基于 Dinky + FlinkSQL + Flink CDC 同步 MySQL 数据到 Elasticsearch、Kafka

only°夏至besos2025-12-25 8:46

一、概述

Dinky是一个开箱即用的一站式实时计算平台以 Apache Flink 为基础，连接 OLAP 和数据湖等众多框架致力于流批一体和湖仓一体的建设与实践。本文以此为FlinkSQL可视化工具。

Flink SQL使得使用标准 SQL 开发流式应用变得简单，免去代码开发。

Flink CDC 本文使用 MySQL CDC 连接器 允许从 MySQL 数据库读取快照数据和增量数据。

环境及主要软件版本说明

二、软件安装部署

本文全部采用单机最简环境。

复制代码 
tar -xzf flink-1.14.6-bin-scala_2.12.tgz
cd flink-1.14.6 && ls -l
### 修改配置 flink-1.14.6/conf/flink-conf.yaml,修改成本机的IP、位置
jobmanager.rpc.address: 192.168.xxx.xxx
taskmanager.host: 192.168.xxx.xxx
jobmanager.memory.process.size: 2048m
taskmanager.memory.process.size: 2048m
taskmanager.numberOfTaskSlots: 10
rest.port: 8888
rest.address: 0.0.0.0
web.tmpdir: /home/soft/flink_/flink-1.14.6/tmp
akka.ask.timeout: 600s
### 将flink-sql-connector-kafka_2.12-1.14.6.jar、flink-sql-connector-elasticsearch7_2.12-1.14.6.jar、
### flink-sql-connector-mysql-cdc-2.3.0.jar、放到flink-1.14.6/lib/ 下
### 启动
./

启动成功后，即可访问Flink UI 可视化界面 http://192.168.xxx.xxx:8888/

2.Kafka

复制代码 
tar -xzf kafka_2.12-3.4.0.tgz
cd kafka_2.12-3.4.0
bin/ random-uuid > uuid
vi config/kraft/server.properties
controller.quorum.voters=1@192.168.XXX.XXX:9093
advertised.listeners=PLAINTEXT://192.168.XXX.XXX:9092
log.dirs=/home/soft/kafka_2.12-3.4.0/kraft-combined-logs
### 使用前面生成的uuid格式化Kafka日志目录
bin/ format -t `cat uuid` -c config/kraft/server.properties
Formatting /home/soft/kafka_2.12-3.4.0/kraft-combined-logs with metadata.version 3.4-IV0.

### Start the Kafka Server
bin/ -daemon config/kraft/server.properties
### Kafka 服务器成功启动后，您将拥有一个正在运行并可以使用的基本 Kafka 环境。
jps -l
14085 kafka.Kafka
#### 启动命令说明###############################################################
### 1、前台启动命令:
### bin/ config/server.properties
### 描述：
### 此种方式是窗口运行。一旦窗口关闭或者执行了CTR+C，那么kafka进程就被kill了，kafka服务端就被关闭了
### 2、后台启动命令：
### (1) nohup bin/ config/server.properties 2>&1 &
### (2)bin/ -daemon config/server.properties

3.Dinky

部署参考：http://www.dlink.top/docs/next/deploy_guide/build

  1. 数据库初始化

    复制代码 
     #登录mysql
 mysql -uroot -p123456
 #创建数据库
 mysql>
 create database dinky;
 #授权
 mysql>
 grant all privileges on dinky.* to 'dinky'@'%' identified by 'dinky@123' with grant option;
 mysql>
 flush privileges;
 #此处用 dinky 用户登录
 mysql -h fdw1 -udinky -pdinky@123
 mysql>
 use dinky;
 # sql文件在下载的压缩包中获取即可
 mysql> source /home/soft/dinky/sql/dinky.sql

  2. Dinky部署

    tar -zxvf dlink-release-0.7.3.tar.gz
    mv dlink-release-0.7.3.tar.gz dinky
    #切换目录,修改数据库配置及端口信息
    cd /home/soft/dinky/config/
    vi application.yml

    -------------------------------------------------------------------------------

    spring:
    datasource:
    url: jdbc:mysql://{MYSQL_ADDR:192.168.XXX.XXX:3306}/{MYSQL_DATABASE:dinky}?useUnicode=true&characterEncoding=UTF-8&autoReconnect=true&useSSL=false&zeroDateTimeBehavior=convertToNull&serverTime
    zone=Asia/Shanghai&allowPublicKeyRetrieval=true
    username: {MYSQL_USERNAME:dinky} password: {MYSQL_PASSWORD:dinky@123}
    server:
    port: 8899

