mirror of
https://github.com/eclipse-cdt/cdt
synced 2025-07-24 01:15:29 +02:00
More efficient way of reading large files, bug 263210.
This commit is contained in:
parent
e2908ac0e2
commit
251635be7d
2 changed files with 217 additions and 156 deletions
|
@ -1,14 +1,15 @@
|
|||
/*******************************************************************************
|
||||
* Copyright (c) 2005, 2008 IBM Corporation and others.
|
||||
* Copyright (c) 2005, 2009 IBM Corporation and others.
|
||||
* All rights reserved. This program and the accompanying materials
|
||||
* are made available under the terms of the Eclipse Public License v1.0
|
||||
* which accompanies this distribution, and is available at
|
||||
* http://www.eclipse.org/legal/epl-v10.html
|
||||
*
|
||||
* Contributors:
|
||||
* IBM Rational Software - Initial API and implementation
|
||||
* Cheong, Jeong-Sik - fix for 162381
|
||||
* Valeri Atamaniouk - fix for 170398
|
||||
* John Camelon (IBM Rational Software) - Initial API and implementation
|
||||
* Cheong, Jeong-Sik - fix for 162381
|
||||
* Valeri Atamaniouk - fix for 170398
|
||||
* Markus Schorn (Wind River Systems)
|
||||
*******************************************************************************/
|
||||
package org.eclipse.cdt.core.parser;
|
||||
|
||||
|
@ -17,32 +18,33 @@ import java.io.IOException;
|
|||
import java.io.InputStream;
|
||||
import java.nio.ByteBuffer;
|
||||
import java.nio.CharBuffer;
|
||||
import java.nio.MappedByteBuffer;
|
||||
import java.nio.channels.FileChannel;
|
||||
import java.nio.charset.Charset;
|
||||
import java.nio.charset.CharsetDecoder;
|
||||
import java.nio.charset.CoderResult;
|
||||
import java.nio.charset.CodingErrorAction;
|
||||
|
||||
import org.eclipse.cdt.core.parser.util.CharArrayUtils;
|
||||
|
||||
/**
|
||||
* @author jcamelon
|
||||
* Reads the content of a file into a char[] buffer.
|
||||
*
|
||||
* @noextend This class is not intended to be subclassed by clients.
|
||||
*/
|
||||
public class CodeReader {
|
||||
public static final String SYSTEM_DEFAULT_ENCODING = System.getProperty( "file.encoding" ); //$NON-NLS-1$
|
||||
public static final String SYSTEM_DEFAULT_ENCODING = System.getProperty("file.encoding"); //$NON-NLS-1$
|
||||
|
||||
private static final int MB = 1024*1024;
|
||||
private static final String NF = "<text>"; //$NON-NLS-1$
|
||||
private static final char [] NOFILE = NF.toCharArray();
|
||||
|
||||
/**
|
||||
* Maximum number of bytes from the source file to load. If the file length
|
||||
* exceeds this value, the content is truncated
|
||||
*
|
||||
* @see #load(String, FileInputStream)
|
||||
*/
|
||||
private static final int MAX_FILE_SIZE = Integer.MAX_VALUE;
|
||||
|
||||
private static final char[] NOFILE = NF.toCharArray();
|
||||
private static final int MAX_FILE_SIZE;
|
||||
static {
|
||||
MAX_FILE_SIZE = (int) Math.min(Integer.MAX_VALUE, (Runtime.getRuntime().maxMemory()) / 4);
|
||||
}
|
||||
public final char[] buffer;
|
||||
public final char[] filename;
|
||||
|
||||
// If you already have preloaded the buffer, e.g. working copy
|
||||
|
||||
// If you already have the buffer, e.g. working copy
|
||||
public CodeReader(String filename, char[] buffer) {
|
||||
this.filename = filename.toCharArray();
|
||||
this.buffer = buffer;
|
||||
|
@ -52,12 +54,11 @@ public class CodeReader {
|
|||
public CodeReader(char[] buffer) {
|
||||
this(NF, buffer);
|
||||
}
|
||||
|
||||
|
||||
// If you are loading up a file normally
|
||||
public CodeReader(String filename) throws IOException
|
||||
{
|
||||
public CodeReader(String filename) throws IOException {
|
||||
this.filename = filename.toCharArray();
