPrefix = strings[beginPrefix];
int beginURI = cInt32();
int beginURI = is.readInt32();
nsURI = strings[beginURI];
}
private void parseNameSpaceEnd() throws IOException {
if (cInt16() != 0x10) {
if (is.readInt16() != 0x10) {
die("NAMESPACE header is not 0x0010");
}
if (cInt32() != 0x18) {
if (is.readInt32() != 0x18) {
die("NAMESPACE header chunk is not 0x18 big");
}
int endLineNumber = cInt32();
int comment = cInt32();
int endPrefix = cInt32();
int endLineNumber = is.readInt32();
int comment = is.readInt32();
int endPrefix = is.readInt32();
nsPrefix = strings[endPrefix];
int endURI = cInt32();
int endURI = is.readInt32();
nsURI = strings[endURI];
}
......@@ -203,15 +246,15 @@ public class BinaryXMLParser {
} else {
writer.incIndent();
}
if (cInt16() != 0x10) {
if (is.readInt16() != 0x10) {
die("ELEMENT HEADER SIZE is not 0x10");
}
// TODO: Check element chunk size
cInt32();
int elementBegLineNumber = cInt32();
int comment = cInt32();
int startNS = cInt32();
int startNSName = cInt32(); // actually is elementName...
is.readInt32();
int elementBegLineNumber = is.readInt32();
int comment = is.readInt32();
int startNS = is.readInt32();
int startNSName = is.readInt32(); // actually is elementName...
if (!wasOneLiner && !"ERROR".equals(currentTag) && !currentTag.equals(strings[startNSName])) {
writer.add(">");
}
......@@ -219,18 +262,18 @@ public class BinaryXMLParser {
currentTag = strings[startNSName];
writer.startLine("<").add(strings[startNSName]);
writer.attachSourceLine(elementBegLineNumber);
int attributeStart = cInt16();
int attributeStart = is.readInt16();
if (attributeStart != 0x14) {
die("startNS's attributeStart is not 0x14");
}
int attributeSize = cInt16();
int attributeSize = is.readInt16();
if (attributeSize != 0x14) {
die("startNS's attributeSize is not 0x14");
}
int attributeCount = cInt16();
int idIndex = cInt16();
int classIndex = cInt16();
int styleIndex = cInt16();
int attributeCount = is.readInt16();
int idIndex = is.readInt16();
int classIndex = is.readInt16();
int styleIndex = is.readInt16();
if ("manifest".equals(strings[startNSName])) {
writer.add(" xmlns:\"").add(nsURI).add("\"");
}
......@@ -247,18 +290,18 @@ public class BinaryXMLParser {
}
private void parseAttribute(int i) throws IOException {
int attributeNS = cInt32();
int attributeName = cInt32();
int attributeRawValue = cInt32();
int attrValSize = cInt16();
int attributeNS = is.readInt32();
int attributeName = is.readInt32();
int attributeRawValue = is.readInt32();
int attrValSize = is.readInt16();
if (attrValSize != 0x08) {
die("attrValSize != 0x08 not supported");
}
if (cInt8() != 0) {
if (is.readInt8() != 0) {
die("res0 is not 0");
}
int attrValDataType = cInt8();
int attrValData = cInt32();
int attrValDataType = is.readInt8();
int attrValData = is.readInt32();
if (attributeNS != -1) {
writer.add(nsPrefix).add(':');
}
......@@ -319,16 +362,16 @@ public class BinaryXMLParser {
}
private void parseElementEnd() throws IOException {
if (cInt16() != 0x10) {
if (is.readInt16() != 0x10) {
die("ELEMENT END header is not 0x10");
}
if (cInt32() != 0x18) {
if (is.readInt32() != 0x18) {
die("ELEMENT END header chunk is not 0x18 big");
}
int endLineNumber = cInt32();
int comment = cInt32();
int elementNS = cInt32();
int elementName = cInt32();
int endLineNumber = is.readInt32();
int comment = is.readInt32();
int elementNS = is.readInt32();
int elementName = is.readInt32();
if (currentTag.equals(strings[elementName])) {
writer.add(" />");
wasOneLiner = true;
......@@ -345,38 +388,8 @@ public class BinaryXMLParser {
}
}
private int cInt8() throws IOException {
return input.read();
}
private int cInt16() throws IOException {
int b1 = input.read();
int b2 = input.read();
return (b2 & 0xFF) << 8 | (b1 & 0xFF);
}
private int cInt32() throws IOException {
InputStream in = input;
int b1 = in.read();
int b2 = in.read();
int b3 = in.read();
int b4 = in.read();
return b4 << 24 | (b3 & 0xFF) << 16 | (b2 & 0xFF) << 8 | (b1 & 0xFF);
}
private void readToArray(byte[] arr) throws IOException {
int count = arr.length;
int pos = input.read(arr, 0, count);
while (pos < count) {
int read = input.read(arr, pos, count - pos);
if (read == -1) {
throw new IOException("No data, can't read " + count + " bytes");
}
pos += read;
}
}
private void die(String message) {
throw new JadxRuntimeException("Decode error: " + message);
throw new JadxRuntimeException("Decode error: " + message
+ ", position: 0x" + Long.toHexString(is.getPos()));
}
}
package jadx.core.xmlgen;
import java.io.IOException;
import java.io.InputStream;
public class ParserStream {
private final InputStream input;
private long readPos = 0;
public ParserStream(InputStream inputStream) {
this.input = inputStream;
}
public long getPos() {
return readPos;
}
public int readInt8() throws IOException {
readPos++;
return input.read();
}
public int readInt16() throws IOException {
readPos += 2;
int b1 = input.read();
int b2 = input.read();
return (b2 & 0xFF) << 8 | (b1 & 0xFF);
}
public int readInt32() throws IOException {
readPos += 4;
InputStream in = input;
int b1 = in.read();
int b2 = in.read();
int b3 = in.read();
int b4 = in.read();
return b4 << 24 | (b3 & 0xFF) << 16 | (b2 & 0xFF) << 8 | (b1 & 0xFF);
}
public byte[] readArray(int count) throws IOException {
readPos += count;
byte[] arr = new byte[count];
int pos = input.read(arr, 0, count);
while (pos < count) {
int read = input.read(arr, pos, count - pos);
if (read == -1) {
throw new IOException("No data, can't read " + count + " bytes");
}
pos += read;
}
return arr;
}
public void skip(long count) throws IOException {
readPos += count;
long pos = input.skip(count);
while (pos < count) {
long skipped = input.skip(count - pos);
if (skipped == -1) {
throw new IOException("No data, can't skip " + count + " bytes");
}
pos += skipped;
}
}
public int decodeLength8() throws IOException {
int len = readInt8();
if ((len & 0x80) != 0) {
len = ((len & 0x7F) << 8) | readInt8();
}
return len;
}
public int decodeLength16() throws IOException {
int len = readInt16();
if ((len & 0x8000) != 0) {
len = ((len & 0x7FFF) << 16) | readInt16();
}
return len;
}
}
......@@ -144,7 +144,6 @@ public class JResource extends JNode implements Comparable<JResource> {
private boolean isSupportedForView(ResourceType type) {
switch (type) {
case CODE:
case XML:
case ARSC:
case FONT:
case IMG:
......@@ -152,6 +151,7 @@ public class JResource extends JNode implements Comparable<JResource> {
return false;
case MANIFEST:
case XML:
case UNKNOWN:
return true;
}
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册