    -------------------------------------------------------------------------------

    加载依赖Dinky 需要具备自身的 Flink 环境，该 Flink 环境的实现需要用户自己在 Dinky 根目录下

    plugins/flink${FLINK_VERSION} 文件夹并上传相关的 Flink 依赖，如 flink-dist, flink-table 等

    复制代码 
     cp ${FLINK_HOME}/lib/flink-*.jar  /home/soft/dinky/plugins/flink1.14/

    包含如下的jar

    flink-csv-1.14.6.jar
    flink-dist_2.12-1.14.6.jar
    flink-json-1.14.6.jar
    flink-shaded-zookeeper-3.4.14.jar
    flink-sql-connector-elasticsearch7_2.12-1.14.6.jar
    flink-sql-connector-kafka_2.12-1.14.6.jar
    flink-sql-connector-mysql-cdc-2.3.0.jar

    完成后启动

    sh auto.sh start 1.14

启动完成后访问 http://192.168.XXX.XXX:8899/ 通过admin/admin即可登录进系统进行可视化操作。

三、任务创建提交

场景说明：目前 MySQL 中有四张表，t1、t2、t3、t4。需要将 t1、t2 关联查询后结果同步到 Elasticsearch 索引 search01_index，t3、t4 关联查询后结果也同步到Elasticsearch索引 search01_index。Elasticsearch 索引信息是通过Spring Elasticsearch Data 项目启动时自动生成好了。只需要将数据同步即可。另外推送 Kafka 供给其他源处理。

Flink SQL和数据库字段类型对照说明：https://nightlies.apache.org/flink/flink-docs-release-1.14/docs/connectors/table/jdbc/#data-type-mapping

MySQL 同步到 Elasticsearch
复制代码 
### 进入flink/bin 目录下，通过 ./ 命令启动sql客户端工具                 
Flink SQL> SET execution.checkpointing.interval = 3s;

### source 1
Flink SQL> CREATE TABLE IF NOT EXISTS test1 (
    id STRING,
    name STRING,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test1的schame',
    'table-name' = 'test1'
);

### source 2
Flink SQL> CREATE TABLE IF NOT EXISTS test2 (
    id   STRING,
    doc_id STRING,
    event_detail STRING,
    country STRING,
    `time` STRING,
    entity_type STRING,
    entity_name STRING,
    type TINYINT,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test2的schame',
    'table-name' = 'test2'
);

### source 3
Flink SQL>CREATE TABLE IF NOT EXISTS test3 (
    id STRING,
    title STRING,
    type TINYINT,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test3的schame',
    'table-name' = 'test3'
);

### source 4
Flink SQL>CREATE TABLE IF NOT EXISTS test4 (
     id STRING,
     `time` STRING,
     country STRING,
     entity_type STRING,
     entity_name STRING,
     detail STRING,
     fo_id STRING,
     del_flag TINYINT,
     PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test4的schame',
    'table-name' = 'test4'
);

### sink
Flink SQL>CREATE TABLE IF NOT EXISTS search01_index (
    id           STRING,
    name         STRING,
    eventDetail STRING,
    country      STRING,
    `time`       STRING,
    entityType  STRING,
    entityName  STRING,
    type         TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'elasticsearch-7',
    'hosts' = 'http://192.168.XXX.XXX:9200',
    'index' = 'search01_index',
    'sink.bulk-flush.max-actions' = '1'
);

### transform 1 ，防止进同一个索引时表 t1 和 t3 的 ID 相同，将 ID 进行拼接处理。
Flink SQL>INSERT INTO search01_index
    SELECT CONCAT('doc_', , '_', ) AS id,
    ,
    t2.event_detail                   AS eventDetail,
    t2.country,
    t2.`time`,
    t2.entity_type                    AS entityType,
    t2.entity_name                    AS entityName,
    t2.type
    FROM test1 t1
    LEFT JOIN test2 t2 ON  = t2.doc_id
    WHERE t1.del_flag = 0
    AND t2.del_flag = 0;
    
