xinb
/
ip2region_private
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

🎨: building\java 中 Searcher 实现 Closeable 接口

This commit is contained in:
jiahao 2023-08-10 14:09:48 +08:00
parent 6ed8bf0118
commit fb8e0ad9c9
6 changed files with 257 additions and 258 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
binding/java/ReadMe.md
View File

@ -15,35 +15,29 @@
```java
import org.lionsoul.ip2region.xdb.Searcher;
import java.io.*;
import java.util.concurrent.TimeUnit;
public class SearcherTest {
public static void main(String[] args) {
// 1、创建 searcher 对象
String dbPath = "ip2region.xdb file path";
Searcher searcher = null;
try {
searcher = Searcher.newWithFileOnly(dbPath);
} catch (IOException e) {
System.out.printf("failed to create searcher with `%s`: %s\n", dbPath, e);
return;
}
String ip = "1.2.3.4";
// 2、查询
try {
String ip = "1.2.3.4";
// 1、创建 searcher 对象
try (Searcher searcher = Searcher.newWithFileOnly(dbPath)) {
// 2、查询
long sTime = System.nanoTime();
String region = searcher.search(ip);
long cost = TimeUnit.NANOSECONDS.toMicros((long) (System.nanoTime() - sTime));
long cost = TimeUnit.NANOSECONDS.toMicros(System.nanoTime() - sTime);
System.out.printf("{region: %s, ioCount: %d, took: %d μs}\n", region, searcher.getIOCount(), cost);
} catch (IOException e) {
System.out.printf("failed to create searcher with `%s`: %s\n", dbPath, e);
} catch (Exception e) {
System.out.printf("failed to search(%s): %s\n", ip, e);
}
// 3、关闭资源（这里通过 try-with-resources 自动关闭）
// 3、关闭资源
searcher.close();
// 备注：并发使用，每个线程需要创建一个独立的 searcher 对象单独使用。
}
}
@ -54,12 +48,14 @@ public class SearcherTest {
我们可以提前从 `xdb` 文件中加载出来 `VectorIndex` 数据，然后全局缓存，每次创建 Searcher 对象的时候使用全局的 VectorIndex 缓存可以减少一次固定的 IO 操作，从而加速查询，减少 IO 压力。
```java
import org.lionsoul.ip2region.xdb.Searcher;
import java.io.*;
import java.util.concurrent.TimeUnit;
public class SearcherTest {
public static void main(String[] args) {
String dbPath = "ip2region.xdb file path";
String ip = "1.2.3.4";
// 1、从 dbPath 中预先加载 VectorIndex 缓存，并且把这个得到的数据作为全局变量，后续反复使用。
byte[] vIndex;
@ -71,27 +67,17 @@ public class SearcherTest {
}
// 2、使用全局的 vIndex 创建带 VectorIndex 缓存的查询对象。
Searcher searcher;
try {
searcher = Searcher.newWithVectorIndex(dbPath, vIndex);
} catch (Exception e) {
System.out.printf("failed to create vectorIndex cached searcher with `%s`: %s\n", dbPath, e);
return;
}
// 3、查询
try {
String ip = "1.2.3.4";
try (Searcher searcher = Searcher.newWithVectorIndex(dbPath, vIndex)) {
long sTime = System.nanoTime();
String region = searcher.search(ip);
long cost = TimeUnit.NANOSECONDS.toMicros((long) (System.nanoTime() - sTime));
long cost = TimeUnit.NANOSECONDS.toMicros(System.nanoTime() - sTime);
System.out.printf("{region: %s, ioCount: %d, took: %d μs}\n", region, searcher.getIOCount(), cost);
} catch (IOException e) {
System.out.printf("failed to create vectorIndex cached searcher with `%s`: %s\n", dbPath, e);
} catch (Exception e) {
System.out.printf("failed to search(%s): %s\n", ip, e);
}
// 4、关闭资源
searcher.close();
// 3、关闭资源（这里通过 try-with-resources 自动关闭）
// 备注：每个线程需要单独创建一个独立的 Searcher 对象，但是都共享全局的制度 vIndex 缓存。
}
@ -103,12 +89,14 @@ public class SearcherTest {
