Actions
User story #20130
closedrudder language generate_lib should use the existing ncf python library instead of implement itself the ncf methods parsing
Pull Request:
UX impact:
Suggestion strength:
User visibility:
Infrequent - complex configurations | third party integrations
Effort required:
Name check:
To do
Fix check:
To do
Regression:
Description
We already have a parser written in python for the methods in ncf, we should use it whenever possible and not maintain several code bases.
The ticket also fix several parsing issues, all the changes will be listed below:
The ticket also fix several parsing issues, all the changes will be listed below:
- all methods now have a
@documentation
even if empty - fix missing method renaming
- variable resource parameters are now always "prefix" and "name" instead of "prefix|variable_prefix" and "name"
- dsc_built_in_resource now has its parameters in the correct order
- the Json object used to generate the stdlib can have its resource parameters description modified based on method parsing order (but I did not find anything using it except the doc)
- the generate_lib script now outputs the resulting Json in its stdout, this make debugging much easier but we need to supress output when calling it from other scripts (and for instance only keep the .data.errors)
Actions