com.android.dx.command.dexer.Main Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of dalvik-dx Show documentation
Show all versions of dalvik-dx Show documentation
A standalone packaging of AOSP's platform/dalvik dx library.
/*
* Copyright (C) 2007 The Android Open Source Project
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.android.dx.command.dexer;
import com.android.dex.Dex;
import com.android.dex.DexException;
import com.android.dex.DexFormat;
import com.android.dex.util.FileUtils;
import com.android.dx.Version;
import com.android.dx.cf.code.SimException;
import com.android.dx.cf.direct.ClassPathOpener;
import com.android.dx.cf.direct.ClassPathOpener.FileNameFilter;
import com.android.dx.cf.direct.DirectClassFile;
import com.android.dx.cf.direct.StdAttributeFactory;
import com.android.dx.cf.iface.ParseException;
import com.android.dx.command.UsageException;
import com.android.dx.dex.DexOptions;
import com.android.dx.dex.cf.CfOptions;
import com.android.dx.dex.cf.CfTranslator;
import com.android.dx.dex.code.PositionList;
import com.android.dx.dex.file.ClassDefItem;
import com.android.dx.dex.file.DexFile;
import com.android.dx.dex.file.EncodedMethod;
import com.android.dx.merge.CollisionPolicy;
import com.android.dx.merge.DexMerger;
import com.android.dx.rop.annotation.Annotation;
import com.android.dx.rop.annotation.Annotations;
import com.android.dx.rop.annotation.AnnotationsList;
import com.android.dx.rop.code.RegisterSpec;
import com.android.dx.rop.cst.CstNat;
import com.android.dx.rop.cst.CstString;
import com.android.dx.rop.cst.CstType;
import com.android.dx.rop.type.Prototype;
import com.android.dx.rop.type.Type;
import java.io.BufferedReader;
import java.io.ByteArrayInputStream;
import java.io.ByteArrayOutputStream;
import java.io.File;
import java.io.FileOutputStream;
import java.io.FileReader;
import java.io.IOException;
import java.io.OutputStream;
import java.io.OutputStreamWriter;
import java.io.PrintWriter;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collection;
import java.util.HashMap;
import java.util.HashSet;
import java.util.List;
import java.util.Map;
import java.util.Set;
import java.util.TreeMap;
import java.util.concurrent.ArrayBlockingQueue;
import java.util.concurrent.Callable;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;
import java.util.concurrent.ThreadPoolExecutor;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.atomic.AtomicInteger;
import java.util.jar.Attributes;
import java.util.jar.JarEntry;
import java.util.jar.JarOutputStream;
import java.util.jar.Manifest;
/**
* Main class for the class file translator.
*/
public class Main {
/**
* File extension of a {@code .dex} file.
*/
private static final String DEX_EXTENSION = ".dex";
/**
* File name prefix of a {@code .dex} file automatically loaded in an
* archive.
*/
private static final String DEX_PREFIX = "classes";
/**
* {@code non-null;} the lengthy message that tries to discourage
* people from defining core classes in applications
*/
private static final String IN_RE_CORE_CLASSES =
"Ill-advised or mistaken usage of a core class (java.* or javax.*)\n" +
"when not building a core library.\n\n" +
"This is often due to inadvertently including a core library file\n" +
"in your application's project, when using an IDE (such as\n" +
"Eclipse). If you are sure you're not intentionally defining a\n" +
"core class, then this is the most likely explanation of what's\n" +
"going on.\n\n" +
"However, you might actually be trying to define a class in a core\n" +
"namespace, the source of which you may have taken, for example,\n" +
"from a non-Android virtual machine project. This will most\n" +
"assuredly not work. At a minimum, it jeopardizes the\n" +
"compatibility of your app with future versions of the platform.\n" +
"It is also often of questionable legality.\n\n" +
"If you really intend to build a core library -- which is only\n" +
"appropriate as part of creating a full virtual machine\n" +
"distribution, as opposed to compiling an application -- then use\n" +
"the \"--core-library\" option to suppress this error message.\n\n" +
"If you go ahead and use \"--core-library\" but are in fact\n" +
"building an application, then be forewarned that your application\n" +
"will still fail to build or run, at some point. Please be\n" +
"prepared for angry customers who find, for example, that your\n" +
"application ceases to function once they upgrade their operating\n" +
"system. You will be to blame for this problem.\n\n" +
"If you are legitimately using some code that happens to be in a\n" +
"core package, then the easiest safe alternative you have is to\n" +
"repackage that code. That is, move the classes in question into\n" +
"your own package namespace. This means that they will never be in\n" +
"conflict with core system classes. JarJar is a tool that may help\n" +
"you in this endeavor. If you find that you cannot do this, then\n" +
"that is an indication that the path you are on will ultimately\n" +
"lead to pain, suffering, grief, and lamentation.\n";
/**
* {@code non-null;} name of the standard manifest file in {@code .jar}
* files
*/
private static final String MANIFEST_NAME = "META-INF/MANIFEST.MF";
/**
* {@code non-null;} attribute name for the (quasi-standard?)
* {@code Created-By} attribute
*/
private static final Attributes.Name CREATED_BY =
new Attributes.Name("Created-By");
/**
* {@code non-null;} list of {@code javax} subpackages that are considered
* to be "core". Note:: This list must be sorted, since it
* is binary-searched.
*/
private static final String[] JAVAX_CORE = {
"accessibility", "crypto", "imageio", "management", "naming", "net",
"print", "rmi", "security", "sip", "sound", "sql", "swing",
"transaction", "xml"
};
/* Array.newInstance may be added by RopperMachine,
* ArrayIndexOutOfBoundsException. may be added by EscapeAnalysis */
private static final int MAX_METHOD_ADDED_DURING_DEX_CREATION = 2;
/* .TYPE */
private static final int MAX_FIELD_ADDED_DURING_DEX_CREATION = 9;
/** number of errors during processing */
private AtomicInteger errors = new AtomicInteger(0);
/** {@code non-null;} parsed command-line arguments */
private Arguments args;
/** {@code non-null;} output file in-progress */
private DexFile outputDex;
/**
* {@code null-ok;} map of resources to include in the output, or
* {@code null} if resources are being ignored
*/
private TreeMap outputResources;
/** Library .dex files to merge into the output .dex. */
private final List libraryDexBuffers = new ArrayList();
/** Thread pool object used for multi-thread class translation. */
private ExecutorService classTranslatorPool;
/** Single thread executor, for collecting results of parallel translation,
* and adding classes to dex file in original input file order. */
private ExecutorService classDefItemConsumer;
/** Futures for {@code classDefItemConsumer} tasks. */
private List> addToDexFutures =
new ArrayList>();
/** Thread pool object used for multi-thread dex conversion (to byte array).
