去雙引號：

sed -i "s/"//g" textName working: perl -p -i -e "s/ /,/g" ./wuhan_feiy_end_result.csv

出現 分組無效情況：
select patient_sn,year,sex,cast(year as int)-cast(birth_date as int) from tableName  limit 10;
select trim(patient_sn) as patient_sn,trim(visit_type) as visit_type,trim(year) as year,trim(sex) as sex,trim(age) as age,sum(fee_sum) from tableName  group by
trim(visit_type) ,trim(patient_sn), trim(year),trim(sex),trim(age) order by patient_sn,year,sex,age,visit_type ;
--結論：可能是 分組資料存在空格導致分組無效，使用trim（）函式解決。

建表語句， 載入csv檔案：

create table tableName (JZJLH string,SDYWH string) ROW FORMAT SERDE 'org.apache.hadoop.hive.serde2.OpenCSVSerde'  WITH SERDEPROPERTIES (  "separatorChar" = ",",   "quoteChar" = "\"",   "escapeChar" = "\\"   )

--建表語句模板：

create table tableName(JZJLH string,SDYWH string) row format delimited fields terminated by '","' lines terminated by '\n' stored as textfile;

載入資料到表中：

OVERWRITE INTO TABLE tableName; load本地新檔案到hive：utf-8 load data local INPATH path OVERWRITE INTO TABLE tableName;

查詢結果只 生成一個檔案：（部分欄位）匯出到本地

hive -e "SELECT concat(jzjlh,',',fldmmc,',',zje) FROM zhangheng.tableName;">/opt/textName

將 查詢到資料插入本地檔案：

INSERT OVERWRITE LOCAL DIRECTORY '/opt/zhangheng' row format delimited fields terminated by "," SELECT jzjlh, fyrq, fldm,sum(cast( zje as double)),fldmmc FROM  tableName group by jzjlh, fyrq, fldm ,fldmmc ORDER BY  jzjlh desc,  fldm desc,  fyrq desc ;

查詢結果插入hive表：

insert overwrite table tableName SELECT jzjlh, fyrq, fldm,sum(zje),fldmmc,count(zje) FROM  tableName group by jzjlh, fyrq, fldm ,fldmmc ORDER BY  jzjlh desc,  fldm desc,  fyrq desc ;