#!/bin/bash

# Parameters File by Pasei Ansah
# Version:		0.1	26/01/2012	PNAnsah	Initial Version
#				0.2 20/02/2012	PNAnsah	Add parameters for a post diff script
#				0.3 20/02/2012	PNAnsah	Rename post POST_DIFF_DATE_FORMAT parameter
#
# Description:	File to store execution parameters for the Pre-Processing scripts
# Usage: 		Reference this file at the start of the script to load the functions
#               and all the parameters below
#

die () {
	echo `date '+%Y-%m-%d %T ' ` "[ERROR] $@" >> $LOG_DIR/$LOG_FILE
	echo >&2 "[ERROR] $@"
	exit 1
}
log () {
	[ $MANUAL_RUN ] && echo `date '+%Y-%m-%d %T ' ` "$@"
	echo `date '+%Y-%m-%d %T ' ` "$@" >> $LOG_DIR/$LOG_FILE
}

# For development the source files were received in .txt form.  In prod this will be .gz - set in properties file?
EXTENSION=.txt.gz


# Hadoop code directory location
HADOOP_BIN_DIR=/home/hadoop/hadoop/bin

# PIG code directory location
PIG_BIN_DIR=/home/hadoop/pig/bin

# Hadoop File System base data directory
HDFS_HADOOP_DATA_DIR=/user/hadoop

# LINUX file system base data directory
FILE_BASE_DIR=/data/newco/data

# Landing folder where Siebel files are delivered and from where they will be retrieved by this script
FILE_LANDING_DIR=$FILE_BASE_DIR/$SOURCE

# Define log directory
LOG_DIR=/data/newco/logs

# Define the name and location for the daily log file
LOG_FILE=Preprocess_Script_$(date '+%Y%m%d').log

# Create log file for the day if it does not already exist
[ -f $LOG_DIR/$LOG_FILE ] || touch $LOG_DIR/$LOG_FILE

# Define length of time in days to store logs before they are purged
LOG_PURGE=90

# Regex search and replace pattern to convert the data date in filename in the format of YYYYMMDD
siebel_DATA_DATE_FORMAT='^\(..\)\(..\)\(..\).*/20\3\2\1'
uim_DATA_DATE_FORMAT='^\(........\).*/\1'

DDF_SOURCE="${SOURCE}_DATA_DATE_FORMAT"
DATA_DATE_FORMAT=${!DDF_SOURCE}

# List of possible source systems
SOURCE_LIST="siebel uim"

# Check that a file containing the following text is present in each daily directory
siebel_FILE_LIST="S_ADDR_PER S_ASSET S_CONTACT S_CON_ADDR S_INV_PROF S_ORG_EXT S_PROD_INT"
uim_FILE_LIST="uimsubscription"

FL_SOURCE="${SOURCE}_FILE_LIST"
FILE_LIST=${!FL_SOURCE}

# Join script to use for source system, if no join is required then create a blank variable
siebel_JOIN_SCRIPT="Asset_Account_Join.pig"
uim_JOIN_SCRIPT="UIM_Drop_Columns.pig"

JS_SOURCE="${SOURCE}_JOIN_SCRIPT"
JOIN_SCRIPT=${!JS_SOURCE}

# Diff script to use for source system
siebel_DIFF_SCRIPT="Asset_Account_Diff.sh"
uim_DIFF_SCRIPT="UIM_Diff.sh"

DS_SOURCE="${SOURCE}_DIFF_SCRIPT"
DIFF_SCRIPT=${!DS_SOURCE}

# Files to diff for each feed
siebel_DIFF_FILES="BillPayer Owner User"
uim_DIFF_FILES="uimsubsfiltered"

DF_SOURCE="${SOURCE}_DIFF_FILES"
DIFF_FILE_LIST=${!DF_SOURCE}

# Set in file delimeter
siebel_DELIMITER='\u00C7'
uim_DELIMITER='\u00E7'

D_SOURCE="${SOURCE}_DELIMITER"
DELIMITER=${!D_SOURCE}

# Set the name of output file to be ingested by ID.
# The ? character will be replaced by the date.
# The # character will be replaced by a suffix if specified
siebel_OUTPUT_FILENAME="siebelsubscriber_?%"
uim_OUTPUT_FILENAME="uimsubscriptions_?"

OF_SOURCE="${SOURCE}_OUTPUT_FILENAME"
OUTPUT_FILENAME=${!OF_SOURCE}

# Define date format to be used in the Post Diff Script
uim_POST_DIFF_DATE_FORMAT='\(....\)\(..\)\(..\).*|\2\/\3\/\1'

PDDF_SOURCE="${SOURCE}_POST_DIFF_DATE_FORMAT"
POST_DIFF_DATE_FORMAT=${!PDDF_SOURCE}

# FTP Details for sending files to ID
HOST=localhost
USER=hadoop
PASS=hadoop


# Set ID Landing Directory
siebel_ID_LANDING_DIR='/data/working/landing/siebel/deltas'
uim_ID_LANDING_DIR='/data/working/landing/UIM/subscription_deltas'

IDLD_SOURCE="${SOURCE}_ID_LANDING_DIR"
ID_LANDING_DIR=${!IDLD_SOURCE}

# Number of parallel Processed to be used for pig jobs
PARALLEL_PROCS=6

# Post diff script(s)
uim_POST_DIFF_SCRIPT="UIM_Add_Date.sh"

PDS_SOURCE="${SOURCE}_POST_DIFF_SCRIPT"
POST_DIFF_SCRIPT=${!PDS_SOURCE}