DolphinScheduler JSON拆解详解

本次活动邀请DolphinScheduler社区活跃贡献者，开源积极分子，现就职于政采云大数据部门，从事大数据平台架构工作的李进勇同学给大家分享相关内容。

同时也特别感谢示说网对本次直播活动的大力支持。

PS：本章内容篇幅较长，请大家耐心阅读。

为什么拆解json

在DolphinScheduler 1.3.x及以前的工作流中的任务及任务关系保存时是以大json的方式保存到数据库中process_definiton表的process_definition_json字段，如果某个工作流很大比如有100或者1000个任务，这个json字段也就非常大，在使用时需要解析json，非常耗费性能，且任务没法重用；基于大json，在工作流版本及任务版本上也没有好的实现方案，否则会导致数据大量冗余。

故社区计划启动json拆解项目，实现的需求目标：

• 大json完全拆分

• 新增工作流及任务版本

• 引入全局唯一键(code)

如何设计拆解后的表

3.6版本工作流

比如在当前1.3.6版本创建个a-->b的工作流以下是processDefiniton save 接口在controller入口打印的入参日志

create  process definition, project name: hadoop, process definition name: ab, process_definition_json: {"globalParams":[],"tasks":[{"type":"SHELL","id":"tasks-77643","name":"a","params":{"resourceList":[],"localParams":[{"prop":"yesterday","direct":"IN","type":"VARCHAR","value":"${system.biz.date}"}],"rawScript":"echo ${yesterday}"},"description":"","timeout":{"strategy":"","interval":null,"enable":false},"runFlag":"NORMAL","conditionResult":{"successNode":[""],"failedNode":[""]},"dependence":{},"maxRetryTimes":"0","retryInterval":"1","taskInstancePriority":"MEDIUM","workerGroup":"default","preTasks":[]},{"type":"SHELL","id":"tasks-99814","name":"b","params":{"resourceList":[],"localParams":[{"prop":"today","direct":"IN","type":"VARCHAR","value":"${system.biz.curdate}"}],"rawScript":"echo ${today}"},"description":"","timeout":{"strategy":"","interval":null,"enable":false},"runFlag":"NORMAL","conditionResult":{"successNode":[""],"failedNode":[""]},"dependence":{},"maxRetryTimes":"0","retryInterval":"1","taskInstancePriority":"MEDIUM","workerGroup":"default","preTasks":["a"]}],"tenantId":1,"timeout":0}, desc:  locations:{"tasks-77643":{"name":"a","targetarr":"","nodenumber":"1","x":251,"y":166},"tasks-99814":{"name":"b","targetarr":"tasks-77643","nodenumber":"0","x":533,"y":161}}, connects:[{"endPointSourceId":"tasks-77643","endPointTargetId":"tasks-99814"}]

依赖节点的工作流，dep是依赖节点

以下是processDefiniton save 接口在controller入口打印的入参日志

 create  process definition, project name: hadoop, process definition name: dep_c, process_definition_json: {"globalParams":[],"tasks":[{"type":"SHELL","id":"tasks-69503","name":"c","params":{"resourceList":[],"localParams":[],"rawScript":"echo 11"},"description":"","timeout":{"strategy":"","interval":null,"enable":false},"runFlag":"NORMAL","conditionResult":{"successNode":[""],"failedNode":[""]},"dependence":{},"maxRetryTimes":"0","retryInterval":"1","taskInstancePriority":"MEDIUM","workerGroup":"default","preTasks":["dep"]},{"type":"DEPENDENT","id":"tasks-22756","name":"dep","params":{},"description":"","timeout":{"strategy":"","interval":null,"enable":false},"runFlag":"NORMAL","conditionResult":{"successNode":[""],"failedNode":[""]},"dependence":{"relation":"AND","dependTaskList":[{"relation":"AND","dependItemList":[{"projectId":1,"definitionId":1,"depTasks":"b","cycle":"day","dateValue":"today"}]}]},"maxRetryTimes":"0","retryInterval":"1","taskInstancePriority":"MEDIUM","workerGroup":"default","preTasks":[]}],"tenantId":1,"timeout":0}, desc:  locations:{"tasks-69503":{"name":"c","targetarr":"tasks-22756","nodenumber":"0","x":597,"y":166},"tasks-22756":{"name":"dep","targetarr":"","nodenumber":"1","x":308,"y":164}}, connects:[{"endPointSourceId":"tasks-22756","endPointTargetId":"tasks-69503"}]

