You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We're introducing additional linting rules to enhance the consistency and clarity of our commits.
To maintain the efficiency of our CI runtime, it's advantageous to run all linting stages concurrently within a single step.
Parallel Execution with concurrently or run-all:
Pros:
Faster Execution: All linting processes run at the same time, potentially reducing the total linting time.
Efficiency: Ideal for CI environments where multiple cores are available, as it can utilize resources better.
Cons:
Additional Dependency: Incorporating third-party libraries introduces another dependency which might impact our build/test/install process.
Muddled Output: The output from simultaneous linting processes can become intermixed, making it harder to interpret and resolve issues.
Sequential Execution:
Pros:
Clear Output: Since tasks run one after the other, the output is easier to interpret. It's straightforward to pinpoint issues without logs getting mixed.
Simplicity: No additional dependencies or configurations are needed.
Cons:
Slower Execution: Especially during local development, running linting tasks one by one can be slower.
Given the trade-offs, and considering our CI's feedback loop is roughly a minute, I'm leaning towards option 2 (Sequential Execution). It seems a more suitable choice for our needs, even if it's slightly slower during local development.
Proposed Change
Here's the proposed update to our scripts in the package.json:
Proposal: Improve Linting Workflow
Summary
We're introducing additional linting rules to enhance the consistency and clarity of our commits.
To maintain the efficiency of our CI runtime, it's advantageous to run all linting stages concurrently within a single step.
Parallel Execution with
concurrently
orrun-all
:Pros:
Cons:
Sequential Execution:
Pros:
Cons:
Given the trade-offs, and considering our CI's feedback loop is roughly a minute, I'm leaning towards option 2 (Sequential Execution). It seems a more suitable choice for our needs, even if it's slightly slower during local development.
Proposed Change
Here's the proposed update to our
scripts
in thepackage.json
:The text was updated successfully, but these errors were encountered: