spring JarFileEntries 源码
springboot JarFileEntries 代码
文件路径:/spring-boot-project/spring-boot-tools/spring-boot-loader/src/main/java/org/springframework/boot/loader/jar/JarFileEntries.java
/*
* Copyright 2012-2022 the original author or authors.
*
* 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
*
* https://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 org.springframework.boot.loader.jar;
import java.io.IOException;
import java.io.InputStream;
import java.util.Arrays;
import java.util.Collections;
import java.util.Iterator;
import java.util.LinkedHashMap;
import java.util.Map;
import java.util.NoSuchElementException;
import java.util.jar.Attributes;
import java.util.jar.Attributes.Name;
import java.util.jar.JarInputStream;
import java.util.jar.Manifest;
import java.util.zip.ZipEntry;
import org.springframework.boot.loader.data.RandomAccessData;
/**
* Provides access to entries from a {@link JarFile}. In order to reduce memory
* consumption entry details are stored using arrays. The {@code hashCodes} array stores
* the hash code of the entry name, the {@code centralDirectoryOffsets} provides the
* offset to the central directory record and {@code positions} provides the original
* order position of the entry. The arrays are stored in hashCode order so that a binary
* search can be used to find a name.
* <p>
* A typical Spring Boot application will have somewhere in the region of 10,500 entries
* which should consume about 122K.
*
* @author Phillip Webb
* @author Andy Wilkinson
*/
class JarFileEntries implements CentralDirectoryVisitor, Iterable<JarEntry> {
private static final Runnable NO_VALIDATION = () -> {
};
private static final String META_INF_PREFIX = "META-INF/";
private static final Name MULTI_RELEASE = new Name("Multi-Release");
private static final int BASE_VERSION = 8;
private static final int RUNTIME_VERSION = Runtime.version().feature();
private static final long LOCAL_FILE_HEADER_SIZE = 30;
private static final char SLASH = '/';
private static final char NO_SUFFIX = 0;
protected static final int ENTRY_CACHE_SIZE = 25;
private final JarFile jarFile;
private final JarEntryFilter filter;
private RandomAccessData centralDirectoryData;
private int size;
private int[] hashCodes;
private Offsets centralDirectoryOffsets;
private int[] positions;
private Boolean multiReleaseJar;
private JarEntryCertification[] certifications;
private final Map<Integer, FileHeader> entriesCache = Collections
.synchronizedMap(new LinkedHashMap<>(16, 0.75f, true) {
@Override
protected boolean removeEldestEntry(Map.Entry<Integer, FileHeader> eldest) {
return size() >= ENTRY_CACHE_SIZE;
}
});
JarFileEntries(JarFile jarFile, JarEntryFilter filter) {
this.jarFile = jarFile;
this.filter = filter;
}
@Override
public void visitStart(CentralDirectoryEndRecord endRecord, RandomAccessData centralDirectoryData) {
int maxSize = endRecord.getNumberOfRecords();
this.centralDirectoryData = centralDirectoryData;
this.hashCodes = new int[maxSize];
this.centralDirectoryOffsets = Offsets.from(endRecord);
this.positions = new int[maxSize];
}
@Override
public void visitFileHeader(CentralDirectoryFileHeader fileHeader, long dataOffset) {
AsciiBytes name = applyFilter(fileHeader.getName());
if (name != null) {
add(name, dataOffset);
}
}
private void add(AsciiBytes name, long dataOffset) {
this.hashCodes[this.size] = name.hashCode();
this.centralDirectoryOffsets.set(this.size, dataOffset);
this.positions[this.size] = this.size;
this.size++;
}
@Override
public void visitEnd() {
sort(0, this.size - 1);
int[] positions = this.positions;
this.positions = new int[positions.length];
for (int i = 0; i < this.size; i++) {
this.positions[positions[i]] = i;
}
}
int getSize() {
return this.size;
}
private void sort(int left, int right) {
// Quick sort algorithm, uses hashCodes as the source but sorts all arrays
if (left < right) {
int pivot = this.hashCodes[left + (right - left) / 2];
int i = left;
int j = right;
while (i <= j) {
while (this.hashCodes[i] < pivot) {
i++;
}
while (this.hashCodes[j] > pivot) {
j--;
}
if (i <= j) {
swap(i, j);
i++;
j--;
}
}
if (left < j) {
sort(left, j);
}
if (right > i) {
sort(i, right);
}
}
}
private void swap(int i, int j) {
swap(this.hashCodes, i, j);
this.centralDirectoryOffsets.swap(i, j);
swap(this.positions, i, j);
}
@Override
public Iterator<JarEntry> iterator() {
return new EntryIterator(NO_VALIDATION);
}
Iterator<JarEntry> iterator(Runnable validator) {
return new EntryIterator(validator);
}
boolean containsEntry(CharSequence name) {
return getEntry(name, FileHeader.class, true) != null;
}
JarEntry getEntry(CharSequence name) {
return getEntry(name, JarEntry.class, true);
}