条件判断的工作流

以下是processDefiniton save 接口在controller入口打印的入参日志

create  process definition, project name: hadoop, process definition name: condition_test, process_definition_json: {"globalParams":[],"tasks":[{"type":"SHELL","id":"tasks-68456","name":"d","params":{"resourceList":[],"localParams":[],"rawScript":"echo 11"},"description":"","timeout":{"strategy":"","interval":null,"enable":false},"runFlag":"NORMAL","conditionResult":{"successNode":[""],"failedNode":[""]},"dependence":{},"maxRetryTimes":"0","retryInterval":"1","taskInstancePriority":"MEDIUM","workerGroup":"default","preTasks":[]},{"type":"SHELL","id":"tasks-58183","name":"e","params":{"resourceList":[],"localParams":[],"rawScript":"echo 22"},"description":"","timeout":{"strategy":"","interval":null,"enable":false},"runFlag":"NORMAL","conditionResult":{"successNode":[""],"failedNode":[""]},"dependence":{},"maxRetryTimes":"0","retryInterval":"1","taskInstancePriority":"MEDIUM","workerGroup":"default","preTasks":["cond"]},{"type":"SHELL","id":"tasks-43996","name":"f","params":{"resourceList":[],"localParams":[],"rawScript":"echo 33"},"description":"","timeout":{"strategy":"","interval":null,"enable":false},"runFlag":"NORMAL","conditionResult":{"successNode":[""],"failedNode":[""]},"dependence":{},"maxRetryTimes":"0","retryInterval":"1","taskInstancePriority":"MEDIUM","workerGroup":"default","preTasks":["cond"]},{"type":"CONDITIONS","id":"tasks-38972","name":"cond","params":{},"description":"","timeout":{"strategy":"","interval":null,"enable":false},"runFlag":"NORMAL","conditionResult":{"successNode":["e"],"failedNode":["f"]},"dependence":{"relation":"AND","dependTaskList":[{"relation":"AND","dependItemList":[{"depTasks":"d","status":"SUCCESS"}]}]},"maxRetryTimes":"0","retryInterval":"1","taskInstancePriority":"MEDIUM","workerGroup":"default","preTasks":["d"]}],"tenantId":1,"timeout":0}, desc:  locations:{"tasks-68456":{"name":"d","targetarr":"","nodenumber":"1","x":168,"y":158},"tasks-58183":{"name":"e","targetarr":"tasks-38972","nodenumber":"0","x":573,"y":82},"tasks-43996":{"name":"f","targetarr":"tasks-38972","nodenumber":"0","x":591,"y":288},"tasks-38972":{"name":"cond","targetarr":"tasks-68456","nodenumber":"2","x":382,"y":175}}, connects:[{"endPointSourceId":"tasks-68456","endPointTargetId":"tasks-38972"},{"endPointSourceId":"tasks-38972","endPointTargetId":"tasks-58183"},{"endPointSourceId":"tasks-38972","endPointTargetId":"tasks-43996"}]

从以上三个案例中，我们知道controller的入口参数的每个参数都可以在t_ds_process_definition表中找到对应，故表中数据如下图

拆解后的表设计思路

工作流只是dag的展现形式，任务通过工作流进行组织，组织的同时存在了任务之间的关系，也就是依赖。就好比一个画板，画板上有些图案，工作流就是画板，图案就是任务，图案之间的关系就是依赖。而调度的核心是调度任务，依赖只是表述调度的先后顺序。当前定时还是对整个工作流进行的定时，拆解后就方便对单独任务进行调度。正是基于这个思想设计了拆解的思路，所以这就需要三张表，工作流定义表、任务定义表、任务关系表。