* Used in combination with multi-dex support, to allow outputing
* a completed dex file, in parallel with continuing processing. */
private ExecutorService dexOutPool;
/** Futures for {@code dexOutPool} task. */
private List> dexOutputFutures = new ArrayList>();
/** Lock object used to to coordinate dex file rotation, and
* multi-threaded translation. */
private Object dexRotationLock = new Object();
/** Record the number if method indices "reserved" for files
* committed to translation in the context of the current dex
* file, but not yet added. */
private int maxMethodIdsInProcess = 0;
/** Record the number if field indices "reserved" for files
* committed to translation in the context of the current dex
* file, but not yet added. */
private int maxFieldIdsInProcess = 0;
/** true if any files are successfully processed */
private volatile boolean anyFilesProcessed;
/** class files older than this must be defined in the target dex file. */
private long minimumFileAge = 0;
private Set classesInMainDex = null;
private List dexOutputArrays = new ArrayList();
private OutputStreamWriter humanOutWriter = null;
private final DxContext context;
public Main(DxContext context) {
this.context = context;
}
/**
* Run and exit if something unexpected happened.
* @param argArray the command line arguments
*/
public static void main(String[] argArray) throws IOException {
DxContext context = new DxContext();
Arguments arguments = new Arguments(context);
arguments.parse(argArray);
int result = new Main(context).runDx(arguments);
if (result != 0) {
System.exit(result);
}
}
public static void clearInternTables() {
Prototype.clearInternTable();
RegisterSpec.clearInternTable();
CstType.clearInternTable();
Type.clearInternTable();
}
/**
* Run and return a result code.
* @param arguments the data + parameters for the conversion
* @return 0 if success > 0 otherwise.
*/
public static int run(Arguments arguments) throws IOException {
return new Main(new DxContext()).runDx(arguments);
}
public int runDx(Arguments arguments) throws IOException {
// Reset the error count to start fresh.
errors.set(0);
// empty the list, so that tools that load dx and keep it around
// for multiple runs don't reuse older buffers.
libraryDexBuffers.clear();
args = arguments;
args.makeOptionsObjects();
OutputStream humanOutRaw = null;
if (args.humanOutName != null) {
humanOutRaw = openOutput(args.humanOutName);
humanOutWriter = new OutputStreamWriter(humanOutRaw);
}
try {
if (args.multiDex) {
return runMultiDex();
} else {
return runMonoDex();
}
} finally {
closeOutput(humanOutRaw);
}
}
private int runMonoDex() throws IOException {
File incrementalOutFile = null;
if (args.incremental) {
if (args.outName == null) {
context.err.println(
"error: no incremental output name specified");
return -1;
}
incrementalOutFile = new File(args.outName);
if (incrementalOutFile.exists()) {
minimumFileAge = incrementalOutFile.lastModified();
}
}
if (!processAllFiles()) {
return 1;
}
if (args.incremental && !anyFilesProcessed) {
return 0; // this was a no-op incremental build
}
// this array is null if no classes were defined
byte[] outArray = null;
if (!outputDex.isEmpty() || (args.humanOutName != null)) {
outArray = writeDex(outputDex);
if (outArray == null) {
return 2;
}
}
if (args.incremental) {
outArray = mergeIncremental(outArray, incrementalOutFile);
}
outArray = mergeLibraryDexBuffers(outArray);
if (args.jarOutput) {
// Effectively free up the (often massive) DexFile memory.
outputDex = null;
if (outArray != null) {
outputResources.put(DexFormat.DEX_IN_JAR_NAME, outArray);
}
if (!createJar(args.outName)) {
return 3;
}
} else if (outArray != null && args.outName != null) {
OutputStream out = openOutput(args.outName);
out.write(outArray);
closeOutput(out);
}
return 0;
}
private int runMultiDex() throws IOException {
assert !args.incremental;
if (args.mainDexListFile != null) {
classesInMainDex = new HashSet();
readPathsFromFile(args.mainDexListFile, classesInMainDex);
}
dexOutPool = Executors.newFixedThreadPool(args.numThreads);
if (!processAllFiles()) {
return 1;
}
if (!libraryDexBuffers.isEmpty()) {
throw new DexException("Library dex files are not supported in multi-dex mode");
}
if (outputDex != null) {
// this array is null if no classes were defined
dexOutputFutures.add(dexOutPool.submit(new DexWriter(outputDex)));
// Effectively free up the (often massive) DexFile memory.
outputDex = null;
}
try {
dexOutPool.shutdown();
if (!dexOutPool.awaitTermination(600L, TimeUnit.SECONDS)) {
throw new RuntimeException("Timed out waiting for dex writer threads.");
}
for (Future f : dexOutputFutures) {
dexOutputArrays.add(f.get());
}
} catch (InterruptedException ex) {
dexOutPool.shutdownNow();
throw new RuntimeException("A dex writer thread has been interrupted.");
} catch (Exception e) {
dexOutPool.shutdownNow();
throw new RuntimeException("Unexpected exception in dex writer thread");
}
if (args.jarOutput) {
for (int i = 0; i < dexOutputArrays.size(); i++) {
outputResources.put(getDexFileName(i),
dexOutputArrays.get(i));
}
if (!createJar(args.outName)) {
return 3;
}
} else if (args.outName != null) {
File outDir = new File(args.outName);
assert outDir.isDirectory();
for (int i = 0; i < dexOutputArrays.size(); i++) {
OutputStream out = new FileOutputStream(new File(outDir, getDexFileName(i)));
try {
out.write(dexOutputArrays.get(i));
} finally {
closeOutput(out);
}
}
}
return 0;
}
private static String getDexFileName(int i) {
if (i == 0) {
return DexFormat.DEX_IN_JAR_NAME;
} else {
return DEX_PREFIX + (i + 1) + DEX_EXTENSION;
}
}
private static void readPathsFromFile(String fileName, Collection paths) throws IOException {
BufferedReader bfr = null;
try {
FileReader fr = new FileReader(fileName);
bfr = new BufferedReader(fr);
String line;
while (null != (line = bfr.readLine())) {
paths.add(fixPath(line));
}
} finally {
if (bfr != null) {
bfr.close();
}
}
}
/**
* Merges the dex files {@code update} and {@code base}, preferring
* {@code update}'s definition for types defined in both dex files.
*
* @param base a file to find the previous dex file. May be a .dex file, a
* jar file possibly containing a .dex file, or null.
* @return the bytes of the merged dex file, or null if both the update
* and the base dex do not exist.