InputStream getInputStream(String name) throws IOException {
FileHeader entry = getEntry(name, FileHeader.class, false);
return getInputStream(entry);
}
InputStream getInputStream(FileHeader entry) throws IOException {
if (entry == null) {
return null;
}
InputStream inputStream = getEntryData(entry).getInputStream();
if (entry.getMethod() == ZipEntry.DEFLATED) {
inputStream = new ZipInflaterInputStream(inputStream, (int) entry.getSize());
}
return inputStream;
}
RandomAccessData getEntryData(String name) throws IOException {
FileHeader entry = getEntry(name, FileHeader.class, false);
if (entry == null) {
return null;
}
return getEntryData(entry);
}
private RandomAccessData getEntryData(FileHeader entry) throws IOException {
// aspectjrt-1.7.4.jar has a different ext bytes length in the
// local directory to the central directory. We need to re-read
// here to skip them
RandomAccessData data = this.jarFile.getData();
byte[] localHeader = data.read(entry.getLocalHeaderOffset(), LOCAL_FILE_HEADER_SIZE);
long nameLength = Bytes.littleEndianValue(localHeader, 26, 2);
long extraLength = Bytes.littleEndianValue(localHeader, 28, 2);
return data.getSubsection(entry.getLocalHeaderOffset() + LOCAL_FILE_HEADER_SIZE + nameLength + extraLength,
entry.getCompressedSize());
}
private <T extends FileHeader> T getEntry(CharSequence name, Class<T> type, boolean cacheEntry) {
T entry = doGetEntry(name, type, cacheEntry, null);
if (!isMetaInfEntry(name) && isMultiReleaseJar()) {
int version = RUNTIME_VERSION;
AsciiBytes nameAlias = (entry instanceof JarEntry jarEntry) ? jarEntry.getAsciiBytesName()
: new AsciiBytes(name.toString());
while (version > BASE_VERSION) {
T versionedEntry = doGetEntry("META-INF/versions/" + version + "/" + name, type, cacheEntry, nameAlias);
if (versionedEntry != null) {
return versionedEntry;
}
version--;
}
}
return entry;
}
private boolean isMetaInfEntry(CharSequence name) {
return name.toString().startsWith(META_INF_PREFIX);
}
private boolean isMultiReleaseJar() {
Boolean multiRelease = this.multiReleaseJar;
if (multiRelease != null) {
return multiRelease;
}
try {
Manifest manifest = this.jarFile.getManifest();
if (manifest == null) {
multiRelease = false;
}
else {
Attributes attributes = manifest.getMainAttributes();
multiRelease = attributes.containsKey(MULTI_RELEASE);
}
}
catch (IOException ex) {
multiRelease = false;
}
this.multiReleaseJar = multiRelease;
return multiRelease;
}
private <T extends FileHeader> T doGetEntry(CharSequence name, Class<T> type, boolean cacheEntry,
AsciiBytes nameAlias) {
int hashCode = AsciiBytes.hashCode(name);
T entry = getEntry(hashCode, name, NO_SUFFIX, type, cacheEntry, nameAlias);
if (entry == null) {
hashCode = AsciiBytes.hashCode(hashCode, SLASH);
entry = getEntry(hashCode, name, SLASH, type, cacheEntry, nameAlias);
}
return entry;
}
private <T extends FileHeader> T getEntry(int hashCode, CharSequence name, char suffix, Class<T> type,
boolean cacheEntry, AsciiBytes nameAlias) {
int index = getFirstIndex(hashCode);
while (index >= 0 && index < this.size && this.hashCodes[index] == hashCode) {
T entry = getEntry(index, type, cacheEntry, nameAlias);
if (entry.hasName(name, suffix)) {
return entry;
}
index++;
}
return null;
}
@SuppressWarnings("unchecked")
private <T extends FileHeader> T getEntry(int index, Class<T> type, boolean cacheEntry, AsciiBytes nameAlias) {
try {
long offset = this.centralDirectoryOffsets.get(index);
FileHeader cached = this.entriesCache.get(index);
FileHeader entry = (cached != null) ? cached
: CentralDirectoryFileHeader.fromRandomAccessData(this.centralDirectoryData, offset, this.filter);
if (CentralDirectoryFileHeader.class.equals(entry.getClass()) && type.equals(JarEntry.class)) {
entry = new JarEntry(this.jarFile, index, (CentralDirectoryFileHeader) entry, nameAlias);
}
if (cacheEntry && cached != entry) {
this.entriesCache.put(index, entry);
}
return (T) entry;
}
catch (IOException ex) {
throw new IllegalStateException(ex);
}
}
private int getFirstIndex(int hashCode) {
int index = Arrays.binarySearch(this.hashCodes, 0, this.size, hashCode);
if (index < 0) {
return -1;
}
while (index > 0 && this.hashCodes[index - 1] == hashCode) {
index--;
}
return index;
}
void clearCache() {
this.entriesCache.clear();
}
private AsciiBytes applyFilter(AsciiBytes name) {
return (this.filter != null) ? this.filter.apply(name) : name;
}
JarEntryCertification getCertification(JarEntry entry) throws IOException {
JarEntryCertification[] certifications = this.certifications;
if (certifications == null) {
certifications = new JarEntryCertification[this.size];
// We fall back to use JarInputStream to obtain the certs. This isn't that
// fast, but hopefully doesn't happen too often.
