sqoop Oracle to hive出现 Error Msg = ORA-00933: SQL 命令未正确结束

原sqoop脚本：

sqoop import \
--connect "jdbc:oracle:thin:@//10.82.88.223:61521/LZY2" \
--username LZSHARE \
--password 'LZ^share$5105' \
--driver oracle.jdbc.driver.OracleDriver \
--query "SELECT 
    TO_NUMBER(TO_CHAR(GCRQ, 'YYYY')) AS gcrq_year,
    TO_NUMBER(TO_CHAR(GCRQ, 'MM')) AS gcrq_month,
    TO_NUMBER(TO_CHAR(GCRQ, 'DD')) AS gcrq_day,
    YEAR,
    GCRQ,
    GCZBS,
    HOUR,
    MINUTE,
    CDH,
    XSFX,
    SBSBM,
    DCSJLX,
    SJXH,
    CLZQ,
    XKC,
    DKC,
    XHC,
    ZHC,
    DHC,
    TDH,
    JZX,
    MTC,
    TLJ,
    XKCS,
    DKCS,
    XHCS,
    ZHCS,
    DHCS,
    TDHS,
    JZXS,
    MTCS,
    TLJS,
    GCBFB,
    PJCTJJ,
    SJZYL,
    FLAGS,
    OFF_MINS,
    ERR_CODE,
    ERR_DESC,
    DELETE_BY,
    DELETE_TIME,
    CREATE_BY,
    CREATE_TIME,
    UPDATE_BY,
    UPDATE_TIME,
    INSERT_TIME
FROM LZJHGX.DAT_DCSJ_TIME 
WHERE TO_CHAR(GCRQ , 'YYYY-MM-DD') = TO_CHAR(SYSDATE, 'YYYY-MM-DD') AND \$CONDITIONS" \
--split-by "YEAR" \
--hive-import \
--hive-table ods_pre_dat_dcsj_time \
--hive-partition-key "gcrq_year,gcrq_month,gcrq_day" \
--target-dir /user/hive/warehouse/ods_pre_dat_dcsj_time \
--hive-partition-value "gcrq_year=${gcrq_year},gcrq_month=${gcrq_month},gcrq_day=${gcrq_day}" \
-- --hive-drop-import-delims

 出现

关键在于sqoop在计算split-by切片字段时：OriginalSql = SELECT MIN(YEAR), MAX(YEAR) FROM (SELECT
    TO_NUMBER(TO_CHAR(GCRQ, 'YYYY')) AS gcrq_year,
    TO_NUMBER(TO_CHAR(GCRQ, 'MM')) AS gcrq_month,
    TO_NUMBER(TO_CHAR(GCRQ, 'DD')) AS gcrq_day,
    .......
FROM LZJHGX.DAT_DCSJ_TIME
WHERE TO_CHAR(GCRQ , 'YYYY-MM-DD') = TO_CHAR(SYSDATE, 'YYYY-MM-DD') AND  (1 = 1) ) AS t1,

使用 AS t1的语法，Oracle是不支持select * from (select * from A) as t1，这种子查询AS 命名表名的写法。

那么sqoop脚本中如何避免呢？

解决办法：删除掉driver参数--driver oracle.jdbc.driver.OracleDriver