README.md

CSVC

Categorise CSV cells based on string matching and Regex.

By parsing a config file, CSVC automatically categorizes the CSV cells.

Parameters

noheader means the input CSV file doesn't contain a header, so matching can begin from the first (0 indexed) row

column NUMLIST sets the columns of interest to the list of numbers. e.g column 0,1,2 will mean CSVC searches through columns 0,1, and 2 to attempt to match.

MODE controls how CSVC will write the substitutions. append simply appends the substitution to the end of the line, replace NUM replaces column NUM with the substitution, and inserts NUM QUOTE inserts a column with the name QUOTE at index NUM.

Grammar

The grammar for the config file is as follows:

CONFIG ::= HEADER? COLUMN? MODE CATEGORY+
HEADER ::= header OPTSEMI
COLUMN ::= column NUMLIST OPTSEMI
NUMLIST ::= NUM [, NUM]*
MODE ::= append OPTSEMI replace NUM OPTSEMI | insert NUM QUOTE OPTSEMI
CATEGORY ::= category QUOTE { RULESET* }
RULESET ::= STRINGRULETYPE | REGEXRULETYPE
STRINGRULETYPE ::= RULETYPE { RULE* }
RULETYPE ::= equals | contains
REGEXRULTEYPE ::= matches { REGEXRULE* }
RULE ::= QUOTE OPTSEMI
REGEXRULE ::= REGEX OPTSEMI
QUOTE ::= "[a-zA-Z ]+"
REGEX ::= ".*"
NUM ::= [0-9]+
OPTSEMI ::= ;?

e.g

noheader
column 0,1
replace 1
category "Food And Drink" {
    contains {
        "Rest"
        "Mc Donalds"
        "Maccas"
    }
}
category "Bad" {
    contains {
        "Key"
    }
}
category "Groceries" {
    equals {
        "New World"
    }
    contains {
        "New World"
        "CountDown"
    }
    matches {
        "b.*a"
    }
}