• 工作流定义表：描述工作流的基本信息，比如全局参数、dag中节点的位置信息

• 任务定义表：描述任务的详情信息，比如任务类别、任务容错信息、优先级等

• 任务关系表：描述任务的关系信息，比如当前节点、前置节点等

基于这个设计思想再扩展到版本，无非是对于这三张表，每张表新增个保存版本的日志表。

工作流定义表

现在看案例中save接口日志，现有字段(project、process_definition_name、desc、locations、connects)，对于json中除了task之外的还剩下

{"globalParams":[],"tenantId":1,"timeout":0}

所以可知工作流定义表：

CREATE TABLE `t_ds_process_definition` (
  `id` int(11) NOT NULL AUTO_INCREMENT COMMENT 'self-increasing id',
  `code` bigint(20) NOT NULL COMMENT 'encoding',
  `name` varchar(200) DEFAULT NULL COMMENT 'process definition name',
  `version` int(11) DEFAULT NULL COMMENT 'process definition version',
  `description` text COMMENT 'description',
  `project_code` bigint(20) NOT NULL COMMENT 'project code',
  `release_state` tinyint(4) DEFAULT NULL COMMENT 'process definition release state：0:offline,1:online',
  `user_id` int(11) DEFAULT NULL COMMENT 'process definition creator id',
  `global_params` text COMMENT 'global parameters',
  `flag` tinyint(4) DEFAULT NULL COMMENT '0 not available, 1 available',
  `locations` text COMMENT 'Node location information',
  `connects` text COMMENT 'Node connection information',
  `warning_group_id` int(11) DEFAULT NULL COMMENT 'alert group id',
  `timeout` int(11) DEFAULT '0' COMMENT 'time out, unit: minute',
  `tenant_id` int(11) NOT NULL DEFAULT '-1' COMMENT 'tenant id',
  `create_time` datetime NOT NULL COMMENT 'create time',
  `update_time` datetime DEFAULT NULL COMMENT 'update time',
  PRIMARY KEY (`id`,`code`),
  UNIQUE KEY `process_unique` (`name`,`project_code`) USING BTREE
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

CREATE TABLE `t_ds_process_definition_log` (
  `id` int(11) NOT NULL AUTO_INCREMENT COMMENT 'self-increasing id',
  `code` bigint(20) NOT NULL COMMENT 'encoding',
  `name` varchar(200) DEFAULT NULL COMMENT 'process definition name',
  `version` int(11) DEFAULT NULL COMMENT 'process definition version',
  `description` text COMMENT 'description',
  `project_code` bigint(20) NOT NULL COMMENT 'project code',
  `release_state` tinyint(4) DEFAULT NULL COMMENT 'process definition release state：0:offline,1:online',
  `user_id` int(11) DEFAULT NULL COMMENT 'process definition creator id',
  `global_params` text COMMENT 'global parameters',
  `flag` tinyint(4) DEFAULT NULL COMMENT '0 not available, 1 available',
  `locations` text COMMENT 'Node location information',
  `connects` text COMMENT 'Node connection information',
  `warning_group_id` int(11) DEFAULT NULL COMMENT 'alert group id',
  `timeout` int(11) DEFAULT '0' COMMENT 'time out,unit: minute',
  `tenant_id` int(11) NOT NULL DEFAULT '-1' COMMENT 'tenant id',
  `operator` int(11) DEFAULT NULL COMMENT 'operator user id',
  `operate_time` datetime DEFAULT NULL COMMENT 'operate time',
  `create_time` datetime NOT NULL COMMENT 'create time',
  `update_time` datetime DEFAULT NULL COMMENT 'update time',
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