我们也可以预先加载整个 ip2region.xdb 的数据到内存，然后基于这个数据创建查询对象来实现完全基于文件的查询，类似之前的 memory search。
```java
import org.lionsoul.ip2region.xdb.Searcher;
import java.io.*;
import java.util.concurrent.TimeUnit;
public class SearcherTest {
public static void main(String[] args) {
String dbPath = "ip2region.xdb file path";
String ip = "1.2.3.4";
// 1、从 dbPath 加载整个 xdb 到内存。
byte[] cBuff;
@ -123,22 +111,21 @@ public class SearcherTest {
Searcher searcher;
try {
searcher = Searcher.newWithBuffer(cBuff);
} catch (Exception e) {
} catch (IOException e) {
System.out.printf("failed to create content cached searcher: %s\n", e);
return;
}
// 3、查询
try {
String ip = "1.2.3.4";
long sTime = System.nanoTime();
String region = searcher.search(ip);
long cost = TimeUnit.NANOSECONDS.toMicros((long) (System.nanoTime() - sTime));
long cost = TimeUnit.NANOSECONDS.toMicros(System.nanoTime() - sTime);
System.out.printf("{region: %s, ioCount: %d, took: %d μs}\n", region, searcher.getIOCount(), cost);
} catch (Exception e) {
System.out.printf("failed to search(%s): %s\n", ip, e);
}
// 4、关闭资源 - 该 searcher 对象可以安全用于并发，等整个服务关闭的时候再关闭 searcher
// searcher.close();

10
binding/java/pom.xml
View File

@ -41,6 +41,7 @@
<properties>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
<project.reporting.outputEncoding>UTF-8</project.reporting.outputEncoding>
<java.version>1.8</java.version>
<maven.compiler.source>1.6</maven.compiler.source>
<maven.compiler.target>1.6</maven.compiler.target>
</properties>
@ -112,6 +113,15 @@
</execution>
</executions>
</plugin>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-compiler-plugin</artifactId>
<configuration>
<source>${java.version}</source>
<target>${java.version}</target>
</configuration>
</plugin>
</plugins>
</build>

163
binding/java/src/main/java/org/lionsoul/ip2region/SearchTest.java
View File

@ -8,7 +8,10 @@ package org.lionsoul.ip2region;
import org.lionsoul.ip2region.xdb.Searcher;
import java.io.*;
import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.util.concurrent.TimeUnit;
public class SearchTest {
@ -65,7 +68,7 @@ public class SearchTest {
}
}
if (dbPath.length() < 1) {
if (dbPath.isEmpty()) {
System.out.print("java -jar ip2region-{version}.jar search [command options]\n");
System.out.print("options:\n");
System.out.print(" --db string ip2region binary xdb file path\n");
@ -73,32 +76,30 @@ public class SearchTest {
return;
}
Searcher searcher = createSearcher(dbPath, cachePolicy);
final BufferedReader reader = new BufferedReader(new InputStreamReader(System.in));
System.out.printf("ip2region xdb searcher test program, cachePolicy: %s\ntype 'quit' to exit\n", cachePolicy);
while ( true ) {
System.out.print("ip2region>> ");
String line = reader.readLine().trim();
if ( line.length() < 2 ) {
continue;
}
try (Searcher searcher = createSearcher(dbPath, cachePolicy);
BufferedReader reader = new BufferedReader(new InputStreamReader(System.in))) {
System.out.printf("ip2region xdb searcher test program, cachePolicy: %s\ntype 'quit' to exit\n", cachePolicy);
while (true) {
System.out.print("ip2region>> ");
String line = reader.readLine().trim();
if (line.length() < 2) {
continue;
}
if ( line.equalsIgnoreCase("quit") ) {