*/
private byte[] mergeIncremental(byte[] update, File base) throws IOException {
Dex dexA = null;
Dex dexB = null;
if (update != null) {
dexA = new Dex(update);
}
if (base.exists()) {
dexB = new Dex(base);
}
Dex result;
if (dexA == null && dexB == null) {
return null;
} else if (dexA == null) {
result = dexB;
} else if (dexB == null) {
result = dexA;
} else {
result = new DexMerger(new Dex[] {dexA, dexB}, CollisionPolicy.KEEP_FIRST, context).merge();
}
ByteArrayOutputStream bytesOut = new ByteArrayOutputStream();
result.writeTo(bytesOut);
return bytesOut.toByteArray();
}
/**
* Merges the dex files in library jars. If multiple dex files define the
* same type, this fails with an exception.
*/
private byte[] mergeLibraryDexBuffers(byte[] outArray) throws IOException {
ArrayList dexes = new ArrayList();
if (outArray != null) {
dexes.add(new Dex(outArray));
}
for (byte[] libraryDex : libraryDexBuffers) {
dexes.add(new Dex(libraryDex));
}
if (dexes.isEmpty()) {
return null;
}
Dex merged = new DexMerger(dexes.toArray(new Dex[dexes.size()]), CollisionPolicy.FAIL, context).merge();
return merged.getBytes();
}
/**
* Constructs the output {@link DexFile}, fill it in with all the
* specified classes, and populate the resources map if required.
*
* @return whether processing was successful
*/
private boolean processAllFiles() {
createDexFile();
if (args.jarOutput) {
outputResources = new TreeMap();
}
anyFilesProcessed = false;
String[] fileNames = args.fileNames;
Arrays.sort(fileNames);
// translate classes in parallel
classTranslatorPool = new ThreadPoolExecutor(args.numThreads,
args.numThreads, 0, TimeUnit.SECONDS,
new ArrayBlockingQueue(2 * args.numThreads, true),
new ThreadPoolExecutor.CallerRunsPolicy());
// collect translated and write to dex in order
classDefItemConsumer = Executors.newSingleThreadExecutor();
try {
if (args.mainDexListFile != null) {
// with --main-dex-list
FileNameFilter mainPassFilter = args.strictNameCheck ? new MainDexListFilter() :
new BestEffortMainDexListFilter();
// forced in main dex
for (int i = 0; i < fileNames.length; i++) {
processOne(fileNames[i], mainPassFilter);
}
if (dexOutputFutures.size() > 0) {
throw new DexException("Too many classes in " + Arguments.MAIN_DEX_LIST_OPTION
+ ", main dex capacity exceeded");
}
if (args.minimalMainDex) {
// start second pass directly in a secondary dex file.
// Wait for classes in progress to complete
synchronized(dexRotationLock) {
while(maxMethodIdsInProcess > 0 || maxFieldIdsInProcess > 0) {
try {
dexRotationLock.wait();
} catch(InterruptedException ex) {
/* ignore */
}
}
}
rotateDexFile();
}
// remaining files
FileNameFilter filter = new RemoveModuleInfoFilter(new NotFilter(mainPassFilter));
for (int i = 0; i < fileNames.length; i++) {
processOne(fileNames[i], filter);
}
} else {
// without --main-dex-list
FileNameFilter filter = new RemoveModuleInfoFilter(ClassPathOpener.acceptAll);
for (int i = 0; i < fileNames.length; i++) {
processOne(fileNames[i], filter);
}
}
} catch (StopProcessing ex) {
/*
* Ignore it and just let the error reporting do
* their things.
*/
}
try {
classTranslatorPool.shutdown();
classTranslatorPool.awaitTermination(600L, TimeUnit.SECONDS);
classDefItemConsumer.shutdown();
classDefItemConsumer.awaitTermination(600L, TimeUnit.SECONDS);
for (Future f : addToDexFutures) {
try {
f.get();
} catch(ExecutionException ex) {
// Catch any previously uncaught exceptions from
// class translation and adding to dex.
int count = errors.incrementAndGet();
if (count < 10) {
if (args.debug) {
context.err.println("Uncaught translation error:");
ex.getCause().printStackTrace(context.err);
} else {
context.err.println("Uncaught translation error: " + ex.getCause());
}
} else {
throw new InterruptedException("Too many errors");
}
}
}
} catch (InterruptedException ie) {
classTranslatorPool.shutdownNow();
classDefItemConsumer.shutdownNow();
throw new RuntimeException("Translation has been interrupted", ie);
} catch (Exception e) {
classTranslatorPool.shutdownNow();
classDefItemConsumer.shutdownNow();
e.printStackTrace(context.out);
throw new RuntimeException("Unexpected exception in translator thread.", e);
}
int errorNum = errors.get();
if (errorNum != 0) {
context.err.println(errorNum + " error" +
((errorNum == 1) ? "" : "s") + "; aborting");
return false;
}
if (args.incremental && !anyFilesProcessed) {
return true;
}
if (!(anyFilesProcessed || args.emptyOk)) {
context.err.println("no classfiles specified");
return false;
}
if (args.optimize && args.statistics) {
context.codeStatistics.dumpStatistics(context.out);
}
return true;
}
private void createDexFile() {
outputDex = new DexFile(args.dexOptions);
if (args.dumpWidth != 0) {
outputDex.setDumpWidth(args.dumpWidth);
}
}
private void rotateDexFile() {
if (outputDex != null) {
if (dexOutPool != null) {
dexOutputFutures.add(dexOutPool.submit(new DexWriter(outputDex)));
} else {
dexOutputArrays.add(writeDex(outputDex));
}
}
createDexFile();
}
/**
* Processes one pathname element.
*
* @param pathname {@code non-null;} the pathname to process. May
* be the path of a class file, a jar file, or a directory
* containing class files.
* @param filter {@code non-null;} A filter for excluding files.
*/
private void processOne(String pathname, FileNameFilter filter) {
ClassPathOpener opener;
opener = new ClassPathOpener(pathname, true, filter, new FileBytesConsumer());
if (opener.process()) {
updateStatus(true);
}
}
private void updateStatus(boolean res) {
anyFilesProcessed |= res;
}
/**
* Processes one file, which may be either a class or a resource.
*
* @param name {@code non-null;} name of the file
* @param bytes {@code non-null;} contents of the file
* @return whether processing was successful
*/
private boolean processFileBytes(String name, long lastModified, byte[] bytes) {
boolean isClass = name.endsWith(".class");
boolean isClassesDex = name.equals(DexFormat.DEX_IN_JAR_NAME);
boolean keepResources = (outputResources != null);
if (!isClass && !isClassesDex && !keepResources) {
if (args.verbose) {
context.out.println("ignored resource " + name);
}
return false;
}
if (args.verbose) {
context.out.println("processing " + name + "...");
}
String fixedName = fixPath(name);
if (isClass) {
if (keepResources && args.keepClassesInJar) {
synchronized (outputResources) {
outputResources.put(fixedName, bytes);
}
}
if (lastModified < minimumFileAge) {
return true;
}
processClass(fixedName, bytes);
// Assume that an exception may occur. Status will be updated
// asynchronously, if the class compiles without error.
return false;
} else if (isClassesDex) {
synchronized (libraryDexBuffers) {
libraryDexBuffers.add(bytes);
}
return true;
} else {
synchronized (outputResources) {
outputResources.put(fixedName, bytes);
}
return true;
}
}
/**
* Processes one classfile.