|
||||
|
||||
|
||||
FileInputStream stream = new FileInputStream(filename);
|
||||
try {
|
||||
buffer = load(SYSTEM_DEFAULT_ENCODING, stream);
|
||||
|
@ -65,10 +66,10 @@ public class CodeReader {
|
|||
stream.close();
|
||||
}
|
||||
}
|
||||
public CodeReader(String filename, String charSet ) throws IOException
|
||||
{
|
||||
|
||||
public CodeReader(String filename, String charSet) throws IOException {
|
||||
this.filename = filename.toCharArray();
|
||||
|
||||
|
||||
FileInputStream stream = new FileInputStream(filename);
|
||||
try {
|
||||
buffer = load(charSet, stream);
|
||||
|
@ -76,18 +77,15 @@ public class CodeReader {
|
|||
stream.close();
|
||||
}
|
||||
}
|
||||
|
||||
public CodeReader( String filename, InputStream stream ) throws IOException
|
||||
{
|
||||
this( filename, SYSTEM_DEFAULT_ENCODING, stream );
|
||||
|
||||
public CodeReader(String filename, InputStream stream) throws IOException {
|
||||
this(filename, SYSTEM_DEFAULT_ENCODING, stream);
|
||||
}
|
||||
|
||||
public CodeReader( String fileName, String charSet, InputStream stream ) throws IOException {
|
||||
filename = fileName.toCharArray();
|
||||
|
||||
FileInputStream fstream =
|
||||
(stream instanceof FileInputStream)
|
||||
? (FileInputStream)stream
|
||||
|
||||
public CodeReader(String fileName, String charSet, InputStream stream) throws IOException {
|
||||
filename = fileName.toCharArray();
|
||||
|
||||
FileInputStream fstream = (stream instanceof FileInputStream) ? (FileInputStream) stream
|
||||
: new FileInputStream(fileName);
|
||||
try {
|
||||
buffer = load(charSet, fstream);
|
||||
|
@ -99,77 +97,100 @@ public class CodeReader {
|
|||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Load the stream content as a character array. The method loads the stream
|
||||
* content using given character set name. In case if the character set is
|
||||
* not supported, the default one is used.
|
||||
* <p>
|
||||
* If the file is really large, it is silently truncated.
|
||||
* </p>
|
||||
*
|
||||
* @param charSet
|
||||
* Character set name to use for decoding.
|
||||
* @param stream
|
||||
* Input stream
|
||||
* @return Loaded character content
|
||||
* @throws IOException
|
||||
* Load the stream content as a character array. The method loads the stream content using given
|
||||
* character set name. In case if the character set is not supported, the default one is used.
|
||||
*/
|
||||
private char[] load(String charSet, FileInputStream stream)
|
||||
throws IOException {
|
||||
String encoding = Charset.isSupported(charSet) ? charSet
|
||||
: SYSTEM_DEFAULT_ENCODING;
|
||||
|
||||
private char[] load(String charSet, FileInputStream stream) throws IOException {
|
||||
String encoding = Charset.isSupported(charSet) ? charSet : SYSTEM_DEFAULT_ENCODING;
|
||||
FileChannel channel = stream.getChannel();
|
||||
final long lsize = channel.size();
|
||||
final int isize = (int) lsize;
|
||||
if (lsize > MAX_FILE_SIZE) {
|
||||
throw new IOException(
|
||||
"File '" + getPath() + "' is larger than " + MAX_FILE_SIZE / 1024 / 1024 + "mb"); //$NON-NLS-1$//$NON-NLS-2$ //$NON-NLS-3$
|
||||
}
|
||||
|
||||
// In most cases JDK uses Java-written character set decoders. Heap
|
||||
// buffer will work way faster here
|
||||
// Also if the file is larger then 2^31 we truncate it (hope there is
|
||||
// enough heap space)
|
||||
ByteBuffer byteBuffer = ByteBuffer.allocate((int) Math.min(channel
|
||||
.size(), MAX_FILE_SIZE));
|
||||
channel.read(byteBuffer);
|
||||
byteBuffer.flip();
|
||||
|
||||
CharBuffer charBuffer = Charset.forName(encoding).decode(byteBuffer);
|
||||
CharBuffer charBuffer;
