chocolatey / package-validator

Windows service to validate packages conform to package standards
Apache License 2.0
32 stars 29 forks source link

Do we need inverse rule validation #99

Open gep13 opened 8 years ago

gep13 commented 8 years ago

Having just gone through this process:

https://github.com/chocolatey/chocolatey-coreteampackages/pull/118

For adding admin tags when certain Chocolatey Helpers are being used, it made me wonder about whether we need inverse validation rules.

For example, we now have a check to see that if certain Chocolatey Helpers are being used, that the admin tag exists. Should be also check to see if the admin tag exists, that one of the main chocolatey Helpers is actually being used in the script?

What are your thoughts @ferventcoder ?

ferventcoder commented 8 years ago

That makes sense - we may need one for at least that. We probably need some others as well. There is no need to change any design, it's already flexible enough to handle this case without issues. It's just another rule that is added to perform the inverse check. :+1:

gep13 commented 8 years ago

@ferventcoder I will go through the existing rules, and make a checklist in this issue, for any existing rules that need an inverse check, and then any new rules going forward will take this into account.

gep13 commented 8 years ago

@ferventcoder on a train just now, and I have just done the first inverse rule for the situation that I originally thought of when I created this issue. Can you think of any others that it would be required on?

If so, I can add to the PR that I just created.

Working with only 11% battery left, so won't be doing much more PR's tonight though :-(

ferventcoder commented 8 years ago

For this particular, I wonder if the admin tag could still be required EVEN if the user is not using a helper.