从表字段可以看出 日志表仅仅比主表多了两个字段operator、operate_time

任务定义表

案例中ab工作流task的json

  "tasks": [{
    "type": "SHELL",
    "id": "tasks-77643",
    "name": "a",
    "params": {
      "resourceList": [],
      "localParams": [{
        "prop": "yesterday",
        "direct": "IN",
        "type": "VARCHAR",
        "value": "${system.biz.date}"
      }],
      "rawScript": "echo ${yesterday}"
    },
    "description": "",
    "timeout": {
      "strategy": "",
      "interval": null,
      "enable": false
    },
    "runFlag": "NORMAL",
    "conditionResult": {
      "successNode": [""],
      "failedNode": [""]
    },
    "dependence": {},
    "maxRetryTimes": "0",
    "retryInterval": "1",
    "taskInstancePriority": "MEDIUM",
    "workerGroup": "default",
    "preTasks": []
  }, {
    "type": "SHELL",
    "id": "tasks-99814",
    "name": "b",
    "params": {
      "resourceList": [],
      "localParams": [{
        "prop": "today",
        "direct": "IN",
        "type": "VARCHAR",
        "value": "${system.biz.curdate}"
      }],
      "rawScript": "echo ${today}"
    },
    "description": "",
    "timeout": {
      "strategy": "",
      "interval": null,
      "enable": false
    },
    "runFlag": "NORMAL",
    "conditionResult": {
      "successNode": [""],
      "failedNode": [""]
    },
    "dependence": {},
    "maxRetryTimes": "0",
    "retryInterval": "1",
    "taskInstancePriority": "MEDIUM",
    "workerGroup": "default",
    "preTasks": ["a"]
  }]

dep_c工作流task的json

  "tasks": [{
    "type": "SHELL",
    "id": "tasks-69503",
    "name": "c",
    "params": {
      "resourceList": [],
      "localParams": [],
      "rawScript": "echo 11"
    },
    "description": "",
    "timeout": {
      "strategy": "",
      "interval": null,
      "enable": false
    },
    "runFlag": "NORMAL",
    "conditionResult": {
      "successNode": [""],
      "failedNode": [""]
    },
    "dependence": {},
    "maxRetryTimes": "0",
    "retryInterval": "1",
    "taskInstancePriority": "MEDIUM",
    "workerGroup": "default",
    "preTasks": ["dep"]
  }, {
    "type": "DEPENDENT",
    "id": "tasks-22756",
    "name": "dep",
    "params": {},
    "description": "",
    "timeout": {
      "strategy": "",
      "interval": null,
      "enable": false
    },
    "runFlag": "NORMAL",
    "conditionResult": {
      "successNode": [""],
      "failedNode": [""]
    },
    "dependence": {
      "relation": "AND",
      "dependTaskList": [{
        "relation": "AND",
        "dependItemList": [{
          "projectId": 1,
          "definitionId": 1,
          "depTasks": "b",
          "cycle": "day",
          "dateValue": "today"
        }]
      }]
    },
    "maxRetryTimes": "0",
    "retryInterval": "1",
    "taskInstancePriority": "MEDIUM",
    "workerGroup": "default",
    "preTasks": []
  }]

condition_test工作流task的json

  "tasks": [{
    "type": "SHELL",
    "id": "tasks-68456",
    "name": "d",
    "params": {
      "resourceList": [],
      "localParams": [],
      "rawScript": "echo 11"
    },
    "description": "",
    "timeout": {
      "strategy": "",
      "interval": null,
      "enable": false
    },
    "runFlag": "NORMAL",
    "conditionResult": {
      "successNode": [""],
      "failedNode": [""]
    },
    "dependence": {},
    "maxRetryTimes": "0",
    "retryInterval": "1",
    "taskInstancePriority": "MEDIUM",
    "workerGroup": "default",
    "preTasks": []
  }, {
    "type": "SHELL",
    "id": "tasks-58183",
    "name": "e",
    "params": {
      "resourceList": [],
      "localParams": [],
      "rawScript": "echo 22"
    },
    "description": "",
    "timeout": {
      "strategy": "",
      "interval": null,
      "enable": false
    },
    "runFlag": "NORMAL",
    "conditionResult": {
      "successNode": [""],
      "failedNode": [""]
    },
    "dependence": {},
    "maxRetryTimes": "0",
    "retryInterval": "1",
    "taskInstancePriority": "MEDIUM",
    "workerGroup": "default",
    "preTasks": ["cond"]
  }, {
    "type": "SHELL",
    "id": "tasks-43996",
    "name": "f",
    "params": {
      "resourceList": [],
      "localParams": [],
      "rawScript": "echo 33"
    },
    "description": "",
    "timeout": {
      "strategy": "",
      "interval": null,
      "enable": false
    },
    "runFlag": "NORMAL",
    "conditionResult": {
      "successNode": [""],
      "failedNode": [""]
    },
    "dependence": {},
    "maxRetryTimes": "0",
    "retryInterval": "1",
    "taskInstancePriority": "MEDIUM",
    "workerGroup": "default",
    "preTasks": ["cond"]
  }, {
    "type": "CONDITIONS",
    "id": "tasks-38972",
    "name": "cond",
    "params": {},
    "description": "",
    "timeout": {
      "strategy": "",
      "interval": null,
      "enable": false
    },
    "runFlag": "NORMAL",
    "conditionResult": {
      "successNode": ["e"],
      "failedNode": ["f"]
    },
    "dependence": {
      "relation": "AND",
      "dependTaskList": [{
        "relation": "AND",
        "dependItemList": [{
          "depTasks": "d",
          "status": "SUCCESS"
        }]
      }]
    },
    "maxRetryTimes": "0",
    "retryInterval": "1",
    "taskInstancePriority": "MEDIUM",
    "workerGroup": "default",
    "preTasks": ["d"]
  }]

