Experiment to guess the types of CSV data

type_tracker.py

import csv, re


def is_float(s):
    try:
        float(s)
        return True
    except ValueError:
        return False

def is_int(s):
    try:
        int(s)
        return True
    except ValueError:
        return False


date_re = re.compile('^\d{4}-\d{2}-\d{2}$')


class Tracker:
    def __init__(self):
        self.couldbe = {
            "integer": is_int,
            "float": is_float,
            "date": date_re.match
        }
    
    def __repr__(self):
        return repr(self.couldbe)
    
    def evaluate(self, value):
        if not value or not self.couldbe:
            return
        bads = []
        for name, test in self.couldbe.items():
            if not test(value):
                bads.append(name)
        for bad in bads:
            del self.couldbe[bad]

            
lines = ["name,age,weight,dob", "Dog,\"3", "2\",3.5,2001-02-04", "cat,2,1.2,"]
rows = csv.DictReader(lines)

trackers = {}
for row in rows:
    for key, value in row.items():
        tracker = trackers.setdefault(key, Tracker())
        tracker.evaluate(value)

print(trackers)
# {'name': {},
# 'age': {'integer': <function is_int at 0x10a3e1670>, 'float': <function is_float at 0x10a3e11f0>},
# 'weight': {'float': <function is_float at 0x10a3e11f0>},
# 'dob': {'date': <built-in method match of re.Pattern object at 0x1074e5cf0>}}