Question

我有一个运行select的程序（我测试的是56条记录好）然后，当我运行游标时，我想将3个字段传递给一个函数（见上文）查找/从包含1500万条记录（价值10年）的表中选择一条记录。

它返回一个我想要的行类型，然后从这个rowtype记录中提取字段使用第一个选择的记录和获取的其他字段运行插入来自查找功能。

如果我运行该程序，控制台会打印出我的测试消息，但是当我尝试运行时

从v_processed_claim中选择*批量收集到v_tab_proc_claim_recs;

由于错误（97,65）而无法编译：PL / SQL：ORA-00942：表或视图不存在

好像其中任何一个都不是表格。

我这样做是对的......我怎么能这样做，为什么它不能看到我试图提取的表？

我应该以其他方式这样做..

感谢您提供任何帮助/建议：）

功能如下......

create or replace function get_processed_claim_rec(

    p_provider VARCHAR2,
    p_rx VARCHAR2,
    p_record_no  NUMBER 
)


 return i_idb.processed_claim%rowtype
  as
    l_claim_record i_idb.processed_claim%rowtype;
  begin

    select * into l_claim_record from i_idb.processed_claim
      where source_date = p_provider
      AND rx = p_rx
      AND rec_no = p_record_no;


   return(l_claim_record);
  end;

程序是......

create or replace PROCEDURE import_mailer_data

  AS
    -------------------------------

    /**
        for the lookup table
    **/
    v_processed_claim  i_idb.processed_claim%rowtype;
    TYPE proc_claim_recs IS TABLE OF v_processed_claim%ROWTYPE INDEX BY PLS_INTEGER;
    v_tab_proc_claim_recs proc_claim_recs;
    --------------------------------

    CURSOR myCursor
    IS
    SELECT * 
      from
      (   
      SELECT
        j.create_date as open_date,
        case when (j.create_date < (sysdate - 20) )
            then 'POD'
            else 'REG'
        end as priority,
        c.division,
        c.unit,
        --p.refill as days_supply,
        --p.din_name,
       'CM_JOHN' as log_code,
        c.first_name,
        c.last_name,
        --p.UNLISTED_compound,
        --p.intervention_code,
        --p.substitution,
        --p.confirm,
        c.PROVIDER,
        c.rx,
        c.DISPENSE_DATE,
        c.DIN,
        c.QTY,
        c.DIN_COST_PAID,
        c.DISP_FEE_PAID,
        c.PAID_AMOUNT,
        c.SOURCE_DATE,
        c.RECORD_NO,
        c.RELATIONSHIP,
        c.INSURER_NO,
        c.GROUP_NO,
        c.CERTIFICATE,
        c.BIRTH_DATE,
        c.USER_ID,
        --p.rej_code --v_seq_no
          rank() over
          (
            partition by c.provider, c.rx, c.record_no Order by c.provider desc, c.rx desc

          ) as RNK         
      FROM AUDITCOLLECTIONS.MAILER_CLAIMS c,
        AUDITCOLLECTIONS.MAILER_JOBS j
      WHERE MAILER_JOB_DETAIL_ID IN
        (SELECT MAILER_JOB_DETAIL_ID
        FROM AUDITCOLLECTIONS.MAILER_JOB_DETAILS
        WHERE MAILER_JOB_ID IN
          ( SELECT MAILER_JOB_ID FROM AUDITCOLLECTIONS.MAILER_JOBS
          )
        )

        AND ( c.PROVIDER, c.rx, c.record_no ) NOT IN
          ( SELECT provider, rx, rec_no FROM  AUDITCOLLECTIONS.COLLECTION_AUDIT_STAGING
          )
        AND j.create_date > (sysdate - 30)
        AND c.provider = '2010500042'
      ) A_Latest
      where A_Latest.RNK = 1;

    BEGIN
    v_report_id := audit_load.create_loaded_report(v_report_type_id);

      FOR curRec IN myCursor
      LOOP
        BEGIN
          dbms_output.put_line ('===>>>> PRINTING TEST1 = ');        

        v_processed_claim := get_processed_claim_rec(curRec.PROVIDER, curRec.RX, curRec.RECORD_NO);


        select * bulk collect into v_tab_proc_claim_recs from v_processed_claim; 



      END LOOP;

      audit_load.update_status_to_loaded(v_report_id);
    END import_mailer_data;

Answer 1

你可以这样做：

 FOR curRec IN myCursor
  LOOP

      v_processed_claim := 
         get_processed_claim_rec(curRec.PROVIDER, curRec.RX, curRec.RECORD_NO);

      v_tab_proc_claim_recs (v_tab_proc_claim_recs.COUNT+1) := v_processed_claim;

  END LOOP;

或简化为：

 FOR curRec IN myCursor
  LOOP

      v_tab_proc_claim_recs (v_tab_proc_claim_recs.COUNT+1) :=
        get_processed_claim_rec(curRec.PROVIDER, curRec.RX, curRec.RECORD_NO);

  END LOOP;

循环中的Oracle查找函数返回rowtype如何从rowtype获取字段

1 个答案: