Files
redkale/docs/convert.md
2024-10-09 19:28:47 +08:00

441 lines
14 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

# 序列化
  Convert提供Java对象的序列化与反序列化功能。支持JSON(JavaScript Object Notation)、PROTOBUF两种格式化。 两种格式使用方式完全一样其性能都大幅度超过其他JSON框架。同时JSON内置于HTTP服务中PROTOBUF也是SNCP协议数据序列化的基础。
## 基本API
  JSON序列化其操作类主要是JsonConvert配置类主要是JsonFactory、ConvertColumn。JsonFactory采用同ClassLoader类似的双亲委托方式设计。
JsonConvert 序列化encode方法
```java
public String convertTo(Object value);
public String convertTo(Type type, Object value);
public void convertTo(OutputStream out, Object value);
public void convertTo(OutputStream out, Type type, Object value);
public ByteBuffer[] convertTo(Supplier<ByteBuffer> supplier, Object value);
public ByteBuffer[] convertTo(Supplier<ByteBuffer> supplier, Type type, Object value);
public void convertTo(JsonWriter writer, Object value);
public void convertTo(JsonWriter writer, Type type, Object value);
```
JsonConvert 反序列化decode方法
```java
public <T> T convertFrom(Type type, String text);
public <T> T convertFrom(Type type, char[] text);
public <T> T convertFrom(Type type, char[] text, int start, int len);
public <T> T convertFrom(Type type, InputStream in);
public <T> T convertFrom(Type type, ByteBuffer... buffers);
public <T> T convertFrom(Type type, JsonReader reader);
```
## JSON基本用法
```java
@Data
public class UserRecord {
private int userid;
private String username = "";
private String password = "";
public UserRecord() {
}
}
public static void main(String[] args) throws Exception {
UserRecord user = new UserRecord();
user.setUserid(100);
user.setUsername("redkalename");
user.setPassword("123456");
final JsonConvert convert = JsonConvert.root();
String json = convert.convertTo(user);
System.out.println(json); //应该是 {"password":"123456","userid":100,"username":"redkalename"}
UserRecord user2 = convert.convertFrom(UserRecord.class, json);
//应该也是 {"password":"123456","userid":100,"username":"redkalename"}
System.out.println(convert.convertTo(user2));
/**
* 以下功能是为了屏蔽password字段。
* 等价于 public String getPassword() 加上 @ConvertColumn
*
* @ConvertColumn(ignore = true, type = ConvertType.JSON)
* public String getPassword() {
* return password;
* }
**/
final JsonFactory childFactory = JsonFactory.root().createChild();
childFactory.register(UserRecord.class, true, "password"); //屏蔽掉password字段使其不输出
childFactory.reloadCoder(UserRecord.class); //重新加载Coder使之覆盖父Factory的配置
final JsonConvert childConvert = childFactory.getConvert();
json = childConvert.convertTo(user);
System.out.println(json); //应该是 {"userid":100,"username":"redkalename"}
user2 = childConvert.convertFrom(UserRecord.class, json);
//应该也是 {"userid":100,"username":"redkalename"}
System.out.println(childConvert.convertTo(user2));
}
```
&emsp;&emsp;在Redkale里存在默认的JsonConvert、ProtobufConvert对象。 只需在所有Service、Servlet中增加依赖注入资源。
```java
public class XXXService implements Service {
@Resource
private JsonConvert jsonConvert;
@Resource
private ProtobufConvert protobufConvert;
}
public class XXXServlet extends HttpServlet {
@Resource
private JsonConvert jsonConvert;
@Resource
private ProtobufConvert protobufConvert;
}
```
&emsp;&emsp; 同一类型数据通过设置新的JsonFactory可以有不同的输出。
```java
@Data
public class UserSimpleInfo {
private int userid;
private String username = "";
@ConvertColumn(ignore = true, type = ConvertType.JSON)
private long regtime; //注册时间
@ConvertColumn(ignore = true, type = ConvertType.JSON)
private String regaddr = ""; //注册IP
}
public class UserInfoServlet extends HttpBaseServlet {
@Resource
private UserSerice service;
@Resource
private JsonFactory jsonRootFactory;
@Resource
private JsonConvert detailConvert;
@Override
public void init(HttpContext context, AnyValue config) {
