@ldgze How is this going?

@ldgze How is this going?

Hi David,

I’m still working on the issue and plan to commit the code later today.

Hi David @ddfridley,

I have a question about when we should call randomWhys for each mostId and leastId. Specifically, do we need to call randomWhys only when an ID has no existing "why," or should we ensure that each ID has at least 5 "whys" by calling randomWhys to make up any shortfall?

Just want to make sure I align with the requirements. Thanks for clarifying!

Hi David,

I’ve implemented the getWhyRanksAndPoints API based on my understanding of the requirements and have created a PR for it. However, I still need some clarifications to ensure everything aligns correctly.

Implementation Steps

1. Fetch Ranks:
   Query ranks matching userId, discussionId, stage: 'why', and round.

2. Extract Parent IDs from Ranks:
   Gather parentId values from the fetched ranks.

3. Fetch Points Based on Parent IDs:
   Query the Points collection for why-points where parentId matches any of the parentIds from the ranks.

4. Check if All mostIds and leastIds Have Corresponding Why-Points:

   • Step 4.1:
     If all mostIds and leastIds have corresponding why-points:
     • Return ranks and points (Currently, I’m returning the why-points fetched based on ranks, not specifically those matching mostIds and leastIds).
   • Step 4.2:
     If not all mostIds and leastIds have corresponding why-points:
     • Proceed to Step 5.

5. Fetch Missing Whys Using getRandomWhys:
   For any missing mostIds or leastIds, fetch random whys using getRandomWhys.

My Questions

1. For Step 4 and Step 5:

   • Should I return the why-points fetched based on ranks?
   • Or should I return the why-points matching the mostIds and leastIds?

2. Test Case Relationship:

   • When writing test cases, how should I define the relationship between the why-points fetched according to the ranks and those matching the mostIds/leastIds?

Thanks for your time and guidance!

@ldgze Hi! I see my responding to this got dropped, And I haven't seen you for a while. Are you still able to work on it? I hope your having a good new year!

@ldgze Hi! I see my responding to this got dropped, And I haven't seen you for a while. Are you still able to work on it? I hope your having a good new year!

Hi David,

Happy New Year! I’ve been busy with job hunting recently, but I’ll take a look at the code today and provide an update.

mostIds and leastIds are 'main-points' that the user has ranked as most/least important.
The ranks for stage: why will be ranks of why-points (not main points). Some why-points will have been entered by the user, other why points will be from other users.
For question1, the why-points returned should be the ones corresponding to the ranks, and/or ones without corresponding ranks yet because they were gatherd by getRandomWhys.
For question 2, mostIds/leastIds are not why points, they are main points.

We only have one collection for points, because the format is the same, and in other situations there could be deeper nesting, main-point, why-point, sub-why-point, sub-sub-why-point. etc. Ie in another discussion we could ask people why they think that's the most important why point. -- it's common in human discussion.

I hope this helps.

Thanks!