SAP Message BRAIN513 - Break point RSDRC_SELDR_SPLIT reached


A query is split up here into sub-queries according to specific rules.The system finds an optimum aggregate for each sub-query. If asub-query is too complicated, the selections are simplified andgeneralized.

Importing
I_TH_SFC: Description of fields that are selectedor drilled down.
I_TSX_SELDR: Description of selections
I_TH_SFK: Description of key figures

Exporting
E_TX_SELECTION: Table, in which each row describes a sub-query
A row contains the tables: SFC, SELDR, SFK, CHARET. It also containsthe following fields:

  • AGGREGATE: Optimum aggregate for the sub-query

  • SEL_FEMS: Bit string of which selections/FEMS from the original
    • SELDR are considered
    • NODERETFL: Boolean value stating whether the query has the potential
      • to return hierarchy nodes.

      Process
      In the form 'SELDR_ADAPT_TO_PROVIDER', the SELDR table is adjustedaccording to the specific capabilities of the InfoProvider in question(and so can be read either with the KEY or the SID, hierarchy nodes canbe selected, and so on).
      The next step deals with whether the query is to be split intosub-queries:

      • If the query is to be split into sub-queries, the SELDR table is
        • split up in the form 'SELAGGR_GET'. The description of an optimumaggregate is also returned for each sub-query. The result appears inthe table 'L_TH_SELAGG'R.
        • If the query is not to be split into sub-queries, the description of
          • the aggregate is determined in the form 'SELAGGR_SIMPLE_GET'.
          The system searches the best aggregates available for the descriptionsof the optimum aggregates in the form 'AGGREGATES_FIND'.
          Sub-queries that read data from the same aggregate/InfoCube areregrouped as much as possible in the form 'SELAGGR_CONDENSE'.
          The field 'READLEVEL' is always set in form 'SFC_FIT_TO_AGGREGATES'when hierarchy nodes are selected.
          The SELDR table of the sub-queries is determined in the form'SELECTION_SPLIT'. If the SELDR table is too complex, it is simplifiedin the function module 'RSDRC_SELDR_CONVEX_HULL'.
          If the query also reads data that has yet to be rolled up into theaggregates, those queries that read data from the aggregates are splitup into a query for the aggregate and a query for the InfoCube in theform 'REQUID_SPLIT'. The condition is adjusted to the characteristic'0REQUID' accordingly.