### transform 2 ，防止进同一个索引时表 t1 和 t3 的 ID 相同，将 ID 进行拼接处理。
Flink SQL>INSERT INTO search01_index
    SELECT CONCAT('fo_', , '_', ) AS id,
    t3.title                         AS name,
    t4.detail                        AS eventDetail,
    t4.country,
    t4.`time`,
    t4.entity_type                   AS entityType,
    t4.entity_name                   AS entityName,
    t3.type
    FROM test3 t3
    LEFT JOIN test4 t4 ON  = t4.fo_id
    WHERE t3.del_flag = 0
    AND t4.del_flag = 0;

执行完成后访问Flink UI http://192.168.XXX.XXX:8888/#/job/running 可以看到正在运行的

Running Jobs，然后对MySQL 表数据进行增删改操作，可以看到 Elasticsearch 中数据同步效果。

MySQL 同步到 Kafka
复制代码 
### 进入flink/bin 目录下，通过 ./ 命令启动sql客户端工具                 
Flink SQL> SET execution.checkpointing.interval = 3s;

### source 1
Flink SQL> CREATE TABLE IF NOT EXISTS test1 (
    id STRING,
    name STRING,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test1的schame',
    'table-name' = 'test1'
);

### source 2
Flink SQL> CREATE TABLE IF NOT EXISTS test2 (
    id   STRING,
    doc_id STRING,
    event_detail STRING,
    country STRING,
    `time` STRING,
    entity_type STRING,
    entity_name STRING,
    type TINYINT,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test2的schame',
    'table-name' = 'test2'
);

### source 3
Flink SQL>CREATE TABLE IF NOT EXISTS test3 (
    id STRING,
    title STRING,
    type TINYINT,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test3的schame',
    'table-name' = 'test3'
);

### source 4
Flink SQL>CREATE TABLE IF NOT EXISTS test4 (
     id STRING,
     `time` STRING,
     country STRING,
     entity_type STRING,
     entity_name STRING,
     detail STRING,
     fo_id STRING,
     del_flag TINYINT,
     PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test4的schame',
    'table-name' = 'test4'
);

### sink
Flink SQL>CREATE TABLE IF NOT EXISTS test_kakfa (
    id           STRING,
    name         STRING,
    eventDetail STRING,
    country      STRING,
    `time`       STRING,
    entityType  STRING,
    entityName  STRING,
    type         TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'kafka',
    'topic' = 'test_kakfa',
    'properties.bootstrap.servers' = '192.168.XXX.XXX:9092',
    '' = 'testGroup',
    'scan.startup.mode' = 'earliest-offset',
    'format' = 'debezium-json',
    'debezium-json.ignore-parse-errors'='true'
);

### transform 1 ，防止进同一个索引时表 t1 和 t3 的 ID 相同，将 ID 进行拼接处理。
Flink SQL>INSERT INTO test_kakfa
    SELECT CONCAT('doc_', , '_', ) AS id,
    ,
    t2.event_detail                   AS eventDetail,
    t2.country,
    t2.`time`,
    t2.entity_type                    AS entityType,
    t2.entity_name                    AS entityName,
    t2.type
    FROM test1 t1
    LEFT JOIN test2 t2 ON  = t2.doc_id
    WHERE t1.del_flag = 0
    AND t2.del_flag = 0;
    
### transform 2 ，防止进同一个索引时表 t1 和 t3 的 ID 相同，将 ID 进行拼接处理。
Flink SQL>INSERT INTO test_kakfa
    SELECT CONCAT('fo_', , '_', ) AS id,
    t3.title                         AS name,
    t4.detail                        AS eventDetail,
    t4.country,
    t4.`time`,
    t4.entity_type                   AS entityType,
    t4.entity_name                   AS entityName,
    t3.type
    FROM test3 t3
    LEFT JOIN test4 t4 ON  = t4.fo_id
    WHERE t3.del_flag = 0
    AND t4.del_flag = 0;

执行完成后访问 Kafka-map UI 可以看到数据已经进入Kafka，数据格式为：

复制代码 
{"before":null,"after":{"id":"fo_48_50","name":"标题啊","eventDetail":"内容2","country":"日本","time":"2023-06-01","entityType":"实体2","entityName":"名称2","type":3},"op":"c"}