final JsonFactory childFactory = jsonRootFactory.createChild();
childFactory.register(UserSimpleInfo.class, false, "regtime", "regaddr"); //允许输出注册时间与注册地址
childFactory.reloadCoder(UserSimpleInfo.class); //重新加载Coder使之覆盖父Factory的配置
this.detailConvert = childFactory.getConvert();
}
// 获取他人基本信息
@HttpMapping(url = "/user/info/")
public void info(HttpRequest req, HttpResponse resp) throws IOException {
int userid = Integer.parseInt(req.getRequstURILastPath());
UserSimpleInfo user = service.findUserInfo(userid);
resp.finishJson(user); // 不包含用户的注册时间和注册地址字段信息
}
//获取用户自己的信息
@HttpMapping(url = "/user/myinfo")
public void mydetail(HttpRequest req, HttpResponse resp) throws IOException {
int userid = req.currentUser().getUserid(); //获取当前用户ID
UserSimpleInfo user = service.findUserInfo(userid);
resp.finishJson(detailConvert, user); // 包含用户的注册时间和注册地址字段信息
}
}
```
&emsp;&emsp;Convert 支持带参数构造函数。
&emsp;&emsp;&emsp;&emsp;1. public 带参数的构造函数加上 @ConstructorParameters 注解:
```java
public class UserRecord {
private int userid;
private String username = "";
private String password = "";
@ConstructorParameters({"userid", "username", "password"})
public UserRecord(int userid, String username, String password) {
this.userid = userid;
this.username = username;
this.password = password;
}
public int getUserid() {
return userid;
}
public String getUsername() {
return username;
}
public String getPassword() {
return password;
}
}
```
&emsp;&emsp;&emsp;&emsp;2. 自定义Creator
```java
public class UserRecord {
private int userid;
private String username = "";
private String password = "";
UserRecord(int userid, String username, String password) {
this.userid = userid;
this.username = username;
this.password = password;
}
public int getUserid() {
return userid;
}
public String getUsername() {
return username;
}
public String getPassword() {
return password;
}
/**
* 自定义Creator方法。
* 1) 方法名可以随意。
* 2) 方法必须是static。
* 3方法的参数必须为空。
* 4方法的返回类型必须是Creator。
*
* @return
*/
private static Creator<UserRecord> creator() {
return new Creator<UserRecord>() {
@Override
@ConstructorParameters({"userid", "username", "password"}) //带参数的构造函数必须有ConstructorParameters注解
public UserRecord create(Object... params) {
int userid = (params[0] == null ? 0 : (Integer) params[0]);
return new UserRecord(userid, (String) params[1], (String) params[2]);
}
};
}
}
```
&emsp;&emsp;通常JavaBean类默认有个public空参数的构造函数因此大部分情况下不要自定义Creator其实只要不是private的空参数构造函数Convert都能自动支持(其他的框架都仅支持public的构造函数)只有仅含private的构造函数才需要自定义Creator。带参数的构造函数就需要标记@ConstructorParameters常见于Immutable Object。
## 自定义
&emsp;&emsp;Convert 支持自定义Decode、Encode。
&emsp;&emsp;&emsp;&emsp;1. 通过ConvertFactory显式的注册
```java
public class FileSimpleCoder<R extends Reader, W extends Writer> extends SimpledCoder<R, W, File> {
public static final FileSimpleCoder instance = new FileSimpleCoder();
@Override
public void convertTo(W out, File value) {
out.writeString(value == null ? null : value.getPath());
}
@Override
public File convertFrom(R in) {
String path = in.readString();
return path == null ? null : new File(path);
}
}
JsonFactory.root().register(File.class, FileSimpleCoder.instance);
ProtobufFactory.root().register(File.class, FileSimpleCoder.instance);
```
&emsp;&emsp;&emsp;&emsp;2. 通过JavaBean类自定义静态方法自动加载
```java
public class InnerCoderEntity {
private final String val;
private final int id;
private InnerCoderEntity(int id, String value) {
this.id = id;
this.val = value;
}
public static InnerCoderEntity create(int id, String value) {
return new InnerCoderEntity(id, value);
}
/**
* 该方法提供给Convert组件自动加载。
* 1) 方法名可以随意。
* 2) 方法必须是static
* 3方法的参数有且只能有一个 且必须是org.redkale.convert.ConvertFactory或子类。
* —3.1) 参数类型为org.redkale.convert.ConvertFactory 表示适合JSON,PROTOBUF。