从案例中可以知道SHELL/DEPENDENT/CONDITIONS类型的节点的json构成（其他任务类似SHELL），preTasks标识前置依赖节点。conditionResult结构比较固定，而dependence结构复杂，DEPENDENT和CONDITIONS类型任务的dependence结构还不一样，所以为了统一，我们将conditionResult和dependence整体放到params中，params对应表字段的task_params。

这样我们就确定了t_ds_task_definition表

CREATE TABLE `t_ds_task_definition` (
  `id` int(11) NOT NULL AUTO_INCREMENT COMMENT 'self-increasing id',
  `code` bigint(20) NOT NULL COMMENT 'encoding',
  `name` varchar(200) DEFAULT NULL COMMENT 'task definition name',
  `version` int(11) DEFAULT NULL COMMENT 'task definition version',
  `description` text COMMENT 'description',
  `project_code` bigint(20) NOT NULL COMMENT 'project code',
  `user_id` int(11) DEFAULT NULL COMMENT 'task definition creator id',
  `task_type` varchar(50) NOT NULL COMMENT 'task type',
  `task_params` text COMMENT 'job custom parameters',
  `flag` tinyint(2) DEFAULT NULL COMMENT '0 not available, 1 available',
  `task_priority` tinyint(4) DEFAULT NULL COMMENT 'job priority',
  `worker_group` varchar(200) DEFAULT NULL COMMENT 'worker grouping',
  `fail_retry_times` int(11) DEFAULT NULL COMMENT 'number of failed retries',
  `fail_retry_interval` int(11) DEFAULT NULL COMMENT 'failed retry interval',
  `timeout_flag` tinyint(2) DEFAULT '0' COMMENT 'timeout flag:0 close, 1 open',
  `timeout_notify_strategy` tinyint(4) DEFAULT NULL COMMENT 'timeout notification policy: 0 warning, 1 fail',
  `timeout` int(11) DEFAULT '0' COMMENT 'timeout length,unit: minute',
  `delay_time` int(11) DEFAULT '0' COMMENT 'delay execution time,unit: minute',
  `resource_ids` varchar(255) DEFAULT NULL COMMENT 'resource id, separated by comma',
  `create_time` datetime NOT NULL COMMENT 'create time',
  `update_time` datetime DEFAULT NULL COMMENT 'update time',
  PRIMARY KEY (`id`,`code`),
  UNIQUE KEY `task_unique` (`name`,`project_code`) USING BTREE
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

