Get Field Separator from a File - ksh

Question

I have several files and they are different between each other and they use ; and | as separators.
Is there a way to let my program read the file get the ; or | separator to save into a variable and use it later on in my script?

Answer 1

Generally speaking the answer is 'yes' you can dynamically determine the delimiter and use it in later coding.
You haven't mentioned how you'll determine which delimiter is in use (eg, what happens if both characters exist in your data file?) so for the sake of discussion we'll assume the delimiter has been determined and stored in the del variable.
You also haven't stated how you'll use this delimiter in your code so, again for sake of discussion, we'll look at examples using cut and awk.
Let's assume our datafile (mydata) contains the following single line of data:
$ cat mydata
abc;def|ghi;jkl|mno;pqr|stu
We'll now switch our delimiter between ; and | and look at some simple cut and awk examples ...
############################
# set our delimiter to ';'
$ del=";"
##### display 1st field
$ cut -d"${del}" -f1 mydata
abc
$ awk -F"${del}" '{print $1}' mydata
abc
#### display 4th field
$ cut -d"${del}" -f4 mydata
pqr|stu
$ awk -F"${del}" '{print $4}' mydata
pqr|stu
############################
# set our delimiter to '|'
#
$ del="|"
##### display 1st field
$ cut -d"${del}" -f1 mydata
abc;def
$ awk -F"${del}" '{print $1}' mydata
abc;def
##### display 4th field
$ cut -d"${del}" -f4 mydata
stu
$ awk -F"${del}" '{print $4}' mydata
stu