gl_je_lines (je_header_id, je_line_num) -> gl_import_references (je_header_id, je_line_num)
gl_import_references (gl_sl_link_table, gl_sl_link_id) -> xla_ae_lines (gl_sl_link_table, gl_sl_link_id)
xla_ae_lines (applicaiton_id, ae_header_id) -> xla_ae_headers (application_id, ae_header_id)
xla_ae_headers (application_id, event_id) -> xla_events (application_id, event_id)
xla_events (application_id, entity_id) -> xla.xla_transaction_entities (application_id, entity_id)
xla.xla_transaction_entities (source_id_int_1, etc) after filtering by application_id, entity_code and ledger_id -> subledger's table(its key columns mentioned in xla_entity_id_mappings) for that ledger_id
For Example:
xla.xla_transaction_entities (source_id_int_1) filtered by application_id 200, entity_code AP_INVOICES and ledger_id -> ap_invoices_all (invoice_id) for that set_of_books_id.
xla.xla_transaction_entities (source_id_int_1) filtered by application_id 200, entity_code AP_PAYMENTS and ledger_id -> ap_checks_all (check_id) for that set_of_books_id.
xla.xla_transaction_entities (source_id_int_1) filtered by application_id 222, entity_code TRANSACTIONS and ledger_id -> ra_customer_trx_all (customer_trx_id) for that set_of_books_id.
Note:
a) There is an index on xla.xla_transaction_entities on the following columns.
application_id
entity_code
ledger_id
nvl(source_id_int_1,-99)
nvl(source_id_int_2,-99)
nvl(source_id_int_3,-99)
nvl(source_id_int_4,-99)
nvl(source_id_char_1,' ')
...
nvl(source_id_char_4,' ')
b) Use application_id filter wherever it is possible, as above mentioned XLA tables are partitioned by that.
No comments:
Post a Comment