If you have the data in a CSV file you can read the data into a SAS data set using either proc import, which makes some guesses as to variable properties based on the content.
Example code:
proc import datafile="C:\My Documents\myfiles\delimiter.csv"
dbms=dlm
out=mydata
replace
;
delimiter=',';
getnames=yes;
guessingrows = max;
run;
The procedure will generate data step code that you can copy from the log and modify if something isn't quite right.
If you have two different CSV files with the same structure, or common variables, that you will want to combine later you should either verify that the created common variables have all of the same properties, (basically the INFORMAT statement for the variable).
If not you may want to modify both data step code sections to provide common properties.
The data set should at least run with MIANALYZE but if the content is not a simple row/column layout with each column intended to be a single variable then that is another question on how to read a more complex file layout.
Questions about combining datasets are frequent topics here so don't hesitate to ask when you get to that part though the Data step MERGE or a Proc SQL block are the two most likely approaches depending on needs.