CREATE TABLE `t_ds_task_definition_log` (
  `id` int(11) NOT NULL AUTO_INCREMENT COMMENT 'self-increasing id',
  `code` bigint(20) NOT NULL COMMENT 'encoding',
  `name` varchar(200) DEFAULT NULL COMMENT 'task definition name',
  `version` int(11) DEFAULT NULL COMMENT 'task definition version',
  `description` text COMMENT 'description',
  `project_code` bigint(20) NOT NULL COMMENT 'project code',
  `user_id` int(11) DEFAULT NULL COMMENT 'task definition creator id',
  `task_type` varchar(50) NOT NULL COMMENT 'task type',
  `task_params` text COMMENT 'job custom parameters',
  `flag` tinyint(2) DEFAULT NULL COMMENT '0 not available, 1 available',
  `task_priority` tinyint(4) DEFAULT NULL COMMENT 'job priority',
  `worker_group` varchar(200) DEFAULT NULL COMMENT 'worker grouping',
  `fail_retry_times` int(11) DEFAULT NULL COMMENT 'number of failed retries',
  `fail_retry_interval` int(11) DEFAULT NULL COMMENT 'failed retry interval',
  `timeout_flag` tinyint(2) DEFAULT '0' COMMENT 'timeout flag:0 close, 1 open',
  `timeout_notify_strategy` tinyint(4) DEFAULT NULL COMMENT 'timeout notification policy: 0 warning, 1 fail',
  `timeout` int(11) DEFAULT '0' COMMENT 'timeout length,unit: minute',
  `delay_time` int(11) DEFAULT '0' COMMENT 'delay execution time,unit: minute',
  `resource_ids` varchar(255) DEFAULT NULL COMMENT 'resource id, separated by comma',
  `operator` int(11) DEFAULT NULL COMMENT 'operator user id',
  `operate_time` datetime DEFAULT NULL COMMENT 'operate time',
  `create_time` datetime NOT NULL COMMENT 'create time',
  `update_time` datetime DEFAULT NULL COMMENT 'update time',
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

//注意：dev版本和1.3.6版本区别，dev版本已将description换成desc，并且新增了delayTime

{
  "globalParams": [],
  "tasks": [{
      "type": "SHELL",
      "id": "tasks-18200",
      "name": "d",
      "code": "",
      "params": {
        "resourceList": [],
        "localParams": [],
        "rawScript": "echo 5"
      },
      "desc": "",
      "runFlag": "NORMAL",
      "conditionResult": {
        "successNode": [
          ""
        ],
        "failedNode": [
          ""
        ]
      },
      "dependence": {},
      "maxRetryTimes": "0",
      "retryInterval": "1",
      "delayTime": "0",
      "timeout": {
        "strategy": "",
        "interval": null,
        "enable": false
      },
      "waitStartTimeout": {},
      "taskInstancePriority": "MEDIUM",
      "workerGroup": "hadoop",
      "preTasks": [],
      "depList": null
    },
    {
      "type": "SHELL",
      "id": "tasks-55225",
      "name": "e",
      "code": "",
      "params": {
        "resourceList": [],
        "localParams": [],
        "rawScript": "echo 6"
      },
      "desc": "",
      "runFlag": "NORMAL",
      "conditionResult": {
        "successNode": [
          ""
        ],
        "failedNode": [
          ""
        ]
      },
      "dependence": {},
      "maxRetryTimes": "0",
      "retryInterval": "1",
      "delayTime": "0",
      "timeout": {
        "strategy": "",
        "interval": null,
        "enable": false
      },
      "waitStartTimeout": {},
      "taskInstancePriority": "MEDIUM",
      "workerGroup": "hadoop",
      "preTasks": [
        "def"
      ],
      "depList": null
    },
    {
      "type": "SHELL",
      "id": "tasks-67639",
      "name": "f",
      "code": "",
      "params": {
        "resourceList": [],
        "localParams": [],
        "rawScript": "echo 7"
      },
      "desc": "",
      "runFlag": "NORMAL",
      "conditionResult": {
        "successNode": [
          ""
        ],
        "failedNode": [
          ""
        ]
      },
      "dependence": {},
      "maxRetryTimes": "0",
      "retryInterval": "1",
      "delayTime": "0",
      "timeout": {
        "strategy": "",
        "interval": null,
        "enable": false
      },
      "waitStartTimeout": {},
      "taskInstancePriority": "MEDIUM",
      "workerGroup": "hadoop",
      "preTasks": [
        "def"
      ],
      "depList": null
    },
    {
      "type": "CONDITIONS",
      "id": "tasks-67387",
      "name": "def",
      "code": "",
      "params": {},
      "desc": "",
      "runFlag": "NORMAL",
      "conditionResult": {
        "successNode": [
          "e"
        ],
        "failedNode": [
          "f"
        ]
      },
      "dependence": {
        "relation": "AND",
        "dependTaskList": [{
          "relation": "AND",
          "dependItemList": [{
              "depTasks": "d",
              "status": "SUCCESS"
            },
            {
              "depTasks": "d",
              "status": "FAILURE"
            }
          ]
        }]
      },
      "maxRetryTimes": "0",
      "retryInterval": "1",
      "delayTime": "0",
      "timeout": {
        "strategy": "",
        "interval": null,
        "enable": false
      },
      "waitStartTimeout": {},
      "taskInstancePriority": "MEDIUM",
      "workerGroup": "hadoop",
      "preTasks": [
        "d"
      ],
      "depList": null
    }
  ],
  "tenantId": 1,
  "timeout": 0
}