*
* @param name {@code non-null;} name of the file, clipped such that it
* should correspond to the name of the class it contains
* @param bytes {@code non-null;} contents of the file
* @return whether processing was successful
*/
private boolean processClass(String name, byte[] bytes) {
if (! args.coreLibrary) {
checkClassName(name);
}
try {
new DirectClassFileConsumer(name, bytes, null).call(
new ClassParserTask(name, bytes).call());
} catch (ParseException ex) {
// handled in FileBytesConsumer
throw ex;
} catch(Exception ex) {
throw new RuntimeException("Exception parsing classes", ex);
}
return true;
}
private DirectClassFile parseClass(String name, byte[] bytes) {
DirectClassFile cf = new DirectClassFile(bytes, name,
args.cfOptions.strictNameCheck);
cf.setAttributeFactory(StdAttributeFactory.THE_ONE);
cf.getMagic(); // triggers the actual parsing
return cf;
}
private ClassDefItem translateClass(byte[] bytes, DirectClassFile cf) {
try {
return CfTranslator.translate(context, cf, bytes, args.cfOptions,
args.dexOptions, outputDex);
} catch (ParseException ex) {
context.err.println("\ntrouble processing:");
if (args.debug) {
ex.printStackTrace(context.err);
} else {
ex.printContext(context.err);
}
}
errors.incrementAndGet();
return null;
}
private boolean addClassToDex(ClassDefItem clazz) {
synchronized (outputDex) {
outputDex.add(clazz);
}
return true;
}
/**
* Check the class name to make sure it's not a "core library"
* class. If there is a problem, this updates the error count and
* throws an exception to stop processing.
*
* @param name {@code non-null;} the fully-qualified internal-form
* class name
*/
private void checkClassName(String name) {
boolean bogus = false;
if (name.startsWith("java/")) {
bogus = true;
} else if (name.startsWith("javax/")) {
int slashAt = name.indexOf('/', 6);
if (slashAt == -1) {
// Top-level javax classes are verboten.
bogus = true;
} else {
String pkg = name.substring(6, slashAt);
bogus = (Arrays.binarySearch(JAVAX_CORE, pkg) >= 0);
}
}
if (! bogus) {
return;
}
/*
* The user is probably trying to include an entire desktop
* core library in a misguided attempt to get their application
* working. Try to help them understand what's happening.
*/
context.err.println("\ntrouble processing \"" + name + "\":\n\n" +
IN_RE_CORE_CLASSES);
errors.incrementAndGet();
throw new StopProcessing();
}
/**
* Converts {@link #outputDex} into a {@code byte[]} and do whatever
* human-oriented dumping is required.
*
* @return {@code null-ok;} the converted {@code byte[]} or {@code null}
* if there was a problem
*/
private byte[] writeDex(DexFile outputDex) {
byte[] outArray = null;
try {
try {
if (args.methodToDump != null) {
/*
* Simply dump the requested method. Note: The call
* to toDex() is required just to get the underlying
* structures ready.
*/
outputDex.toDex(null, false);
dumpMethod(outputDex, args.methodToDump, humanOutWriter);
} else {
/*
* This is the usual case: Create an output .dex file,
* and write it, dump it, etc.
*/
outArray = outputDex.toDex(humanOutWriter, args.verboseDump);
}
if (args.statistics) {
context.out.println(outputDex.getStatistics().toHuman());
}
} finally {
if (humanOutWriter != null) {
humanOutWriter.flush();
}
}
} catch (Exception ex) {
if (args.debug) {
context.err.println("\ntrouble writing output:");
ex.printStackTrace(context.err);
} else {
context.err.println("\ntrouble writing output: " +
ex.getMessage());
}
return null;
}
return outArray;
}
/**
* Creates a jar file from the resources (including dex file arrays).
*
* @param fileName {@code non-null;} name of the file
* @return whether the creation was successful
*/
private boolean createJar(String fileName) {
/*
* Make or modify the manifest (as appropriate), put the dex
* array into the resources map, and then process the entire
* resources map in a uniform manner.
*/
try {
Manifest manifest = makeManifest();
OutputStream out = openOutput(fileName);
JarOutputStream jarOut = new JarOutputStream(out, manifest);
try {
for (Map.Entry e :
outputResources.entrySet()) {
String name = e.getKey();
byte[] contents = e.getValue();
JarEntry entry = new JarEntry(name);
int length = contents.length;
if (args.verbose) {
context.out.println("writing " + name + "; size " + length + "...");
}
entry.setSize(length);
jarOut.putNextEntry(entry);
jarOut.write(contents);
jarOut.closeEntry();
}
} finally {
jarOut.finish();
jarOut.flush();
closeOutput(out);
}
} catch (Exception ex) {
if (args.debug) {
context.err.println("\ntrouble writing output:");
ex.printStackTrace(context.err);
} else {
context.err.println("\ntrouble writing output: " +
ex.getMessage());
}
return false;
}
return true;
}
/**
* Creates and returns the manifest to use for the output. This may
* modify {@link #outputResources} (removing the pre-existing manifest).
*
* @return {@code non-null;} the manifest
*/
private Manifest makeManifest() throws IOException {
byte[] manifestBytes = outputResources.get(MANIFEST_NAME);
Manifest manifest;
Attributes attribs;
if (manifestBytes == null) {
// We need to construct an entirely new manifest.
manifest = new Manifest();
attribs = manifest.getMainAttributes();
attribs.put(Attributes.Name.MANIFEST_VERSION, "1.0");
} else {
manifest = new Manifest(new ByteArrayInputStream(manifestBytes));
attribs = manifest.getMainAttributes();
outputResources.remove(MANIFEST_NAME);
}
String createdBy = attribs.getValue(CREATED_BY);
if (createdBy == null) {
createdBy = "";
} else {
createdBy += " + ";
}
createdBy += "dx " + Version.VERSION;
attribs.put(CREATED_BY, createdBy);
attribs.putValue("Dex-Location", DexFormat.DEX_IN_JAR_NAME);
return manifest;
}
/**
* Opens and returns the named file for writing, treating "-" specially.
