ruzeynalov commented on PR #6415:
URL: https://github.com/apache/fineract/pull/6415#issuecomment-5583959642

   > Confirmed test-only, and I independently re-derived the schedule math 
against the actual production calculator for every scenario rather than 
trusting the regenerated numbers - all of it reproduces exactly, including the 
rate-segment scenarios, so there's no sign of the non-monotonic-balance issue 
that was found on #6339.
   > 
   > One thing that's actually failing CI though: scenario C102416 ends with 
`And The retrieved amortization schedule has no negative monetary amounts`, but 
that scenario never runs `Admin retrieves the projected amortization schedule` 
- it only uses the `has N periods` step, which doesn't populate the context the 
assertion step reads from. This is the only scenario in the whole suite that 
asserts on "the retrieved schedule" without retrieving it first, and it's 
exactly what's failing shard 10 with a null-response error. Easy fix - either 
add the retrieve step before that line, or drop it.
   > 
   > Non-blocking: C102414 mostly duplicates the existing C98194 scenario (same 
product/amounts, same first few rows) with less coverage of the interesting 
tail - might be worth folding one into the other rather than keeping both.
   > 
   > Recommendation: CHANGES_REQUESTED
   
   @galovics Thank you for your feedback
   
   **C102416 - agreed, fixed.**
   
   **C102414 vs C98194 - I would keep both, and I have made the intent explicit 
in the file.** They share the product and rows 1-3 but exercise different 
things:
   - C98194 calls the explicit `generate` endpoint on a *submitted* loan, pins 
the tail (rows 210-212) and then rejects the loan. C102414 disburses and reads 
back the model persisted at disbursement with no generate call, which is the 
path a real loan takes.
   - C102414 is the only scenario pinning payment 5 (amortization 8.93). That 
cell is the carry-forward diagnostic from the normalization work in #6369: the 
high-precision value 8.924171922530 rounds to 8.92 on its own, and only the 
fraction carried over from days 1-4 gives 8.93. C98194 does not assert that 
cell.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to