break;
}
if ("quit".equalsIgnoreCase(line)) {
break;
}
try {
double sTime = System.nanoTime();
String region = searcher.search(line);
long cost = TimeUnit.NANOSECONDS.toMicros((long) (System.nanoTime() - sTime));
System.out.printf("{region: %s, ioCount: %d, took: %d μs}\n", region, searcher.getIOCount(), cost);
} catch (Exception e) {
System.out.printf("{err: %s, ioCount: %d}\n", e, searcher.getIOCount());
try {
double sTime = System.nanoTime();
String region = searcher.search(line);
long cost = TimeUnit.NANOSECONDS.toMicros((long) (System.nanoTime() - sTime));
System.out.printf("{region: %s, ioCount: %d, took: %d μs}\n", region, searcher.getIOCount(), cost);
} catch (Exception e) {
System.out.printf("{err: %s, ioCount: %d}\n", e, searcher.getIOCount());
}
}
}
reader.close();
searcher.close();
System.out.println("searcher test program exited, thanks for trying");
}
@ -121,19 +122,23 @@ public class SearchTest {
String key = r.substring(2, sIdx);
String val = r.substring(sIdx + 1);
if ("db".equals(key)) {
dbPath = val;
} else if ("src".equals(key)) {
srcPath = val;
} else if ("cache-policy".equals(key)) {
cachePolicy = val;
} else {
System.out.printf("undefined option `%s`\n", r);
return;
switch (key) {
case "db":
dbPath = val;
break;
case "src":
srcPath = val;
break;
case "cache-policy":
cachePolicy = val;
break;
default:
System.out.printf("undefined option `%s`\n", r);
return;
}
}
if (dbPath.length() < 1 || srcPath.length() < 1) {
if (dbPath.isEmpty() || srcPath.isEmpty()) {
System.out.print("java -jar ip2region-{version}.jar bench [command options]\n");
System.out.print("options:\n");
System.out.print(" --db string ip2region binary xdb file path\n");
@ -142,61 +147,59 @@ public class SearchTest {
return;
}
Searcher searcher = createSearcher(dbPath, cachePolicy);
long count = 0, costs = 0, tStart = System.nanoTime();
String line;
final BufferedReader reader = new BufferedReader(new FileReader(srcPath));
while ((line = reader.readLine()) != null) {
String l = line.trim();
String[] ps = l.split("\\|", 3);
if (ps.length != 3) {
System.out.printf("invalid ip segment `%s`\n", l);
return;
}
long sip;
try {
sip = Searcher.checkIP(ps[0]);
} catch (Exception e) {
System.out.printf("check start ip `%s`: %s\n", ps[0], e);
return;
}
long eip;
try {
eip = Searcher.checkIP(ps[1]);
} catch (Exception e) {
System.out.printf("check end ip `%s`: %s\n", ps[1], e);
return;
}
if (sip > eip) {
System.out.printf("start ip(%s) should not be greater than end ip(%s)\n", ps[0], ps[1]);
return;
}
long mip = (sip + eip) >> 1;
for (final long ip : new long[]{sip, (sip + mip) >> 1, mip, (mip + eip) >> 1, eip}) {
long sTime = System.nanoTime();
String region = searcher.search(ip);
costs += System.nanoTime() - sTime;
// check the region info
if (!ps[2].equals(region)) {
System.out.printf("failed search(%s) with (%s != %s)\n", Searcher.long2ip(ip), region, ps[2]);
try (Searcher searcher = createSearcher(dbPath, cachePolicy);
BufferedReader reader = new BufferedReader(new FileReader(srcPath))) {
String line;
while ((line = reader.readLine()) != null) {
String l = line.trim();
String[] ps = l.split("\\|", 3);
if (ps.length != 3) {
