How the sanity and regression testing works toghether in a simple scenario?

I was delving into the topic of smoke and sanity testing, but after many readings, some things are still not clear to me:

Let’s suppose that i have 3 modules:

1. Login
2. Cart
3. Purchase

If, in this sprint, devs changed something in the Cart functionality, i will execute the sanity testing (to check the correct functioning of the software), if this step fails, i will reject the build and i will never execute the regression right?
But if the sanity passes, i will execute the regression on the touched funcionality/modules (with some impact analysis). In this case i will include in my regression the cart functionality.

Now, my doubts are:

1. Is what i wrote above correct?
2. If yes, is it possible to have a test that is contained in both sets (sanity and regression)?
3. Are the sanity tests only containing positive scenarios? Because i read that sanity is superficial.. If this passes, the negative scenarios will be executed in the regression
4. If I understand correctly, smoke test is carried out at the beginning when the software is not yet stable, while sanity test is carried out when the software becomes stable. This implies, for example, that if I have a smoke test "validate login with correct username and password", if the build is stable it automatically becomes Sanity?
5. Is the difference really just that one runs when the build is not stable and the other runs when the build is stable? If so, why not call them both smoke without giving the testers a headache? hahaha

Thank you in advice.

>Solution :

Addressing your doubts:

1. Yes, You are correct. Sanity testing ensures the basic functionality works after changes, while regression testing verifies broader functionality after changes.

2. Yes, it’s possible for a test to be included in both sanity and regression testing sets if it’s relevant for both purposes.

3. Sanity tests typically focus on primary functionality and positive scenarios to quickly validate basic functionality. Negative scenarios are usually covered more comprehensively in regression testing.

4. Smoke tests are indeed conducted early on to ensure basic functionality. As the software stabilizes, smoke tests may evolve into sanity tests, focusing on broader functionality validation.

5. While the distinction between smoke and sanity tests may seem subtle, it helps to categorize tests based on their purpose and timing in the development lifecycle. Using distinct terms can aid in clear communication and understanding among team members.