try (JarInputStream certifiedJarStream = new JarInputStream(this.jarFile.getData().getInputStream())) {
java.util.jar.JarEntry certifiedEntry;
while ((certifiedEntry = certifiedJarStream.getNextJarEntry()) != null) {
// Entry must be closed to trigger a read and set entry certificates
certifiedJarStream.closeEntry();
int index = getEntryIndex(certifiedEntry.getName());
if (index != -1) {
certifications[index] = JarEntryCertification.from(certifiedEntry);
}
}
}
this.certifications = certifications;
}
JarEntryCertification certification = certifications[entry.getIndex()];
return (certification != null) ? certification : JarEntryCertification.NONE;
}
private int getEntryIndex(CharSequence name) {
int hashCode = AsciiBytes.hashCode(name);
int index = getFirstIndex(hashCode);
while (index >= 0 && index < this.size && this.hashCodes[index] == hashCode) {
FileHeader candidate = getEntry(index, FileHeader.class, false, null);
if (candidate.hasName(name, NO_SUFFIX)) {
return index;
}
index++;
}
return -1;
}
private static void swap(int[] array, int i, int j) {
int temp = array[i];
array[i] = array[j];
array[j] = temp;
}
private static void swap(long[] array, int i, int j) {
long temp = array[i];
array[i] = array[j];
array[j] = temp;
}
/**
* Iterator for contained entries.
*/
private final class EntryIterator implements Iterator<JarEntry> {
private final Runnable validator;
private int index = 0;
private EntryIterator(Runnable validator) {
this.validator = validator;
validator.run();
}
@Override
public boolean hasNext() {
this.validator.run();
return this.index < JarFileEntries.this.size;
}
@Override
public JarEntry next() {
this.validator.run();
if (!hasNext()) {
throw new NoSuchElementException();
}
int entryIndex = JarFileEntries.this.positions[this.index];
this.index++;
return getEntry(entryIndex, JarEntry.class, false, null);
}
}
/**
* Interface to manage offsets to central directory records. Regular zip files are
* backed by an {@code int[]} based implementation, Zip64 files are backed by a
* {@code long[]} and will consume more memory.
*/
private interface Offsets {
void set(int index, long value);
long get(int index);
void swap(int i, int j);
static Offsets from(CentralDirectoryEndRecord endRecord) {
int size = endRecord.getNumberOfRecords();
return endRecord.isZip64() ? new Zip64Offsets(size) : new ZipOffsets(size);
}
}
/**
* {@link Offsets} implementation for regular zip files.
*/
private static final class ZipOffsets implements Offsets {
private final int[] offsets;
private ZipOffsets(int size) {
this.offsets = new int[size];
}
@Override
public void swap(int i, int j) {
JarFileEntries.swap(this.offsets, i, j);
}
@Override
public void set(int index, long value) {
this.offsets[index] = (int) value;
}
@Override
public long get(int index) {
return this.offsets[index];
}
}
/**
* {@link Offsets} implementation for zip64 files.
*/
private static final class Zip64Offsets implements Offsets {
private final long[] offsets;
private Zip64Offsets(int size) {
this.offsets = new long[size];
}
@Override
public void swap(int i, int j) {
JarFileEntries.swap(this.offsets, i, j);
}
@Override
public void set(int index, long value) {
this.offsets[index] = value;
}
@Override
public long get(int index) {
return this.offsets[index];
}
}
}
相关信息
相关文章
spring CentralDirectoryEndRecord 源码
spring CentralDirectoryFileHeader 源码
spring CentralDirectoryParser 源码
0
赞
- 所属分类: 后端技术
- 本文标签: Spring Boot Java Spring
热门推荐
-
2、 - 优质文章
-
3、 gate.io
-
8、 golang
-
9、 openharmony
-
10、 Vue中input框自动聚焦