*
* @param name {@code non-null;} the file name
* @return {@code non-null;} the opened file
*/
private OutputStream openOutput(String name) throws IOException {
if (name.equals("-") ||
name.startsWith("-.")) {
return context.out;
}
return new FileOutputStream(name);
}
/**
* Flushes and closes the given output stream, except if it happens to be
* {@link System#out} in which case this method does the flush but not
* the close. This method will also silently do nothing if given a
* {@code null} argument.
*
* @param stream {@code null-ok;} what to close
*/
private void closeOutput(OutputStream stream) throws IOException {
if (stream == null) {
return;
}
stream.flush();
if (stream != context.out) {
stream.close();
}
}
/**
* Returns the "fixed" version of a given file path, suitable for
* use as a path within a {@code .jar} file and for checking
* against a classfile-internal "this class" name. This looks for
* the last instance of the substring {@code "/./"} within
* the path, and if it finds it, it takes the portion after to be
* the fixed path. If that isn't found but the path starts with
* {@code "./"}, then that prefix is removed and the rest is
* return. If neither of these is the case, this method returns
* its argument.
*
* @param path {@code non-null;} the path to "fix"
* @return {@code non-null;} the fixed version (which might be the same as
* the given {@code path})
*/
private static String fixPath(String path) {
/*
* If the path separator is \ (like on windows), we convert the
* path to a standard '/' separated path.
*/
if (File.separatorChar == '\\') {
path = path.replace('\\', '/');
}
int index = path.lastIndexOf("/./");
if (index != -1) {
return path.substring(index + 3);
}
if (path.startsWith("./")) {
return path.substring(2);
}
return path;
}
/**
* Dumps any method with the given name in the given file.
*
* @param dex {@code non-null;} the dex file
* @param fqName {@code non-null;} the fully-qualified name of the
* method(s)
* @param out {@code non-null;} where to dump to
*/
private void dumpMethod(DexFile dex, String fqName,
OutputStreamWriter out) {
boolean wildcard = fqName.endsWith("*");
int lastDot = fqName.lastIndexOf('.');
if ((lastDot <= 0) || (lastDot == (fqName.length() - 1))) {
context.err.println("bogus fully-qualified method name: " +
fqName);
return;
}
String className = fqName.substring(0, lastDot).replace('.', '/');
String methodName = fqName.substring(lastDot + 1);
ClassDefItem clazz = dex.getClassOrNull(className);
if (clazz == null) {
context.err.println("no such class: " + className);
return;
}
if (wildcard) {
methodName = methodName.substring(0, methodName.length() - 1);
}
ArrayList allMeths = clazz.getMethods();
TreeMap meths =
new TreeMap();
/*
* Figure out which methods to include in the output, and get them
* all sorted, so that the printout code is robust with respect to
* changes in the underlying order.
*/
for (EncodedMethod meth : allMeths) {
String methName = meth.getName().getString();
if ((wildcard && methName.startsWith(methodName)) ||
(!wildcard && methName.equals(methodName))) {
meths.put(meth.getRef().getNat(), meth);
}
}
if (meths.size() == 0) {
context.err.println("no such method: " + fqName);
return;
}
PrintWriter pw = new PrintWriter(out);
for (EncodedMethod meth : meths.values()) {
// TODO: Better stuff goes here, perhaps.
meth.debugPrint(pw, args.verboseDump);
/*
* The (default) source file is an attribute of the class, but
* it's useful to see it in method dumps.
*/
CstString sourceFile = clazz.getSourceFile();
if (sourceFile != null) {
pw.println(" source file: " + sourceFile.toQuoted());
}
Annotations methodAnnotations =
clazz.getMethodAnnotations(meth.getRef());
AnnotationsList parameterAnnotations =
clazz.getParameterAnnotations(meth.getRef());
if (methodAnnotations != null) {
pw.println(" method annotations:");
for (Annotation a : methodAnnotations.getAnnotations()) {
pw.println(" " + a);
}
}
if (parameterAnnotations != null) {
pw.println(" parameter annotations:");
int sz = parameterAnnotations.size();
for (int i = 0; i < sz; i++) {
pw.println(" parameter " + i);
Annotations annotations = parameterAnnotations.get(i);
for (Annotation a : annotations.getAnnotations()) {
pw.println(" " + a);
}
}
}
}
pw.flush();
}
private static class NotFilter implements FileNameFilter {
private final FileNameFilter filter;
private NotFilter(FileNameFilter filter) {
this.filter = filter;
}
@Override
public boolean accept(String path) {
return !filter.accept(path);
}
}
/**
* Filters "module-info.class" out of the paths accepted by delegate.
*/
private static class RemoveModuleInfoFilter implements FileNameFilter {
protected final FileNameFilter delegate;
public RemoveModuleInfoFilter(FileNameFilter delegate) {
this.delegate = delegate;
}
@Override
public boolean accept(String path) {
return delegate.accept(path) && !("module-info.class".equals(path));
}
}
/**
* A quick and accurate filter for when file path can be trusted.
*/
private class MainDexListFilter implements FileNameFilter {
@Override
public boolean accept(String fullPath) {
if (fullPath.endsWith(".class")) {
String path = fixPath(fullPath);
return classesInMainDex.contains(path);
} else {
return true;
}
}
}
/**
* A best effort conservative filter for when file path can not be trusted.
*/
private class BestEffortMainDexListFilter implements FileNameFilter {
Map> map = new HashMap>();
public BestEffortMainDexListFilter() {
for (String pathOfClass : classesInMainDex) {
String normalized = fixPath(pathOfClass);
String simple = getSimpleName(normalized);
List fullPath = map.get(simple);
if (fullPath == null) {
fullPath = new ArrayList(1);
map.put(simple, fullPath);
}
fullPath.add(normalized);
}
}
@Override
public boolean accept(String path) {
if (path.endsWith(".class")) {
String normalized = fixPath(path);
String simple = getSimpleName(normalized);
List fullPaths = map.get(simple);
if (fullPaths != null) {
for (String fullPath : fullPaths) {
if (normalized.endsWith(fullPath)) {
return true;
}
}
}
return false;
} else {
return true;
}
}
private String getSimpleName(String path) {
int index = path.lastIndexOf('/');
if (index >= 0) {
return path.substring(index + 1);
} else {
return path;
}
}
}
/**
* Exception class used to halt processing prematurely.
*/
private static class StopProcessing extends RuntimeException {
// This space intentionally left blank.
}
/**
* Command-line argument parser and access.