System.out.printf("invalid ip segment `%s`\n", l);
return;
}
count++;
long sip;
try {
sip = Searcher.checkIP(ps[0]);
} catch (Exception e) {
System.out.printf("check start ip `%s`: %s\n", ps[0], e);
return;
}
long eip;
try {
eip = Searcher.checkIP(ps[1]);
} catch (Exception e) {
System.out.printf("check end ip `%s`: %s\n", ps[1], e);
return;
}
if (sip > eip) {
System.out.printf("start ip(%s) should not be greater than end ip(%s)\n", ps[0], ps[1]);
return;
}
long mip = (sip + eip) >> 1;
for (final long ip : new long[]{sip, (sip + mip) >> 1, mip, (mip + eip) >> 1, eip}) {
long sTime = System.nanoTime();
String region = searcher.search(ip);
costs += System.nanoTime() - sTime;
// check the region info
if (!ps[2].equals(region)) {
System.out.printf("failed search(%s) with (%s != %s)\n", Searcher.long2ip(ip), region, ps[2]);
return;
}
count++;
}
}
}
reader.close();
searcher.close();
long took = System.nanoTime() - tStart;
System.out.printf("Bench finished, {cachePolicy: %s, total: %d, took: %ds, cost: %d μs/op}\n",
cachePolicy, count, TimeUnit.NANOSECONDS.toSeconds(took),
count == 0 ? 0 : TimeUnit.NANOSECONDS.toMicros(costs/count));
count == 0 ? 0 : TimeUnit.NANOSECONDS.toMicros(costs / count));
}
public static void main(String[] args) {

2
binding/java/src/main/java/org/lionsoul/ip2region/UtilTest.java
View File

@ -12,7 +12,7 @@ public class UtilTest {
public static void testIP2Long() {
String ip = "1.2.3.4";
long ipAddr = 0;
long ipAddr;
try {
ipAddr = Searcher.checkIP(ip);
} catch (Exception e) {

15
binding/java/src/main/java/org/lionsoul/ip2region/xdb/Header.java
View File

@ -24,13 +24,14 @@ public class Header {
buffer = buff;
}
@Override public String toString() {
@Override
public String toString() {
return "{" +
"Version: " + version + ',' +
"IndexPolicy: " + indexPolicy + ',' +
"CreatedAt: " + createdAt + ',' +
"StartIndexPtr: " + startIndexPtr + ',' +
"EndIndexPtr: " + endIndexPtr +
'}';
"Version: " + version + ',' +
"IndexPolicy: " + indexPolicy + ',' +
"CreatedAt: " + createdAt + ',' +
"StartIndexPtr: " + startIndexPtr + ',' +
"EndIndexPtr: " + endIndexPtr +
'}';
}
}

270
binding/java/src/main/java/org/lionsoul/ip2region/xdb/Searcher.java
View File

@ -9,22 +9,21 @@ package org.lionsoul.ip2region.xdb;
// @Date 2022/06/23
import java.io.Closeable;
import java.io.IOException;
import java.io.RandomAccessFile;
import java.nio.charset.StandardCharsets;
public class Searcher {
public class Searcher implements Closeable {
// constant defined copied from the xdb maker
public static final int HeaderInfoLength = 256;
public static final int VectorIndexRows = 256;
public static final int VectorIndexCols = 256;
public static final int VectorIndexSize = 8;
public static final int VectorIndexRows = 256;
public static final int VectorIndexCols = 256;
public static final int VectorIndexSize = 8;
public static final int SegmentIndexSize = 14;
public static final byte[] shiftIndex = {24, 16, 8, 0};
// random access file handle for file based search
private final RandomAccessFile handle;
private int ioCount = 0;
// vector index.
// use the byte[] instead of VectorIndex entry array to keep
// the minimal memory allocation.