|
||||
if (isize < MB) {
|
||||
charBuffer= decodeSmallFile(channel, isize, encoding);
|
||||
} else {
|
||||
charBuffer= decodeLargeFile(channel, isize, encoding);
|
||||
}
|
||||
if (charBuffer.hasArray() && charBuffer.arrayOffset() == 0) {
|
||||
char[] buff= charBuffer.array();
|
||||
if (buff.length == charBuffer.remaining())
|
||||
char[] buff = charBuffer.array();
|
||||
if (buff.length == charBuffer.remaining())
|
||||
return buff;
|
||||
}
|
||||
char[] buff = new char[charBuffer.remaining()];
|
||||
charBuffer.get(buff);
|
||||
return buff;
|
||||
}
|
||||
|
||||
protected char[] xload(FileInputStream stream) throws IOException {
|
||||
FileChannel channel = stream.getChannel();
|
||||
MappedByteBuffer map = channel.map(FileChannel.MapMode.READ_ONLY, 0, channel.size());
|
||||
|
||||
// TODO use the real encoding
|
||||
CharBuffer charBuffer = Charset.forName(SYSTEM_DEFAULT_ENCODING).decode(map);
|
||||
if (charBuffer.hasArray())
|
||||
return charBuffer.array();
|
||||
|
||||
// Got to copy it out
|
||||
char[] buff = new char[charBuffer.length()];
|
||||
charBuffer.get(buff);
|
||||
return buff;
|
||||
|
||||
private CharBuffer decodeSmallFile(FileChannel channel, final int isize, String encoding) throws IOException {
|
||||
ByteBuffer byteBuffer = ByteBuffer.allocate(isize);
|
||||
channel.read(byteBuffer);
|
||||
byteBuffer.flip();
|
||||
|
||||
return Charset.forName(encoding).decode(byteBuffer);
|
||||
}
|
||||
|
||||
|
||||
private CharBuffer decodeLargeFile(FileChannel channel, final int isize, String encoding) throws IOException {
|
||||
int chunk = Math.min(isize, MB);
|
||||
final ByteBuffer in = ByteBuffer.allocate(chunk);
|
||||
final Charset charset = Charset.forName(encoding);
|
||||
final CharsetDecoder decoder = charset.newDecoder().onMalformedInput(CodingErrorAction.REPLACE)
|
||||
.onUnmappableCharacter(CodingErrorAction.REPLACE);
|
||||
|
||||
int n = (int) (isize * (double) decoder.averageCharsPerByte()); // avoid rounding errors.
|
||||
CharBuffer out = CharBuffer.allocate(n);
|
||||
|
||||
int offset = 0;
|
||||
while (offset < isize) {
|
||||
channel.read(in);
|
||||
in.flip();
|
||||
offset += in.limit();
|
||||
|
||||
CoderResult cr = decoder.decode(in, out, offset >= isize);
|
||||
final int remainingBytes = in.remaining();
|
||||
if (cr.isOverflow()) {
|
||||
int totalRemainingBytes= isize-offset + remainingBytes;
|
||||
if (totalRemainingBytes > 0) {
|
||||
n+= (int) (totalRemainingBytes * (double) decoder.maxCharsPerByte()); // avoid rounding errors.
|
||||
CharBuffer o = CharBuffer.allocate(n);
|
||||
out.flip();
|
||||
o.put(out);
|
||||
out = o;
|
||||
}
|
||||
} else if (!cr.isUnderflow()) {
|
||||
cr.throwException();
|
||||
}
|
||||
|
||||
if (remainingBytes == 0) {
|
||||
in.clear();
|
||||
} else {
|
||||
byte[] rest = new byte[remainingBytes];
|
||||
in.get(rest);
|
||||
in.clear();
|
||||
in.put(rest);
|
||||
offset -= remainingBytes;
|
||||
}
|
||||
}
|
||||
out.flip();
|
||||
return out;
|
||||
}
|
||||
|
||||
public boolean isFile() {
|
||||
return !CharArrayUtils.equals( filename, NOFILE );
|
||||
return !CharArrayUtils.equals(filename, NOFILE);
|
||||
}
|
||||
|
||||
@Override
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
return getPath();
|
||||
}
|
||||
|
||||
public String getPath()
|
||||
{
|
||||
return new String( filename );
|
||||
}
|
||||
|
||||
|
||||
public String getPath() {
|
||||
return new String(filename);
|
||||
}
|
||||
}
|
||||
|
|
|
@ -1,5 +1,5 @@
|
|||
/*******************************************************************************
|
||||
* Copyright (c) 2000, 2008 IBM Corporation and others.