* —3.2) 参数类型为org.redkale.convert.json.JsonFactory 表示仅适合JSON。
* —3.3) 参数类型为org.redkale.convert.pb.ProtobufFactory 表示仅适合PROTOBUF。
* 4方法的返回类型必须是 Decodeable/Encodeable/SimpledCoder
* 若返回类型不是SimpledCoder, 就必须提供两个方法: 一个返回Decodeable 一个返回 Encodeable。
*
* @param factory
* @return
*/
private static SimpledCoder<Reader, Writer, InnerCoderEntity> createConvertCoder(final ConvertFactory factory) {
return new SimpledCoder<Reader, Writer, InnerCoderEntity>() {
//必须与EnMember[] 顺序一致
private final DeMember[] deMembers = new DeMember[]{
DeMember.create(factory, InnerCoderEntity.class, "id"),
DeMember.create(factory, InnerCoderEntity.class, "val")};
//必须与DeMember[] 顺序一致
private final EnMember[] enMembers = new EnMember[]{
EnMember.create(factory, InnerCoderEntity.class, "id"),
EnMember.create(factory, InnerCoderEntity.class, "val")};
@Override
public void convertTo(Writer out, InnerCoderEntity value) {
if (value == null) {
out.writeObjectNull(InnerCoderEntity.class);
return;
}
out.writeObjectB(value);
for (EnMember member : enMembers) {
out.writeObjectField(member, value);
}
out.writeObjectE(value);
}
@Override
public InnerCoderEntity convertFrom(Reader in) {
if (in.readObjectB(InnerCoderEntity.class) == null) return null;
int index = 0;
final Object[] params = new Object[deMembers.length];
while (in.hasNext()) {
DeMember member = in.readFieldName(deMembers); //读取字段名
in.readBlank(); //读取字段名与字段值之间的间隔符JSON则是跳过冒号:
if (member == null) {
in.skipValue(); //跳过不存在的字段的值, 一般不会发生
} else {
params[index++] = member.read(in);
}
}
in.readObjectE(InnerCoderEntity.class);
return InnerCoderEntity.create(params[0] == null ? 0 : (Integer) params[0], (String) params[1]);
}
};
}
public int getId() {
return id;
}
public String getVal() {
return val;
}
@Override
public String toString() {
return JsonConvert.root().convertTo(this);
}
}
```
## RestConvert
&emsp;&emsp;```@RestConvert```、```@RestConvertCoder```是针对```@RestService```接口进行自定义序列化注解
```java
@Data
public class RestConvertItem {
private long createTime;
@ConvertColumn(ignore = true)
private String aesKey;
}
@Data
public class RestConvertBean {
private int id;
private boolean enable;
private String name;
private RestConvertItem content;
}
//将boolean类型值转换成0/1的int值
public class RestConvertBoolCoder<R extends Reader, W extends Writer> extends SimpledCoder<R, W, Boolean> {
@Override
public void convertTo(W out, Boolean value) {
out.writeInt(value == null || !value ? 0 : 1);
}
@Override
public Boolean convertFrom(R in) {
return in.readInt() == 1;
}
}
@RestService(name = "test", autoMapping = true)
public class RestConvertService extends AbstractService {
//输出: {"content":{"createTime":100},"enable":true,"id":123,"name":"haha"}
public RestConvertBean load1() {
return createBean();
}
//输出: {"content":{"aesKey":"keykey","createTime":100},"enable":true,"id":123,"name":"haha"}
//aesKey字段也会被输出
@RestConvert(type = RestConvertItem.class, skipIgnore = true)
public RestConvertBean load2() {
return createBean();
}
//输出: {"id":123}
//只输出id字段
@RestConvert(type = RestConvertBean.class, onlyColumns = "id")
public RestConvertBean load3() {
return createBean();
}
//输出: {"content":{"createTime":100},"enable":1,"id":123,"name":"haha"}
//enable字段输出1而不是true
@RestConvertCoder(type = RestConvertBean.class, field = "enable", coder = RestConvertBoolCoder.class)
public RestConvertBean load4() {
return createBean();
}
private RestConvertBean createBean() {
RestConvertBean bean = new RestConvertBean();
bean.setId(123);
bean.setName("haha");
bean.setEnable(true);
RestConvertItem item = new RestConvertItem();
item.setCreateTime(100);
item.setAesKey("keykey");
bean.setContent(item);
return bean;
}
}
```