@ -34,20 +33,7 @@ public class Searcher {
private final byte[] contentBuff;
// --- static method to create searchers
public static Searcher newWithFileOnly(String dbPath) throws IOException {
return new Searcher(dbPath, null, null);
}
public static Searcher newWithVectorIndex(String dbPath, byte[] vectorIndex) throws IOException {
return new Searcher(dbPath, vectorIndex, null);
}
public static Searcher newWithBuffer(byte[] cBuff) throws IOException {
return new Searcher(null, null, cBuff);
}
// --- End of creator
private int ioCount = 0;
public Searcher(String dbFile, byte[] vectorIndex, byte[] cBuff) throws IOException {
if (cBuff != null) {
@ -61,6 +47,130 @@ public class Searcher {
}
}
public static Searcher newWithFileOnly(String dbPath) throws IOException {
return new Searcher(dbPath, null, null);
}
// --- End of creator
public static Searcher newWithVectorIndex(String dbPath, byte[] vectorIndex) throws IOException {
return new Searcher(dbPath, vectorIndex, null);
}
public static Searcher newWithBuffer(byte[] cBuff) throws IOException {
return new Searcher(null, null, cBuff);
}
public static Header loadHeader(RandomAccessFile handle) throws IOException {
handle.seek(0);
final byte[] buff = new byte[HeaderInfoLength];
handle.read(buff);
return new Header(buff);
}
public static Header loadHeaderFromFile(String dbPath) throws IOException {
final RandomAccessFile handle = new RandomAccessFile(dbPath, "r");
final Header header = loadHeader(handle);
handle.close();
return header;
}
public static byte[] loadVectorIndex(RandomAccessFile handle) throws IOException {
handle.seek(HeaderInfoLength);
int len = VectorIndexRows * VectorIndexCols * VectorIndexSize;
final byte[] buff = new byte[len];
int rLen = handle.read(buff);
if (rLen != len) {
throw new IOException("incomplete read: read bytes should be " + len);
}
return buff;
}
public static byte[] loadVectorIndexFromFile(String dbPath) throws IOException {
final RandomAccessFile handle = new RandomAccessFile(dbPath, "r");
final byte[] vIndex = loadVectorIndex(handle);
handle.close();
return vIndex;
}
// --- static cache util function
public static byte[] loadContent(RandomAccessFile handle) throws IOException {
handle.seek(0);
final byte[] buff = new byte[(int) handle.length()];
int rLen = handle.read(buff);
if (rLen != buff.length) {
throw new IOException("incomplete read: read bytes should be " + buff.length);
}
return buff;
}
public static byte[] loadContentFromFile(String dbPath) throws IOException {
final RandomAccessFile handle = new RandomAccessFile(dbPath, "r");
final byte[] content = loadContent(handle);
handle.close();
return content;
}
/* get an int from a byte array start from the specified offset */
public static long getIntLong(byte[] b, int offset) {
return (
((b[offset++] & 0x000000FFL)) |
((b[offset++] << 8) & 0x0000FF00L) |
((b[offset++] << 16) & 0x00FF0000L) |
((b[offset] << 24) & 0xFF000000L)
);
}
public static int getInt(byte[] b, int offset) {
return (
((b[offset++] & 0x000000FF)) |
((b[offset++] << 8) & 0x0000FF00) |
((b[offset++] << 16) & 0x00FF0000) |
((b[offset] << 24) & 0xFF000000)
);
}
public static int getInt2(byte[] b, int offset) {
return (
((b[offset++] & 0x000000FF)) |
((b[offset] << 8) & 0x0000FF00)
);
}
/* long int to ip string */
public static String long2ip(long ip) {
return String.valueOf((ip >> 24) & 0xFF) + '.' +
((ip >> 16) & 0xFF) + '.' + ((ip >> 8) & 0xFF) + '.' + ((ip) & 0xFF);
}
// --- End cache load util function
// --- static util method
/* check the specified ip address */
public static long checkIP(String ip) throws Exception {
String[] ps = ip.split("\\.");
if (ps.length != 4) {