*/
public static class Arguments {
private static final String MINIMAL_MAIN_DEX_OPTION = "--minimal-main-dex";
private static final String MAIN_DEX_LIST_OPTION = "--main-dex-list";
private static final String MULTI_DEX_OPTION = "--multi-dex";
private static final String NUM_THREADS_OPTION = "--num-threads";
private static final String INCREMENTAL_OPTION = "--incremental";
private static final String INPUT_LIST_OPTION = "--input-list";
public final DxContext context;
/** whether to run in debug mode */
public boolean debug = false;
/** whether to emit warning messages */
public boolean warnings = true;
/** whether to emit high-level verbose human-oriented output */
public boolean verbose = false;
/** whether to emit verbose human-oriented output in the dump file */
public boolean verboseDump = false;
/** whether we are constructing a core library */
public boolean coreLibrary = false;
/** {@code null-ok;} particular method to dump */
public String methodToDump = null;
/** max width for columnar output */
public int dumpWidth = 0;
/** {@code null-ok;} output file name for binary file */
public String outName = null;
/** {@code null-ok;} output file name for human-oriented dump */
public String humanOutName = null;
/** whether strict file-name-vs-class-name checking should be done */
public boolean strictNameCheck = true;
/**
* whether it is okay for there to be no {@code .class} files
* to process
*/
public boolean emptyOk = false;
/**
* whether the binary output is to be a {@code .jar} file
* instead of a plain {@code .dex}
*/
public boolean jarOutput = false;
/**
* when writing a {@code .jar} file, whether to still
* keep the {@code .class} files
*/
public boolean keepClassesInJar = false;
/** what API level to target */
public int minSdkVersion = DexFormat.API_NO_EXTENDED_OPCODES;
/** how much source position info to preserve */
public int positionInfo = PositionList.LINES;
/** whether to keep local variable information */
public boolean localInfo = true;
/** whether to merge with the output dex file if it exists. */
public boolean incremental = false;
/** whether to force generation of const-string/jumbo for all indexes,
* to allow merges between dex files with many strings. */
public boolean forceJumbo = false;
/** whether default and static interface methods can be invoked at any API level. */
public boolean allowAllInterfaceMethodInvokes = false;
/** {@code non-null} after {@link #parse}; file name arguments */
public String[] fileNames;
/** whether to do SSA/register optimization */
public boolean optimize = true;
/** Filename containg list of methods to optimize */
public String optimizeListFile = null;
/** Filename containing list of methods to NOT optimize */
public String dontOptimizeListFile = null;
/** Whether to print statistics to stdout at end of compile cycle */
public boolean statistics;
/** Options for class file transformation */
public CfOptions cfOptions;
/** Options for dex file output */
public DexOptions dexOptions;
/** number of threads to run with */
public int numThreads = 1;
/** generation of multiple dex is allowed */
public boolean multiDex = false;
/** Optional file containing a list of class files containing classes to be forced in main
* dex */
public String mainDexListFile = null;
/** Produce the smallest possible main dex. Ignored unless multiDex is true and
* mainDexListFile is specified and non empty. */
public boolean minimalMainDex = false;
public int maxNumberOfIdxPerDex = DexFormat.MAX_MEMBER_IDX + 1;
/** Optional list containing inputs read in from a file. */
private List inputList = null;
private boolean outputIsDirectory = false;
private boolean outputIsDirectDex = false;
public Arguments(DxContext context) {
this.context = context;
}
public Arguments() {
this(new DxContext());
}
private static class ArgumentsParser {
/** The arguments to process. */
private final String[] arguments;
/** The index of the next argument to process. */
private int index;
/** The current argument being processed after a {@link #getNext()} call. */
private String current;
/** The last value of an argument processed by {@link #isArg(String)}. */
private String lastValue;
public ArgumentsParser(String[] arguments) {
this.arguments = arguments;
index = 0;
}
public String getCurrent() {
return current;
}
public String getLastValue() {
return lastValue;
}
/**
* Moves on to the next argument.
* Returns false when we ran out of arguments that start with --.
*/
public boolean getNext() {
if (index >= arguments.length) {
return false;
}
current = arguments[index];
if (current.equals("--") || !current.startsWith("--")) {
return false;
}
index++;
return true;
}
/**
* Similar to {@link #getNext()}, this moves on the to next argument.
* It does not check however whether the argument starts with --
* and thus can be used to retrieve values.
*/
private boolean getNextValue() {
if (index >= arguments.length) {
return false;
}
current = arguments[index];
index++;
return true;
}
/**
* Returns all the arguments that have not been processed yet.
*/
public String[] getRemaining() {
int n = arguments.length - index;
String[] remaining = new String[n];
if (n > 0) {
System.arraycopy(arguments, index, remaining, 0, n);
}
return remaining;
}
/**
* Checks the current argument against the given prefix.
* If prefix is in the form '--name=', an extra value is expected.
* The argument can then be in the form '--name=value' or as a 2-argument
* form '--name value'.
*/
public boolean isArg(String prefix) {
int n = prefix.length();
if (n > 0 && prefix.charAt(n-1) == '=') {
// Argument accepts a value. Capture it.
if (current.startsWith(prefix)) {
// Argument is in the form --name=value, split the value out
lastValue = current.substring(n);
return true;
} else {
// Check whether we have "--name value" as 2 arguments
prefix = prefix.substring(0, n-1);
if (current.equals(prefix)) {
if (getNextValue()) {
lastValue = current;
return true;
} else {
System.err.println("Missing value after parameter " + prefix);
throw new UsageException();
}
}
return false;
}
} else {
// Argument does not accept a value.
