Script to get data from CSV file(s)

[Inbhir Pheofharain]

Hi everyone,

I was wondering if anyone can help me modify my shell script.

It's basically an enquiry tool which is used to get details of any UK railway station.

Currently I input a 3 letter code and it calls up another shell script (the station info), which in turn reverts back to my script. This means I need to create over 2,500 scripts for each station.

rather than that I have 2 csv files which contain all the data that I need, but haven't got a clue how to integrate them. After having Googled for a few hours I've noticed that awk and IFS look the most likely options to use.

What I want to achieve is after typed in the input i want the script to determine whether it's 1.) a 3 character (letters only) input, or 2) a 4 character (numbers or alphanumeric) input. If its 1. then file1 gets parsed, if 2 then file2 gets parsed. The input is the first entry in the CSV files. After the files are parsed I want the rest of the entries on each row to populate various fields.

Here is my existing code:

#!/bin/sh 
clear


echo
echo
echo
echo -e '\E[37;44m				'"\033[1m Welcome to StationMaster \033[0m" #The 1 enables bold text
echo
echo "				         Version 1.0 "
echo
echo "				   All Rights Reserved"
echo "				 ========================"
echo
echo "				   For help type HELP"
echo
echo
echo
echo
echo
echo
echo
echo
echo -n "Enter Station Code or NLC: " 




read FILE 
FILE=${FILE}".sh" 
if [ -f ${FILE} ]; then 
    ./${FILE} 
else
	cd NLC
   if [ -a ${FILE} ]; then 
    ./${FILE} 
    else
    cd ..
	
    echo "Station not found." 
    fi
fi 


#Pause function

function pause(){ 
   read -n 1 -p "$*" 
   echo 
} 
pause 'Press any key to continue...'
./sm4.sh

[sathiya]

You can use perl scripts which can do it simply...

If you dont know how to write perl program, explain the requirement better, somebody else can do for you.

But for running that, you would wanted a perl interpreter.

[Inbhir Pheofharain]

basically what I want to achieve is:

1. Run my script, a "data entry" screen prompt appears.
2. Enter either a) a 3 letter code or b) a 4 digit number.
3. The script determines what has been entered then it searches the relevant csv file.
4. The if the search finds a match (i.e. the input matches the first column on the relevant csv file) it brings up the data for that station (the entries on the remeinder of the row in the file), if no match is found it returns an error.
5. After either outcome a pause function is run then returns to 1.

e.g.

sample from file 1:

EDB, Edinburgh, EDB, 9856, First ScotRail,

sample from file 2:

9856, Edinburgh, EDB, 9856, First ScotRail,

At the Prompt I enter either EDB or 9856

The relevant file is searched

It returns:

Station Name: Edinburgh
CRS Code: EDB
NLC: 9856
Operator: First ScoRail

can anyone help?

[burschik]

I don't see why you need two files in the first place. The data seems to be the same in both cases.

[Inbhir Pheofharain]

I was thinking it may be faster to use 2 files as using one file there would be more data to parse.

If I only need to use one file, can anyone advise me how I go about my script?

Regards

[burschik]

Try something along these lines:

awk -v KEY=/EDB/ 'KEY { print "Station Name: " $2 }' file1

Oh, and of course, "All rights reserved."

[Inbhir Pheofharain]

I assume that the first instance of KEY is the input I type in?

In the example I gave only 1 line of my csv file the input could be any 3 letter code or any 4 digit number.

would that work?

[burschik]

I assume that the first instance of KEY is the input I type in?

In the example I gave only 1 line of my csv file the input could be any 3 letter code or any 4 digit number.

would that work?

That is correct. My simple example will work as long as the key does not occur in any other line. I don't know whether that is the case with regard to your data. The following might be more appropriate:

awk -F", *" -v KEY=EDB '$3 == KEY { print whatever; }'

This uses ", *" as a field separator and then checks whether field number three equals KEY (your input). If so, the code enclosed in braces can format and print the record.

[Inbhir Pheofharain]

Ah I'm with you now.

My input will vary as it can be any 3 letter code or any 4 digit number not just 1 set value.

Basically the csv file is a collection of records and the first entry on each row is the 'input', after this is inputted I want the rest of the row displayed on the screen under various headings. So I would type EDB or 9856 if I wanted to view details about Edinburgh, but the input would be different if I wanted to view details about Glasgow or Aberdeen etc.

[creepy cripple]

while [ true ]
do
echo "Enter Station:"
read input
length=${#input}
if [ $length -eq 3 ]
then
SN=$(grep $input data1 | cut -d, -f2)
CRS=$(grep $input data1 | cut -d, -f3)
NLC=$(grep $input data1 | cut -d, -f4)
OP=$(grep $input data1 | cut -d, -f5)
else
SN=$(grep $input data2 | cut -d, -f2)
CRS=$(grep $input data2 | cut -d, -f3)
NLC=$(grep $input data2 | cut -d, -f4)
OP=$(grep $input data2 | cut -d, -f5)
fi
echo "Station Name: $SN"
echo "CRS Code: $CRS"
echo "NLC: $NLC"
echo "Operator: $OP"
echo; echo
done