Use the store plugin in BKCI
After the plug-in is developed, the plug-in is published to the R&D Store through the R&D store workbench for users to add to the pipeline.
entry
When it was first released, the entry was called "On Shelves"
If the current version is not in the end state, possible operations are as follows:
Launch: The first version enters the launch process
Progress: The release process page is displayed
You can add a new version only when the latest version is in the terminal state
Fill in plug-in related information / upload plug-in release package
Applicable Job Type:
This parameter corresponds to the pipeline Job type. Select this parameter based on the plug-in's actual application
If incorrect, you need to add a new version
Release Package:
The atomCode in task.json must be the same as the identifier entered when adding a plug-in in [create-plugin/]; otherwise, the upload will fail
Incompatible upgrade:
Used when the plug-in input and output are different or the same input and output but the functional logic has changed significantly and is incompatible with the older version
After this version is released, the pipeline that has used this plug-in will not automatically upgrade the version. You need to manually change the version number
Major version number +1
Compatible function update:
The input and output of the plug-in are compatible with older versions, and can only be used when the function is updated or newly added (does not affect existing users )
After this version is released, the pipeline that uses the plug-in and version number is [major version.latest] automatically uses the new version of the plug-in without manually editing the pipeline
Minor version number +1
Compatible problem correction:
Plug-in I/O compatible with older versions, only do problem fixing
After this version is released, the pipeline that uses the plug-in and version number is [major version.latest] automatically uses the new version of the plug-in without manually editing the pipeline
Correction sign +1
Test/release plugin
Test: Click and jump to the pipeline service of plug-in debugging project. You can add the current plug-in to the pipeline to verify whether UI and functions meet expectations
Retransmit packets: If a problem is found in the test, upload the released package again and perform the test again
Continue: After the test is OK and the expectation is met, confirm the submission and release
Cancel the release: You can terminate the release at any time during the release process
Note: When the plug-in development language is Python or Nodejs, the plug-in execution environment has certain requirements. For details, see Preparations for the Third-Party Builder Environment
Last updated