saltfishlife opened a new issue #4092:
URL: https://github.com/apache/incubator-dolphinscheduler/issues/4092


   In addition to the community's friends, they all mentioned setting 
parameters and global parameters when starting the stream. In actual use, there 
are also project-level parameter requirements.
   
    For general parameters that are not easily modified, they can be set as 
global parameters or project-level parameters. Then use it directly in the task 
without additional definition. The parameters of the flow level should also 
have corresponding saving functions. 
   
   To sort out, the process of parsing parameter values ​​should be as follows: 
   
   1. Find the parameter value from the task level (the parameter value is set 
in the task definition interface); if the acquisition fails, go to the next 
step;
    2. Find the parameter value from the flow level (flow start The pop-up 
interface when the time is set can be set); if the acquisition fails, go to the 
next step; 
   3. Find the parameter value from the project level (there is a corresponding 
interface for management); if the acquisition fails, go to the next step; 
   4. Search from the global parameter level Parameter value (there is a 
corresponding interface for management); if the acquisition fails, then go to 
one step; 
   5. Parameter value search fails, the parameter value is empty, end.
   
   
--------------------------------------------------------------------------------------------------------------------------------------
   
   除了社区的小伙伴都有提到的在启动流时设置参数、全局参数。实际使用中,还有项目级别的参数需求。
   
   对于通用的,不轻易修改的参数,可以设置为全局参数、项目级别、流级别的参数。然后在任务中直接使用,不需要额外去定义。
   
   梳理一下,解析参数值时的流程应该如下:
   
   1.从任务级别寻找参数值(参数值在任务定义界面设置);获取失败,则转入下一步;
   2.从流级别寻找参数值(流启动时的弹出界面可设置);获取失败,则转入下一步;
   3.从项目级别寻找参数值(有相应的界面进行管理);获取失败,则转入下一步;
   4.从全局参数级别寻找参数值(有相应的界面进行管理);获取失败,则转入一步;
   5.参数值查找失败,参数值为空,结束。
   
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]


Reply via email to