Buffer logic in One Order header extension Read
In one of my previous blog A small tip to analyze code by code and export to markdown format it is introduced that the following header extension are supported in One order read function module CRM_ORDER_READ.
When the given header extension is read via CRM_ORDER_READ, take OPPORT_H for example:
DATA: lt_header_guid TYPE crmt_object_guid_tab, lv_guid TYPE crmt_object_guid, lv_object_id TYPE CRMT_OBJECT_ID_DB value '21', lt_oppt TYPE CRMT_OPPORT_H_WRKT, lt_partner TYPE crmt_partner_external_wrkt. START-OF-SELECTION. CALL FUNCTION 'CONVERSION_EXIT_ALPHA_INPUT' EXPORTING input = lv_object_id IMPORTING output = lv_object_id. SELECT SINGLE guid INTO lv_guid FROM crmd_orderadm_h WHERE object_id = lv_object_id AND process_type = 'CXOP'. CHECK sy-subrc = 0. APPEND lv_guid TO lt_header_guid. CALL FUNCTION 'CRM_ORDER_READ' EXPORTING it_header_guid = lt_header_guid IMPORTING * et_partner = lt_partner et_opport_h = lt_oppt.
The calling hierarchy could be found from below:
Every header extension has one corresponding read function module acting as entry point for read which will be called by CRM_ORDER_READ_OW with naming convention CRM_<Object name>_READ_OB. In OB ( object buffer ) read function module, the corresponding object buffer is evaluated. If object buffer is not hit for current read, another DB read function module ( CRM_<Object name>_READ_DB is called. Within this DB read function module, another internal table which represents Database buffer is evaluated again. The real read access on database table is only performed when this second fold buffer check fails.
Take OPPORT_H read for example, I draw a picture to demonstrate this two-fold buffer evaluation logic:
In the runtime, OPPORT_H object buffer could be monitored in the context of CRM_OPPORT_H_READ_OB,
and database buffer in CRM_OPPORT_H_READ_DB accordingly.
The object buffer is declared in function group CRM_OPPORT_H_OB:
and database buffer is defined in function group CRM_OPPORT_H_DB:
I have written a series of blogs to explain how One Order API works. The blogs are written based on a simple scenario: read, change and save field “Closing Date” in Opportunity header level.