任务关系表

preTasks标识前置依赖节点，当前节点在关系表中使用postTask标识。由于当前节点肯定存在而前置节点不一定存在，所以post不可能为空，而preTask有可能为空

CREATE TABLE `t_ds_process_task_relation` (
  `id` int(11) NOT NULL AUTO_INCREMENT COMMENT 'self-increasing id',
  `name` varchar(200) DEFAULT NULL COMMENT 'relation name',
  `process_definition_version` int(11) DEFAULT NULL COMMENT 'process version',
  `project_code` bigint(20) NOT NULL COMMENT 'project code',
  `process_definition_code` bigint(20) NOT NULL COMMENT 'process code',
  `pre_task_code` bigint(20) NOT NULL COMMENT 'pre task code',
  `pre_task_version` int(11) NOT NULL COMMENT 'pre task version',
  `post_task_code` bigint(20) NOT NULL COMMENT 'post task code',
  `post_task_version` int(11) NOT NULL COMMENT 'post task version',
  `condition_type` tinyint(2) DEFAULT NULL COMMENT 'condition type : 0 none, 1 judge 2 delay',
  `condition_params` text COMMENT 'condition params(json)',
  `create_time` datetime NOT NULL COMMENT 'create time',
  `update_time` datetime DEFAULT NULL COMMENT 'update time',
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

CREATE TABLE `t_ds_process_task_relation_log` (
  `id` int(11) NOT NULL AUTO_INCREMENT COMMENT 'self-increasing id',
  `name` varchar(200) DEFAULT NULL COMMENT 'relation name',
  `process_definition_version` int(11) DEFAULT NULL COMMENT 'process version',
  `project_code` bigint(20) NOT NULL COMMENT 'project code',
  `process_definition_code` bigint(20) NOT NULL COMMENT 'process code',
  `pre_task_code` bigint(20) NOT NULL COMMENT 'pre task code',
  `pre_task_version` int(11) NOT NULL COMMENT 'pre task version',
  `post_task_code` bigint(20) NOT NULL COMMENT 'post task code',
  `post_task_version` int(11) NOT NULL COMMENT 'post task version',
  `condition_type` tinyint(2) DEFAULT NULL COMMENT 'condition type : 0 none, 1 judge 2 delay',
  `condition_params` text COMMENT 'condition params(json)',
  `operator` int(11) DEFAULT NULL COMMENT 'operator user id',
  `operate_time` datetime DEFAULT NULL COMMENT 'operate time',
  `create_time` datetime NOT NULL COMMENT 'create time',
  `update_time` datetime DEFAULT NULL COMMENT 'update time',
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

对于依赖关系复杂的场景

API模块如何改造

api模块进行save操作时

• 通过雪花算法生成13位的数字作为process_definition_code，工作流定义保存至process_definition（主表）和process_definition_log（日志表），这两个表保存的是同样的数据，工作流定义版本为1

• 通过雪花算法生成13位的数字作为task_definition_code，任务定义表保存至task_definition（主表）和task_definition_log（日志表），也是保存同样的数据，任务定义版本为1