|
||||
* Copyright (c) 2000, 2009 IBM Corporation and others.
|
||||
* All rights reserved. This program and the accompanying materials
|
||||
* are made available under the terms of the Eclipse Public License v1.0
|
||||
* which accompanies this distribution, and is available at
|
||||
|
@ -84,6 +84,9 @@ import org.osgi.framework.BundleContext;
|
|||
|
||||
/**
|
||||
* CCorePlugin is the life-cycle owner of the core plug-in, and starting point for access to many core APIs.
|
||||
*
|
||||
* @noextend This class is not intended to be subclassed by clients.
|
||||
* @noinstantiate This class is not intended to be instantiated by clients.
|
||||
*/
|
||||
public class CCorePlugin extends Plugin {
|
||||
|
||||
|
@ -206,28 +209,13 @@ public class CCorePlugin extends Plugin {
|
|||
}
|
||||
|
||||
/**
|
||||
* Answers the shared working copies currently registered for this buffer factory.
|
||||
* Working copies can be shared by several clients using the same buffer factory,see
|
||||
* <code>IWorkingCopy.getSharedWorkingCopy</code>.
|
||||
*
|
||||
* @param factory the given buffer factory
|
||||
* @return the list of shared working copies for a given buffer factory
|
||||
* @see IWorkingCopy
|
||||
* Returns the shared working copies currently registered for the default buffer factory.
|
||||
* @since 5.1
|
||||
*/
|
||||
public static IWorkingCopy[] getSharedWorkingCopies(IBufferFactory factory){
|
||||
|
||||
// if factory is null, default factory must be used
|
||||
if (factory == null) factory = BufferManager.getDefaultBufferManager().getDefaultBufferFactory();
|
||||
Map<IBufferFactory, Map<ITranslationUnit, WorkingCopy>> sharedWorkingCopies = CModelManager.getDefault().sharedWorkingCopies;
|
||||
|
||||
Map<ITranslationUnit, WorkingCopy> perFactoryWorkingCopies = sharedWorkingCopies.get(factory);
|
||||
if (perFactoryWorkingCopies == null) return CModelManager.NoWorkingCopy;
|
||||
Collection<WorkingCopy> copies = perFactoryWorkingCopies.values();
|
||||
IWorkingCopy[] result = new IWorkingCopy[copies.size()];
|
||||
copies.toArray(result);
|
||||
return result;
|
||||
public static IWorkingCopy[] getSharedWorkingCopies() {
|
||||
return getSharedWorkingCopies(null);
|
||||
}
|
||||
|
||||
|
||||
public static String getResourceString(String key) {
|
||||
try {
|
||||
return fgResourceBundle.getString(key);
|
||||
|
@ -262,41 +250,6 @@ public class CCorePlugin extends Plugin {
|
|||
return fgCPlugin;
|
||||
}
|
||||
|
||||
public static void log(String e) {
|
||||
log(createStatus(e));
|
||||
}
|
||||
|
||||
public static void log(Throwable e) {
|
||||
log("Error", e); //$NON-NLS-1$
|
||||
}
|
||||
|
||||
public static void log(String message, Throwable e) {
|
||||
Throwable nestedException;
|
||||
if (e instanceof CModelException
|
||||
&& (nestedException = ((CModelException)e).getException()) != null) {
|
||||
e = nestedException;
|
||||
}
|
||||
log(createStatus(message, e));
|
||||
}
|
||||
|
||||
public static IStatus createStatus(String msg) {
|
||||
return createStatus(msg, null);
|
||||
}
|
||||
|
||||
public static IStatus createStatus(String msg, Throwable e) {
|
||||
return new Status(IStatus.ERROR, PLUGIN_ID, IStatus.ERROR, msg, e);
|
||||
}
|
||||
|
||||
public static void log(IStatus status) {
|
||||
getDefault().getLog().log(status);
|
||||
}
|
||||
|
||||
// ------ CPlugin
|
||||
|
||||
public CCorePlugin() {
|
||||
super();
|
||||
fgCPlugin = this;
|
||||
}
|
||||
|
||||
/**
|
||||
* @see Plugin#shutdown
|
||||
|
@ -646,14 +599,6 @@ public class CCorePlugin extends Plugin {
|
|||
return fCoreModel;
|
||||
}
|
||||
|
||||
/**
|
||||
* @deprecated use getIndexManager().
