It would be useful if we could make some engines triggered after all the others have run. For example, we could then keep track of statistics of the linting without changing linting engines or we could rework file stats engine to not fake POST /file as "analysis" but to simply be invoked after all other analysis has run.
These engines should be defined is a separate section of lazy.yaml file and the definition should also mark their order. It stands to reason to have the engines process the output sequentially rather than concurrently - for example we could have an engine that filters out all the warnings that the users has turned off but that should be triggered after the statistics engine which should keep track of everything no matter how inane it seems to the user.
ierceg
commented
7 years ago
It would be useful if we could make some engines triggered after all the others have run. For example, we could then keep track of statistics of the linting without changing linting engines or we could rework file stats engine to not fake
POST /fileas "analysis" but to simply be invoked after all other analysis has run.These engines should be defined is a separate section of lazy.yaml file and the definition should also mark their order. It stands to reason to have the engines process the output sequentially rather than concurrently - for example we could have an engine that filters out all the warnings that the users has turned off but that should be triggered after the statistics engine which should keep track of everything no matter how inane it seems to the user.