• 工作流任务关系保存在 process_task_relation（主表）和process_task_relation_log（日志表），该表保存的code和version是工作流的code和version，因为任务是通过工作流进行组织，以工作流来画dag。也是通过post_task_code和post_task_version标识dag的当前节点，这个节点的前置依赖通过pre_task_code和pre_task_version来标识，如果没有依赖，pre_task_code和pre_task_version为0

api模块进行update操作时，工作流定义和任务定义直接更新主表数据，更新后的数据insert到日志表。关系表主表先删除然后再插入新的关系，关系表日志表直接插入新的关系

api模块进行delete操作时，工作流定义、任务定义和关系表直接删除主表数据，日志表数据不变动

api模块进行switch操作时，直接将日志表中对应version数据覆盖到主表

数据交互如何改造

在json拆分一期api模块controller层整体未变动，传入的大json还是在service层映射为ProcessData对象。insert或update操作在公共Service模块通过ProcessService.saveProcessDefiniton()入口完成保存数据库操作，按照task_definition、process_task_relation、process_definition的顺序保存。保存时，如果该任务已经存在并且关联的工作流未上线，则更改任务；如果任务关联的工作流已上线，则不允许更改任务

api查询操作时，当前还是通过工作流id来查询，在公共Service模块通过ProcessService.genTaskNodeList()入口完成数据组装，还是组装为ProcessData对象，进而生成json返回

Server模块（Master）也是通过公共Service模块ProcessService.genTaskNodeList()获得TaskNodeList生成调度dag，把当前任务所有信息放到 MasterExecThread.readyToSubmitTaskQueue队列，以便生成taskInstance，dispatch给worker

当前json还需做什么

• controller对外restAPI接口改造

• ui模块dag改造

• ui模块新增task操作页面

processDefinition

备注：taskRelationJson格式：

[{"name":"","pre_task_code":0,"pre_task_version":0,"post_task_code":123456789,"post_task_version":1,"condition_type":0,"condition_params":{}},{"name":"","pre_task_code":123456789,"pre_task_version":1,"post_task_code":123451234,"post_task_version":1,"condition_type":0,"condition_params":{}}]

同理其他接口请求参数processDefinitionId换成code

schedule

taskDefinition(新增)

taskDefinitionJson：[{"name":"test","description":"","task_type":"SHELL","task_params":[],"flag":0,"task_priority":0,"worker_group":"default","fail_retry_times":0,"fail_retry_interval":0,"timeout_flag":0,"timeout_notify_strategy":0,"timeout":0,"delay_time":0,"resource_ids":""}]

对应需求issue

• [Feature][JsonSplit-api] api module controller design #5498

• [Feature][JsonSplit-api]processDefinition save/update interface  #5499

• [Feature][JsonSplit-api]processDefinition switch interface #5501

• [Feature][JsonSplit-api]processDefinition delete interface #5502

• [Feature][JsonSplit-api]processDefinition copy interface #5503

• [Feature][JsonSplit-api]processDefinition export interface #5504

• [Feature][JsonSplit-api]processDefinition list-paging interface #5505

• [Feature][JsonSplit-api]processDefinition move interface #5506

• [Feature][JsonSplit-api]processDefinition queryProcessDefinitionAllByProjectId interface #5507

• [Feature][JsonSplit-api]processDefinition select-by-id interface #5508

• [Feature][JsonSplit-api]processDefinition view-tree interface #5509

• [Feature][JsonSplit-api]schedule create interface #5510

• [Feature][JsonSplit-api]schedule list-paging interface #5511

• [Feature][JsonSplit-api]schedule update interface #5512

• [Feature][JsonSplit-api]taskDefinition save interface #5513

• [Feature][JsonSplit-api]taskDefinition update interface #5514

• [Feature][JsonSplit-api]taskDefinition switch interface #5515

• [Feature][JsonSplit-api]taskDefinition query interface #5516

• [Feature][JsonSplit-api]taskDefinition delete interface #5517

• [Feature][JsonSplit-api]WorkFlowLineage interface #5518

• [Feature][JsonSplit-api]analysis interface #5519

• [Feature][JsonSplit-api]executors interface #5520

• [Feature][JsonSplit-api]processInstance interface #5521

• [Feature][JsonSplit-api]project interface #5522

https://github.com/apache/dolphinscheduler/issues

Github issues

示说网

https://www.slidestalk.com/

点击阅读原文查看本次分享视频