|
||||
*/
|
||||
@Deprecated
|
||||
public static IPDOMManager getPDOMManager() {
|
||||
return getDefault().pdomManager;
|
||||
}
|
||||
|
||||
public static IIndexManager getIndexManager() {
|
||||
return getDefault().pdomManager;
|
||||
}
|
||||
|
@ -1200,5 +1145,100 @@ public class CCorePlugin extends Plugin {
|
|||
return UserVarSupplier.getInstance();
|
||||
}
|
||||
|
||||
// NON-API
|
||||
|
||||
/**
|
||||
* @noreference This constructor is not intended to be referenced by clients.
|
||||
*/
|
||||
public CCorePlugin() {
|
||||
super();
|
||||
fgCPlugin = this;
|
||||
}
|
||||
|
||||
/**
|
||||
* Answers the shared working copies currently registered for this buffer factory.
|
||||
* Working copies can be shared by several clients using the same buffer factory,see
|
||||
* <code>IWorkingCopy.getSharedWorkingCopy</code>.
|
||||
*
|
||||
* @param factory the given buffer factory
|
||||
* @return the list of shared working copies for a given buffer factory
|
||||
* @see IWorkingCopy
|
||||
* @noreference This method is not intended to be referenced by clients.
|
||||
*/
|
||||
public static IWorkingCopy[] getSharedWorkingCopies(IBufferFactory factory) {
|
||||
// if factory is null, default factory must be used
|
||||
if (factory == null)
|
||||
factory = BufferManager.getDefaultBufferManager().getDefaultBufferFactory();
|
||||
Map<IBufferFactory, Map<ITranslationUnit, WorkingCopy>> sharedWorkingCopies = CModelManager
|
||||
.getDefault().sharedWorkingCopies;
|
||||
|
||||
Map<ITranslationUnit, WorkingCopy> perFactoryWorkingCopies = sharedWorkingCopies.get(factory);
|
||||
if (perFactoryWorkingCopies == null)
|
||||
return CModelManager.NoWorkingCopy;
|
||||
Collection<WorkingCopy> copies = perFactoryWorkingCopies.values();
|
||||
IWorkingCopy[] result = new IWorkingCopy[copies.size()];
|
||||
copies.toArray(result);
|
||||
return result;
|
||||
}
|
||||
|
||||
/**
|
||||
* @noreference This method is not intended to be referenced by clients.
|
||||
*/
|
||||
public static void log(String e) {
|
||||
log(createStatus(e));
|
||||
}
|
||||
|
||||
/**
|
||||
* @noreference This method is not intended to be referenced by clients.
|
||||
*/
|
||||
public static void log(Throwable e) {
|
||||
String msg= e.getMessage();
|
||||
if (msg == null) {
|
||||
log("Error", e); //$NON-NLS-1$
|
||||
} else {
|
||||
log("Error: " + msg, e); //$NON-NLS-1$
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* @noreference This method is not intended to be referenced by clients.
|
||||
*/
|
||||
public static void log(String message, Throwable e) {
|
||||
Throwable nestedException;
|
||||
if (e instanceof CModelException
|
||||
&& (nestedException = ((CModelException)e).getException()) != null) {
|
||||
e = nestedException;
|
||||
}
|
||||
log(createStatus(message, e));
|
||||
}
|
||||
|
||||
/**
|
||||
* @noreference This method is not intended to be referenced by clients.
|
||||
*/
|
||||
public static IStatus createStatus(String msg) {
|
||||
return createStatus(msg, null);
|
||||
}
|
||||
|
||||
/**
|
||||
* @noreference This method is not intended to be referenced by clients.
|
||||
*/
|
||||
public static IStatus createStatus(String msg, Throwable e) {
|
||||
return new Status(IStatus.ERROR, PLUGIN_ID, IStatus.ERROR, msg, e);
|
||||
}
|
||||
|
||||
/**
|
||||
* @noreference This method is not intended to be referenced by clients.
|
||||
*/
|
||||
public static void log(IStatus status) {
|
||||
getDefault().getLog().log(status);
|
||||
}
|
||||
|
||||
/**
|
||||
* @deprecated use getIndexManager().
|
||||
* @noreference This method is not intended to be referenced by clients.
|
||||
*/
|
||||
@Deprecated
|
||||
public static IPDOMManager getPDOMManager() {
|
||||
return getDefault().pdomManager;
|
||||
}
|
||||
}
|
Loading…
Add table
Reference in a new issue