We export a task selecting the "Without endpoints" option. When reimporting the task on a different QLIK server the endpoint names reappear. However those endpoint names are different fore each server.
Is there a way to configure this behaviour and simply export the task without any endpoint names at all ?
Well, < export a task selecting the "Without endpoints" option.> means that there will be no <databases> section in the json to define those entrypoint, only the task itself, with entrypoints named - but not defined, as an (only) entry under <tasks>
The expectation is that you have are working on a DEV box pointing to a DEV database and want to move (import) to QA or PROD. The definition of the task itself need (should!) not change, but the database should be the on for QA or PROD - UNDER THE SAME NAME. When doing a full export and subsequent import on a server with different goal importing a full export would blow away any existing QA (prod) database definition resulting in being the dev task with its dev endpoints just on a different server.
By NOT importing the endpoints you can move the tasks itself unchanged from DEV to QA to PROD and back as desired.
Now many developers choose to have the the endpoint names reflect the usage environment and/or database name. For example HR_XYZ_SRC_DEV vs HR_XYZ_SRC_PROD. So they have to lightly edit the JSON to reflect this. I'd like to argue that it is easier to do (automate) with a present endpoint name than a removed one as you seem to request/suggest. IMHO It is easier and more clear to search and replace <_DEV> than to look for an empty string.
Well, < export a task selecting the "Without endpoints" option.> means that there will be no <databases> section in the json to define those entrypoint, only the task itself, with entrypoints named - but not defined, as an (only) entry under <tasks>
The expectation is that you have are working on a DEV box pointing to a DEV database and want to move (import) to QA or PROD. The definition of the task itself need (should!) not change, but the database should be the on for QA or PROD - UNDER THE SAME NAME. When doing a full export and subsequent import on a server with different goal importing a full export would blow away any existing QA (prod) database definition resulting in being the dev task with its dev endpoints just on a different server.
By NOT importing the endpoints you can move the tasks itself unchanged from DEV to QA to PROD and back as desired.
Now many developers choose to have the the endpoint names reflect the usage environment and/or database name. For example HR_XYZ_SRC_DEV vs HR_XYZ_SRC_PROD. So they have to lightly edit the JSON to reflect this. I'd like to argue that it is easier to do (automate) with a present endpoint name than a removed one as you seem to request/suggest. IMHO It is easier and more clear to search and replace <_DEV> than to look for an empty string.
If I understand what you're trying to do; we have a similar need.
Server 1 is our DEV server and an end point name is something like SRC_DEV_DATABASE_01
When moving to Server 2; we need this end point name to match the environment like SRC_UAT_DATABASE_01.
To achieve this on our side; we use a python script to export the task without endpoints; modify the json definition for the task to change SRC_DEV_DATABASE_01 to SRC_<environment>_DATABASE_01 depending on what environment it needs to go to.