throw new Exception("invalid ip address `" + ip + "`");
}
long ipDst = 0;
for (int i = 0; i < ps.length; i++) {
int val = Integer.parseInt(ps[i]);
if (val > 255) {
throw new Exception("ip part `" + ps[i] + "` should be less then 256");
}
ipDst |= ((long) val << shiftIndex[i]);
}
return ipDst & 0xFFFFFFFFL;
}
@Override
public void close() throws IOException {
if (this.handle != null) {
this.handle.close();
@ -135,7 +245,7 @@ public class Searcher {
// load and return the region data
final byte[] regionBuff = new byte[dataLen];
read(dataPtr, regionBuff);
return new String(regionBuff, "utf-8");
return new String(regionBuff, StandardCharsets.UTF_8);
}
protected void read(int offset, byte[] buffer) throws IOException {
@ -157,116 +267,4 @@ public class Searcher {
}
}
// --- static cache util function
public static Header loadHeader(RandomAccessFile handle) throws IOException {
handle.seek(0);
final byte[] buff = new byte[HeaderInfoLength];
handle.read(buff);
return new Header(buff);
}
public static Header loadHeaderFromFile(String dbPath) throws IOException {
final RandomAccessFile handle = new RandomAccessFile(dbPath, "r");
final Header header = loadHeader(handle);
handle.close();
return header;
}
public static byte[] loadVectorIndex(RandomAccessFile handle) throws IOException {
handle.seek(HeaderInfoLength);
int len = VectorIndexRows * VectorIndexCols * VectorIndexSize;
final byte[] buff = new byte[len];
int rLen = handle.read(buff);
if (rLen != len) {
throw new IOException("incomplete read: read bytes should be " + len);
}
return buff;
}
public static byte[] loadVectorIndexFromFile(String dbPath) throws IOException {
final RandomAccessFile handle = new RandomAccessFile(dbPath, "r");
final byte[] vIndex = loadVectorIndex(handle);
handle.close();
return vIndex;
}
public static byte[] loadContent(RandomAccessFile handle) throws IOException {
handle.seek(0);
final byte[] buff = new byte[(int) handle.length()];
int rLen = handle.read(buff);
if (rLen != buff.length) {
throw new IOException("incomplete read: read bytes should be " + buff.length);
}
return buff;
}
public static byte[] loadContentFromFile(String dbPath) throws IOException {
final RandomAccessFile handle = new RandomAccessFile(dbPath, "r");
final byte[] content = loadContent(handle);
handle.close();
return content;
}
// --- End cache load util function
// --- static util method
/* get an int from a byte array start from the specified offset */
public static long getIntLong(byte[] b, int offset) {
return (
((b[offset++] & 0x000000FFL)) |
((b[offset++] << 8) & 0x0000FF00L) |
((b[offset++] << 16) & 0x00FF0000L) |
((b[offset ] << 24) & 0xFF000000L)
);
}
public static int getInt(byte[] b, int offset) {
return (
((b[offset++] & 0x000000FF)) |
((b[offset++] << 8) & 0x0000FF00) |
((b[offset++] << 16) & 0x00FF0000) |
((b[offset ] << 24) & 0xFF000000)
);
}
public static int getInt2(byte[] b, int offset) {
return (
((b[offset++] & 0x000000FF)) |
((b[offset ] << 8) & 0x0000FF00)
);
}
/* long int to ip string */
public static String long2ip( long ip )
{
return String.valueOf((ip >> 24) & 0xFF) + '.' +
((ip >> 16) & 0xFF) + '.' + ((ip >> 8) & 0xFF) + '.' + ((ip) & 0xFF);
}
public static final byte[] shiftIndex = {24, 16, 8, 0};
/* check the specified ip address */
public static long checkIP(String ip) throws Exception {
String[] ps = ip.split("\\.");
if (ps.length != 4) {
throw new Exception("invalid ip address `" + ip + "`");
}
long ipDst = 0;
for (int i = 0; i < ps.length; i++) {
int val = Integer.parseInt(ps[i]);
if (val > 255) {
throw new Exception("ip part `"+ps[i]+"` should be less then 256");
}
ipDst |= ((long) val << shiftIndex[i]);
}
return ipDst & 0xFFFFFFFFL;
}
}
}