2. 通过 Dinky 方式

登录Dinky可视化界面后选择【注册中心】菜单添加flink的环境

然后选择【数据开发】菜单配置SQL任务

  • mysql2kafka-1 内容

    复制代码 
    CREATE TABLE IF NOT EXISTS test1 (
    id STRING,
    name STRING,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test1的schame',
    'table-name' = 'test1'
);
 CREATE TABLE IF NOT EXISTS test2 (
    id   STRING,
    doc_id STRING,
    event_detail STRING,
    country STRING,
    `time` STRING,
    entity_type STRING,
    entity_name STRING,
    type TINYINT,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test2的schame',
    'table-name' = 'test2'
);

CREATE TABLE IF NOT EXISTS test_kakfa (
    id           STRING,
    name         STRING,
    eventDetail STRING,
    country      STRING,
    `time`       STRING,
    entityType  STRING,
    entityName  STRING,
    type         TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'kafka',
    'topic' = 'test_kakfa',
    'properties.bootstrap.servers' = '192.168.XXX.XXX:9092',
    '' = 'testGroup',
    'scan.startup.mode' = 'earliest-offset',
    'format' = 'debezium-json',
    'debezium-json.ignore-parse-errors'='true'
);
INSERT INTO test_kakfa
SELECT CONCAT('doc_', , '_', ) AS id,
    ,
    t2.event_detail                   AS eventDetail,
    t2.country,
    t2.`time`,
    t2.entity_type                    AS entityType,
    t2.entity_name                    AS entityName,
    t2.type
FROM test1 t1
LEFT JOIN test2 t2 ON  = t2.doc_id
WHERE t1.del_flag = 0
AND t2.del_flag = 0;

  • mysql2kafka-2 内容

    复制代码 
    CREATE TABLE IF NOT EXISTS test3 (
    id STRING,
    title STRING,
    type TINYINT,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test3的schame',
    'table-name' = 'test3'
);
CREATE TABLE IF NOT EXISTS test4 (
     id STRING,
     `time` STRING,
     country STRING,
     entity_type STRING,
     entity_name STRING,
     detail STRING,
     fo_id STRING,
     del_flag TINYINT,
     PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test4的schame',
    'table-name' = 'test4'
);
CREATE TABLE IF NOT EXISTS test_kakfa (
    id           STRING,
    name         STRING,
    eventDetail STRING,
    country      STRING,
    `time`       STRING,
    entityType  STRING,
    entityName  STRING,
    type         TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'kafka',
    'topic' = 'test_kakfa',
    'properties.bootstrap.servers' = '192.168.6.36:9092',
    '' = 'testGroup',
    'scan.startup.mode' = 'earliest-offset',
    'format' = 'debezium-json',
    'debezium-json.ignore-parse-errors'='true'
);
INSERT INTO test_kakfa
SELECT CONCAT('fo_', , '_', ) AS id,
    t3.title                         AS name,
    t4.detail                        AS eventDetail,
    t4.country,
    t4.`time`,
    t4.entity_type                   AS entityType,
    t4.entity_name                   AS entityName,
    t3.type
FROM test3 t3
LEFT JOIN test4 t4 ON  = t4.fo_id
WHERE t3.del_flag = 0
AND t4.del_flag = 0;

  • mysql2es-1 内容

    复制代码 
    CREATE TABLE IF NOT EXISTS test1 (
    id STRING,
    name STRING,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test1的schame',
    'table-name' = 'test1'
);
 CREATE TABLE IF NOT EXISTS test2 (
    id   STRING,
    doc_id STRING,
    event_detail STRING,
    country STRING,
    `time` STRING,
    entity_type STRING,
    entity_name STRING,
    type TINYINT,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test2的schame',
    'table-name' = 'test2'
);
CREATE TABLE IF NOT EXISTS search01_index (
    id           STRING,
    name         STRING,
    eventDetail STRING,
    country      STRING,
    `time`       STRING,
    entityType  STRING,
    entityName  STRING,
    type         TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'elasticsearch-7',
    'hosts' = 'http://192.168.XXX.XXX:9200',
    'index' = 'search01_index',
    'sink.bulk-flush.max-actions' = '1'
);
INSERT INTO search01_index
SELECT CONCAT('doc_', , '_', ) AS id,
    ,
    t2.event_detail                   AS eventDetail,
    t2.country,
    t2.`time`,
    t2.entity_type                    AS entityType,
    t2.entity_name                    AS entityName,
    t2.type
FROM test1 t1
LEFT JOIN test2 t2 ON  = t2.doc_id
WHERE t1.del_flag = 0
AND t2.del_flag = 0;

