关于shell脚本求解

-----控制文件如下：
[oracle@CMSINTE1 ftp_user]$ cat crm_gw_day.ctl
load data
infile PROD_GW_#599_D_130415_A_NM_F1.ITF
badfile '/oradata/ftp_user/gw_crm_day.bad'
DISCARDMAX 10
append
into table EDW_CRM_GW_DAY_SRC
fields terminated by ","
optionally enclosed by '"'
TRAILING NULLCOLS
(LATN_ID ,    
BUREAU_KEY  ,
PROD_ID ,
PROP_CUST_ID,
CUST_ID ,
PROD_SPEC_ID,
PROD_SPEC_TYPE ,
CRT_DT  timestamp 'yyyy-mm-dd hh24:mi:ss' ,
MDFY_DT timestamp 'yyyy-mm-dd hh24:mi:ss',
STAFF   ,
PROD_ST_CODE,
SERV_ACC_NBR,
ADDR_ID ,
EXCH_ID ,
CUST_NAME   ,
CUST_GRADE_ID  ,
CITY_FLAG  ,
FILE_CYCLE timestamp 'yyyy-mm-dd hh24:mi:ss',
FILE_NBR,
NEW_FLAG_ID ,
TRADE_TYPE_ID  ,
JOIN_DURATION_LEVEL2_ID   ,
MDSE_SPEC_ID,
ACCOUNT ,
REGION_ID   ,
RENT_TYPE   ,
ADDRESS_NAME,
MAIN_GROUP_ID  ,
BRAND_ID,
IS_BILL_USER,
IS_AVAIL_USER  ,
PAY_TYPE,
CHAN_ID ,
NBR_HEAD,
STOP_FLAG   ,
REMOVE_FLAG ,
PREFER_ID   ,
PREFER_NAME ,
PAG_FLAG,
PREFER_VALUE,
PREFER_PROD_ID ,
PREFER_EEF_DT timestamp 'yyyy-mm-dd hh24:mi:ss' ,
PREFER_EXP_DT timestamp 'yyyy-mm-dd hh24:mi:ss' ,
PREFER_CRT_DT timestamp 'yyyy-mm-dd hh24:mi:ss' ,
PREFER_MDFY_DT timestamp 'yyyy-mm-dd hh24:mi:ss' ,
PREFER_STATE,
PRICE_ID,
PRICE_NAME  ,
PRICE_ID_JT ,
CUST_TYPE_LEVEL_4_ID ,
CHANEL_TYPE_ID ,
SHLR_ID ,
FREE_FLG,
TC_CLASS,
CUST_CODE   ,
COMM_ORG_ID ,
DEV_TE_ID   ,
DEAL_TE_ID  ,
DEV_STAFF_ID,
DEAL_STAFF_ID  ,
SALE_DT timestamp 'yyyy-mm-dd hh24:mi:ss',
PROD_FUNC_TYPE ,
CUST_CUR_YEAR_TYPE ,
CUST_SERVICE_LEVEL ,
CUST_CREDIT_LEVEL  ,
CUST_CREDIT_VALUE  ,
BIRTHDAY timestamp 'yyyy-mm-dd hh24:mi:ss',
AGE,
SEX,
CONTACT_NAME,
CONTACT_PHONE）-----SH文件如下：
[oracle@CMSINTE1 ftp_user]$ cat crm_gw_day.sh 
#!/bin/bash
#Setup Oracle Enviroment
SHELL=/bin/bash
PATH=/sbin:/bin:/usr/sbin:/usr/bin
ORACLE_BASE=/oracle/app/oracle/product/10.2.0
export ORACLE_BASE
ORACLE_HOME=/oracle/app/oracle/product/10.2.0
export ORACLE_HOME
NLS_LANG=AMERICAN_AMERICA.ZHS16GBK
export NLS_LANG
PATH=$PATH:$ORACLE_HOME/bin
export PATHfilepath='/oradata/ftp_user/';
filelog='crm_gw_day_imp.log';
filename_src='crm_gw_day_src.ctl';
filename='crm_gw_day.ctl';date>> $filepath/$filelog;echo  "begin">>$filepath/$filelog;cd $filepath;for var in   `find $filepath -name "PROD_GW_#*_D_*_A_NM_F1.ITF" -exec basename {} \;` ; do echo  ${var}>> $filepath/$filelog;if [ -f "test.ctl" ]; then
  
rm  $filepath/$filename;
  
ficp $filepath/$filename_src $filepath/$filename;
sed -i 's/infile/infile '${var}'/g' $filepath/$filename;
sqlldr userid=dw/cms_dw@fj_cms_145 control=$filepath/$filename   direct=y;cat crm_gw_day.log>>crm_gw_day_log.log;mv  $filepath/${var}  /oradata/bakdata;
doneecho "end">>$filepath/$filelog ;
[oracle@CMSINTE1 ftp_user]$ ----------------------------------------------------
求好心人士，详细讲解下以上shell脚本的具体执行业务逻辑，谢谢