https://www.softwarecollections.org/en/不用再GOOGLE寻找安装方法。
安装MYSQL示例：
# 2. Install the collection:
$ sudo yum install rh-mariadb103
# 3. Start using software collections:
$ scl enable rh-mariadb103 bash
$ service rh-mariadb103-mariadb start
$ mysql
$ mysqld
#开机加载命令
cp /opt/rh/rh-mariadb103/enable /etc/profile.d/rh-mariadb103.sh
当CLIENT或用户在KEYCLOAK中成功登录后，会返回JWT字符串，其中默认含有权限的信息，但此信息以内嵌的方式呈现，非常不方便。
"resource_access": {
"app-springboot-confidential": {
"roles": [
"user"
]
},
"test-employee-service": {
"roles": [
"READ_EMPLOYEE"
]
},
"service-springboot": {
"roles": [
"READ_PRODUCTS"
]
},
"account": {
"roles": [
"manage-account",
"manage-account-links"
]
},
"test-department-service": {
"roles": [
"READ_DEPARTMENT"
]
}
}
- 需要将权限的信息输出到一个KEY中，这时可以新增自定义CLIENT SCOPE。Mapper中新增KEYCLOAK已内置的【realm roles/client roles】，定义输出到JTW的字段名：my-roles。
- 授权哪些CLIENT可以读取此CLIENT SCOPE.
- 在登录参数scope中，加入此值：my-roles，这样在输出的JWT就会以平面的方式输出所有roles
"my-roles": [
"user",
"READ_EMPLOYEE",
"READ_PRODUCTS",
"manage-account",
"manage-account-links",
"READ_DEPARTMENT",
"offline_access",
"user"
]
@Bean
public ReactiveJwtAuthenticationConverter jwtAuthenticationConverter(ObjectMapper objectMapper) {
JwtGrantedAuthoritiesConverter jwtGrantedAuthoritiesConverter = new JwtGrantedAuthoritiesConverter();
jwtGrantedAuthoritiesConverter.setAuthorityPrefix("ROLE_");
jwtGrantedAuthoritiesConverter.setAuthoritiesClaimName("my-roles");
// KeycloakRealmRoleConverter keycloakRealmRoleConverter = new KeycloakRealmRoleConverter(objectMapper);
ReactiveJwtGrantedAuthoritiesConverterAdapter reactiveJwtGrantedAuthoritiesConverterAdapter =
new ReactiveJwtGrantedAuthoritiesConverterAdapter(
// new KeycloakRealmRoleConverter(objectMapper);
jwtGrantedAuthoritiesConverter
);
ReactiveJwtAuthenticationConverter jwtConverter = new ReactiveJwtAuthenticationConverter();
jwtConverter.setJwtGrantedAuthoritiesConverter(reactiveJwtGrantedAuthoritiesConverterAdapter);
return jwtConverter;
}
ServerHttpSecurity
.authorizeExchange(
a -> a.pathMatchers("/", "/error").permitAll()
.matchers(PathRequest.toStaticResources().atCommonLocations()).permitAll()
.pathMatchers(HttpMethod.GET, "/protected/**").hasRole("READ_DEPARTMENT")
.anyExchange()
.authenticated()
)
添加依赖，pom.xml
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-webflux</artifactId>
</dependency>
<!-- spring session with mongodb -->
<dependency>
<groupId>org.springframework.session</groupId>
<artifactId>spring-session-data-mongodb</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-data-mongodb-reactive</artifactId>
</dependency>
配置文件，application.yaml
spring:
session:
store-type: mongodb
timeout: 30s
mongodb:
collection-name: WEB_SESSIONS
java配置，HttpSessionConfiguration.java：
package com.paul.testmicroservicecommon.config;
import org.springframework.boot.autoconfigure.session.MongoSessionProperties;
import org.springframework.boot.autoconfigure.session.SessionProperties;
import org.springframework.boot.context.properties.EnableConfigurationProperties;
import org.springframework.context.annotation.Bean;
import org.springframework.session.config.ReactiveSessionRepositoryCustomizer;
import org.springframework.session.data.mongo.ReactiveMongoSessionRepository;
import org.springframework.session.data.mongo.config.annotation.web.reactive.EnableMongoWebSession;
@EnableMongoWebSession
@EnableConfigurationProperties(MongoSessionProperties.class)
public class HttpSessionConfiguration {
@Bean
public ReactiveSessionRepositoryCustomizer<ReactiveMongoSessionRepository> customize(
SessionProperties sessionProperties,
MongoSessionProperties mongoSessionProperties
){
return c -> {
c.setMaxInactiveIntervalInSeconds((int)sessionProperties.getTimeout().getSeconds());
c.setCollectionName(mongoSessionProperties.getCollectionName());
};
}
}
目前SPRING CLOUD(2020)尚未支持REACTIVE FEIGN，但官方推荐使用feign-reactive。
pom.xml
<dependency>
<groupId>com.playtika.reactivefeign</groupId>
<artifactId>feign-reactor-spring-cloud-starter</artifactId>
<version>3.1.2</version>
<type>pom</type>
</dependency>
LoanDecisionClientReactive.java
package com.paul.testspringcloudstream.loancheck.service;
import org.springframework.web.bind.annotation.PostMapping;
import org.springframework.web.bind.annotation.RequestBody;
import com.paul.testspringcloudstream.common.model.Loan;
import reactivefeign.spring.config.ReactiveFeignClient;
import reactor.core.publisher.Mono;
@ReactiveFeignClient(name = "loan-decision")
public interface LoanDecisionClientReactive {
@PostMapping("/loan-decision")
public Mono<Loan> getDecision(@RequestBody Loan loan);
}
LoanCheckConfiguration.java
@Configuration
@Import({
MongoDbConsumerConfiguration.class,
})
@EnableDiscoveryClient
@EnableReactiveFeignClients("com.paul.testspringcloudstream.loancheck.service")
public class LoanCheckConfiguration {
}
使用同feign
@Autowired
private LoanDecisionClientReactive loanDecisionClientReactive;
Reference
https://blog.csdn.net/LCBUSHIHAHA/article/details/113817966官方SAMPLE
https://github.com/kptfh/feign-reactive-sample
升级spring cloud版本之后发现bootstrap.yml 失效了，阅读官方文档得知，需要新增一个引用来开启bootstrap.xml文件的读取，新版spring cloud默认是关闭读取了。
增加依赖如下即可：
<dependency>
<groupId>org.springframework.cloud</groupId>
<artifactId>spring-cloud-starter-bootstrap</artifactId>
</dependency>
官方文档：
https://docs.spring.io/spring-cloud-config/docs/current/reference/html/#config-first-bootstrap
SPRING REACTOR 之Flux和Mono，有点象SPRING INTEGRATION的IntegrationFlow，有如下特点
- 定义了针对某种类型数据的处理流程
- 可以进行类型转换
- 长期运行，除非被要求中止
- 流程中的每种操作可以在新的线程中执行
- 可以正常中止，如果中途有异常，则该流程也会中止
- 要subscribe，流程才开始被启动
- 可以分割成各个子流程
- 可以聚合子流程
- Mono发送一个数据，就发送中止信号
- Flux发送任意数据，由程序决定何时发送中止信号
编程则比较简单，先根据不同的数据类型定义不同的Flux或Mono，业务操作用Function包装后，放在map/flatmap中，再调用subscribe启动流程。