return current.equals(prefix);
}
}
}
private void parseFlags(ArgumentsParser parser) {
while(parser.getNext()) {
if (parser.isArg("--debug")) {
debug = true;
} else if (parser.isArg("--no-warning")) {
warnings = false;
} else if (parser.isArg("--verbose")) {
verbose = true;
} else if (parser.isArg("--verbose-dump")) {
verboseDump = true;
} else if (parser.isArg("--no-files")) {
emptyOk = true;
} else if (parser.isArg("--no-optimize")) {
optimize = false;
} else if (parser.isArg("--no-strict")) {
strictNameCheck = false;
} else if (parser.isArg("--core-library")) {
coreLibrary = true;
} else if (parser.isArg("--statistics")) {
statistics = true;
} else if (parser.isArg("--optimize-list=")) {
if (dontOptimizeListFile != null) {
context.err.println("--optimize-list and "
+ "--no-optimize-list are incompatible.");
throw new UsageException();
}
optimize = true;
optimizeListFile = parser.getLastValue();
} else if (parser.isArg("--no-optimize-list=")) {
if (dontOptimizeListFile != null) {
context.err.println("--optimize-list and "
+ "--no-optimize-list are incompatible.");
throw new UsageException();
}
optimize = true;
dontOptimizeListFile = parser.getLastValue();
} else if (parser.isArg("--keep-classes")) {
keepClassesInJar = true;
} else if (parser.isArg("--output=")) {
outName = parser.getLastValue();
if (new File(outName).isDirectory()) {
jarOutput = false;
outputIsDirectory = true;
} else if (FileUtils.hasArchiveSuffix(outName)) {
jarOutput = true;
} else if (outName.endsWith(".dex") ||
outName.equals("-")) {
jarOutput = false;
outputIsDirectDex = true;
} else {
context.err.println("unknown output extension: " +
outName);
throw new UsageException();
}
} else if (parser.isArg("--dump-to=")) {
humanOutName = parser.getLastValue();
} else if (parser.isArg("--dump-width=")) {
dumpWidth = Integer.parseInt(parser.getLastValue());
} else if (parser.isArg("--dump-method=")) {
methodToDump = parser.getLastValue();
jarOutput = false;
} else if (parser.isArg("--positions=")) {
String pstr = parser.getLastValue().intern();
if (pstr == "none") {
positionInfo = PositionList.NONE;
} else if (pstr == "important") {
positionInfo = PositionList.IMPORTANT;
} else if (pstr == "lines") {
positionInfo = PositionList.LINES;
} else {
context.err.println("unknown positions option: " +
pstr);
throw new UsageException();
}
} else if (parser.isArg("--no-locals")) {
localInfo = false;
} else if (parser.isArg(NUM_THREADS_OPTION + "=")) {
numThreads = Integer.parseInt(parser.getLastValue());
} else if (parser.isArg(INCREMENTAL_OPTION)) {
incremental = true;
} else if (parser.isArg("--force-jumbo")) {
forceJumbo = true;
} else if (parser.isArg(MULTI_DEX_OPTION)) {
multiDex = true;
} else if (parser.isArg(MAIN_DEX_LIST_OPTION + "=")) {
mainDexListFile = parser.getLastValue();
} else if (parser.isArg(MINIMAL_MAIN_DEX_OPTION)) {
minimalMainDex = true;
} else if (parser.isArg("--set-max-idx-number=")) { // undocumented test option
maxNumberOfIdxPerDex = Integer.parseInt(parser.getLastValue());
} else if(parser.isArg(INPUT_LIST_OPTION + "=")) {
File inputListFile = new File(parser.getLastValue());
try {
inputList = new ArrayList();
readPathsFromFile(inputListFile.getAbsolutePath(), inputList);
} catch (IOException e) {
context.err.println(
"Unable to read input list file: " + inputListFile.getName());
// problem reading the file so we should halt execution
throw new UsageException();
}
} else if (parser.isArg("--min-sdk-version=")) {
String arg = parser.getLastValue();
int value;
try {
value = Integer.parseInt(arg);
} catch (NumberFormatException ex) {
value = -1;
}
if (value < 1) {
System.err.println("improper min-sdk-version option: " + arg);
throw new UsageException();
}
minSdkVersion = value;
} else if (parser.isArg("--allow-all-interface-method-invokes")) {
allowAllInterfaceMethodInvokes = true;
} else {
context.err.println("unknown option: " + parser.getCurrent());
throw new UsageException();
}
}
}
/**
* Parses all command-line arguments and updates the state of the {@code Arguments} object
* accordingly.
*
* @param args {@code non-null;} the arguments
*/
private void parse(String[] args) {
ArgumentsParser parser = new ArgumentsParser(args);
parseFlags(parser);
fileNames = parser.getRemaining();
if(inputList != null && !inputList.isEmpty()) {
// append the file names to the end of the input list
inputList.addAll(Arrays.asList(fileNames));
fileNames = inputList.toArray(new String[inputList.size()]);
}
if (fileNames.length == 0) {
if (!emptyOk) {
context.err.println("no input files specified");
throw new UsageException();
}
} else if (emptyOk) {
context.out.println("ignoring input files");
}
if ((humanOutName == null) && (methodToDump != null)) {
humanOutName = "-";
}
if (mainDexListFile != null && !multiDex) {
context.err.println(MAIN_DEX_LIST_OPTION + " is only supported in combination with "
+ MULTI_DEX_OPTION);
throw new UsageException();
}
if (minimalMainDex && (mainDexListFile == null || !multiDex)) {
context.err.println(MINIMAL_MAIN_DEX_OPTION + " is only supported in combination with "
+ MULTI_DEX_OPTION + " and " + MAIN_DEX_LIST_OPTION);
throw new UsageException();
}
if (multiDex && incremental) {
context.err.println(INCREMENTAL_OPTION + " is not supported with "
+ MULTI_DEX_OPTION);
throw new UsageException();
}
if (multiDex && outputIsDirectDex) {
context.err.println("Unsupported output \"" + outName +"\". " + MULTI_DEX_OPTION +
" supports only archive or directory output");
throw new UsageException();
}
if (outputIsDirectory && !multiDex) {
outName = new File(outName, DexFormat.DEX_IN_JAR_NAME).getPath();
}
makeOptionsObjects();
}
/**
* Parses only command-line flags and updates the state of the {@code Arguments} object
* accordingly.
*
* @param flags {@code non-null;} the flags
*/
public void parseFlags(String[] flags) {
parseFlags(new ArgumentsParser(flags));
}
/**
* Copies relevant arguments over into CfOptions and DexOptions instances.
*/
public void makeOptionsObjects() {
cfOptions = new CfOptions();
cfOptions.positionInfo = positionInfo;
cfOptions.localInfo = localInfo;
cfOptions.strictNameCheck = strictNameCheck;
cfOptions.optimize = optimize;
cfOptions.optimizeListFile = optimizeListFile;
cfOptions.dontOptimizeListFile = dontOptimizeListFile;
cfOptions.statistics = statistics;
if (warnings) {
cfOptions.warn = context.err;
} else {
cfOptions.warn = context.noop;
}
dexOptions = new DexOptions(context.err);
dexOptions.minSdkVersion = minSdkVersion;
dexOptions.forceJumbo = forceJumbo;
dexOptions.allowAllInterfaceMethodInvokes = allowAllInterfaceMethodInvokes;
}
}
/**
* Callback class for processing input file bytes, produced by the
* ClassPathOpener.
*/
private class FileBytesConsumer implements ClassPathOpener.Consumer {
@Override
public boolean processFileBytes(String name, long lastModified,
byte[] bytes) {
return Main.this.processFileBytes(name, lastModified, bytes);
}
@Override
public void onException(Exception ex) {
if (ex instanceof StopProcessing) {
throw (StopProcessing) ex;
} else if (ex instanceof SimException) {
context.err.println("\nEXCEPTION FROM SIMULATION:");
context.err.println(ex.getMessage() + "\n");
context.err.println(((SimException) ex).getContext());
} else if (ex instanceof ParseException) {
context.err.println("\nPARSE ERROR:");
ParseException parseException = (ParseException) ex;
if (args.debug) {
parseException.printStackTrace(context.err);
} else {
parseException.printContext(context.err);
}
} else {
context.err.println("\nUNEXPECTED TOP-LEVEL EXCEPTION:");
ex.printStackTrace(context.err);
}
errors.incrementAndGet();
}
@Override
public void onProcessArchiveStart(File file) {
if (args.verbose) {
context.out.println("processing archive " + file + "...");
}
}
}
/** Callable helper class to parse class bytes. */
private class ClassParserTask implements Callable {
String name;
byte[] bytes;
private ClassParserTask(String name, byte[] bytes) {
this.name = name;
this.bytes = bytes;
}
@Override
public DirectClassFile call() throws Exception {
DirectClassFile cf = parseClass(name, bytes);
return cf;
}
}
/**
* Callable helper class used to sequentially collect the results of
* the (optionally parallel) translation phase, in correct input file order.