  • mysql2es-2 内容

    复制代码 
    CREATE TABLE IF NOT EXISTS test3 (
    id STRING,
    title STRING,
    type TINYINT,
    del_flag TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test3的schame',
    'table-name' = 'test3'
);
CREATE TABLE IF NOT EXISTS test4 (
     id STRING,
     `time` STRING,
     country STRING,
     entity_type STRING,
     entity_name STRING,
     detail STRING,
     fo_id STRING,
     del_flag TINYINT,
     PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'mysql-cdc',
    'hostname' = '192.168.XXX.XXX',
    'port' = '3306',
    'username' = 'root',
    'password' = '123456',
    'database-name' = 'test4的schame',
    'table-name' = 'test4'
);
CREATE TABLE IF NOT EXISTS search01_index (
    id           STRING,
    name         STRING,
    eventDetail STRING,
    country      STRING,
    `time`       STRING,
    entityType  STRING,
    entityName  STRING,
    type         TINYINT,
    PRIMARY KEY (id) NOT ENFORCED
) WITH (
    'connector' = 'elasticsearch-7',
    'hosts' = 'http://192.168.XXX.XXX:9200',
    'index' = 'search01_index',
    'sink.bulk-flush.max-actions' = '1'
);
INSERT INTO search01_index
SELECT CONCAT('fo_', , '_', ) AS id,
    t3.title                         AS name,
    t4.detail                        AS eventDetail,
    t4.country,
    t4.`time`,
    t4.entity_type                   AS entityType,
    t4.entity_name                   AS entityName,
    t3.type
FROM test3 t3
LEFT JOIN test4 t4 ON  = t4.fo_id
WHERE t3.del_flag = 0
AND t4.del_flag = 0;
  • 分别点击任务运行后，查看Flink UI
  • 查看Kafka-map
  • 查看Elasticsearch Head 插件

至此就大功告成了。

相关推荐
南墙上的石头6 小时前
麒麟 V10 重装人大金仓 V8R6 踩坑实录（含 MySQL 兼容模式）
数据库·mysql
执子手 吹散苍茫茫烟波8 小时前
RC 隔离级别下 MySQL InnoDB 死锁典型案例
数据库·mysql
峥无11 小时前
深入理解MySQL事务与MVCC机制
数据库·mysql
Elastic 中国社区官方博客14 小时前
跟踪资金流向：使用 ES|QL 和跨集群搜索追踪洗钱网络
大数据·人工智能·安全·elasticsearch·搜索引擎·金融·全文检索
要开心吖ZSH14 小时前
MVCC 进阶：快照读 vs 当前读、幻读与 Next-Key Lock
java·数据库·sql·mysql·mvcc
万亿少女的梦16815 小时前
基于Spring Boot的天空影院电影网站系统设计与实现
java·spring boot·mysql·vue·系统设计
万亿少女的梦16815 小时前
基于Spring Boot的社区管理系统设计与实现
java·spring boot·mysql·vue·系统设计
翔云12345615 小时前
简单概括主库上 Executed_Gtid_Set 是什么时候更新的
数据库·mysql
要开心吖ZSH15 小时前
Java事务与MySQL事务的关系及MVCC通俗解析
java·开发语言·mysql·mvcc
爱喝热水的呀哈喽16 小时前
hypermesh两个网格参数解析
服务器·数据库·mysql
热门推荐
01GitHub 镜像站点022026 年 AI 编程工具终极横评：Cursor vs Claude Code vs Copilot vs Windsurf03AI科技热点日报 | 2026年07月01日04【AI】2026 年具身智能模型和世界模型总结052026 AI 编程工具终极实战指南：Cursor vs Claude Code vs Copilot，开发者该怎么选？062026 年 AI 大模型 & AI 编程工具实战全总结072026年6月AI大模型全景报告：GPT-5.6、Claude Opus 4.8、Gemini 3.5，中美AI三足鼎立谁主沉浮？08Agnes AI 免费 API 接入指南：文本、生图、生视频，一套接口全免费09Claude Code、Codex、Cursor三分天下：2026年AI编程Agent生态全景剖析102026 国产 AI 大模型横评：DeepSeek、通义千问、Kimi、文心一言、星火、豆包谁更能打？