|
Vishwas,
On 1, the issue of having a unique transaction_id for loans, savings, journal entries, you already produce some random string for manual journal entries and reuse the unique autoincrement transaction_id from loan_transactions or savings_transactions - Maybe we should just use the one approach for all three areas that gaurantees uniquenesss i.e. use a GUID/UUID based on the date-time of transaction. It also brings up a bigger question about maybe modelling all transactions in the system in one table with a one-to-one extension table for any differences between them?
On 2, the issue of fetching m_payment_detail data from journalentries/ resource - as you say sander theres a number of nice search by features on that API that you dont want to move to a stretchy report - As the m_payment_detail table is just a simple extension of the m_loan_transaction and m_savings_transaction table, there should be no issue with joining through to get the data. If performance becomes an issue as a result of this we can look at that seperately.
|