* This class is also responsible for coordinating dex file rotation
* with the ClassDefItemConsumer class.
* We maintain invariant that the number of indices used in the current
* dex file plus the max number of indices required by classes passed to
* the translation phase and not yet added to the dex file, is less than
* or equal to the dex file limit.
* For each parsed file, we estimate the maximum number of indices it may
* require. If passing the file to the translation phase would invalidate
* the invariant, we wait, until the next class is added to the dex file,
* and then reevaluate the invariant. If there are no further classes in
* the translation phase, we rotate the dex file.
*/
private class DirectClassFileConsumer implements Callable {
String name;
byte[] bytes;
Future dcff;
private DirectClassFileConsumer(String name, byte[] bytes,
Future dcff) {
this.name = name;
this.bytes = bytes;
this.dcff = dcff;
}
@Override
public Boolean call() throws Exception {
DirectClassFile cf = dcff.get();
return call(cf);
}
private Boolean call(DirectClassFile cf) {
int maxMethodIdsInClass = 0;
int maxFieldIdsInClass = 0;
if (args.multiDex) {
// Calculate max number of indices this class will add to the
// dex file.
// The possibility of overloading means that we can't easily
// know how many constant are needed for declared methods and
// fields. We therefore make the simplifying assumption that
// all constants are external method or field references.
int constantPoolSize = cf.getConstantPool().size();
maxMethodIdsInClass = constantPoolSize + cf.getMethods().size()
+ MAX_METHOD_ADDED_DURING_DEX_CREATION;
maxFieldIdsInClass = constantPoolSize + cf.getFields().size()
+ MAX_FIELD_ADDED_DURING_DEX_CREATION;
synchronized(dexRotationLock) {
int numMethodIds;
int numFieldIds;
// Number of indices used in current dex file.
synchronized(outputDex) {
numMethodIds = outputDex.getMethodIds().items().size();
numFieldIds = outputDex.getFieldIds().items().size();
}
// Wait until we're sure this class will fit in the current
// dex file.
while(((numMethodIds + maxMethodIdsInClass + maxMethodIdsInProcess
> args.maxNumberOfIdxPerDex) ||
(numFieldIds + maxFieldIdsInClass + maxFieldIdsInProcess
> args.maxNumberOfIdxPerDex))) {
if (maxMethodIdsInProcess > 0 || maxFieldIdsInProcess > 0) {
// There are classes in the translation phase that
// have not yet been added to the dex file, so we
// wait for the next class to complete.
try {
dexRotationLock.wait();
} catch(InterruptedException ex) {
/* ignore */
}
} else if (outputDex.getClassDefs().items().size() > 0) {
// There are no further classes in the translation
// phase, and we have a full dex file. Rotate!
rotateDexFile();
} else {
// The estimated number of indices is too large for
// an empty dex file. We proceed hoping the actual
// number of indices needed will fit.
break;
}
synchronized(outputDex) {
numMethodIds = outputDex.getMethodIds().items().size();
numFieldIds = outputDex.getFieldIds().items().size();
}
}
// Add our estimate to the total estimate for
// classes under translation.
maxMethodIdsInProcess += maxMethodIdsInClass;
maxFieldIdsInProcess += maxFieldIdsInClass;
}
}
// Submit class to translation phase.
Future cdif = classTranslatorPool.submit(
new ClassTranslatorTask(name, bytes, cf));
Future res = classDefItemConsumer.submit(new ClassDefItemConsumer(
name, cdif, maxMethodIdsInClass, maxFieldIdsInClass));
addToDexFutures.add(res);
return true;
}
}
/** Callable helper class to translate classes in parallel */
private class ClassTranslatorTask implements Callable {
String name;
byte[] bytes;
DirectClassFile classFile;
private ClassTranslatorTask(String name, byte[] bytes,
DirectClassFile classFile) {
this.name = name;
this.bytes = bytes;
this.classFile = classFile;
}
@Override
public ClassDefItem call() {
ClassDefItem clazz = translateClass(bytes, classFile);
return clazz;
}
}
/**
* Callable helper class used to collect the results of
* the parallel translation phase, adding the translated classes to
* the current dex file in correct (deterministic) file order.
* This class is also responsible for coordinating dex file rotation
* with the DirectClassFileConsumer class.
*/
private class ClassDefItemConsumer implements Callable {
String name;
Future futureClazz;
int maxMethodIdsInClass;
int maxFieldIdsInClass;
private ClassDefItemConsumer(String name, Future futureClazz,
int maxMethodIdsInClass, int maxFieldIdsInClass) {
this.name = name;
this.futureClazz = futureClazz;
this.maxMethodIdsInClass = maxMethodIdsInClass;
this.maxFieldIdsInClass = maxFieldIdsInClass;
}
@Override
public Boolean call() throws Exception {
try {
ClassDefItem clazz = futureClazz.get();
if (clazz != null) {
addClassToDex(clazz);
updateStatus(true);
}
return true;
} catch(ExecutionException ex) {
// Rethrow previously uncaught translation exceptions.
// These, as well as any exceptions from addClassToDex,
// are handled and reported in processAllFiles().
Throwable t = ex.getCause();
throw (t instanceof Exception) ? (Exception) t : ex;
} finally {
if (args.multiDex) {
// Having added our actual indicies to the dex file,
// we subtract our original estimate from the total estimate,
// and signal the translation phase, which may be paused
// waiting to determine if more classes can be added to the
// current dex file, or if a new dex file must be created.
synchronized(dexRotationLock) {
maxMethodIdsInProcess -= maxMethodIdsInClass;
maxFieldIdsInProcess -= maxFieldIdsInClass;
dexRotationLock.notifyAll();
}
}
}
}
}
/** Callable helper class to convert dex files in worker threads */
private class DexWriter implements Callable {
private final DexFile dexFile;
private DexWriter(DexFile dexFile) {
this.dexFile = dexFile;
}
@Override
public byte[] call() throws IOException {
return writeDex(dexFile);
}
}
}