NFIRSparser.py

import csv
import re

def main():
    moduleData = {}
    table = {}
    rows = []
    title = ""

    with open("NFIRS_Spec_Tables_2013.csv", 'r') as spec:
        reader = csv.reader(spec)
        for row in reader:
            if (row[0] == "" and row[1] == "") or ("please note" in row[1].lower()):  # Empty line
                continue
            if row[0] != "" and row[1] != "": # Has useful data
                rows.append(row)
            if "please note" in row[0].lower(): # Has a reference to another module
                continue 
            if "module; begin" in row[1].lower(): # Has module demarcation
                moduleMatch = re.match("[eE]nd(?:\sof\s|\s)(.*) [mM]odule; [bB]egin (.*) [mM]odule", row[1])
                moduleData[moduleMatch.group(1).lower()] = table
                table = {}
            if (row[0] == "" or row[0] == " ") and row[1] != "": # Has title or note
                table[title] = rows
                title = row[1].lower()
                rows = []
        
        print(moduleData["fire"])
        for module in moduleData:
            for table in moduleData[module]:
                print(module, table)
        # With data model built, need to replace some references
        for module in moduleData:
            for table in moduleData[module]:
                for row in moduleData[module][table]:
                    if row[0].lower() == "please note:" and "please see the codes listed" in row[1].lower():
                        print(row[1])
                        codeMatch = re.match(".*used for ([\w\s]*)(?:[\s-]).* in the (.*) [mM]odule", row[1])
                        print(codeMatch.groups())
                        moduleData[module][table] = moduleData[codeMatch.group(2).lower()][codeMatch.group(1).lower()]

if __name__ == "__main__":
    main()