detect_format, train on column before extraction

[DanielJDufour]

Assuming a column in a csv will all be formatted the same, I should be able to train on a column of dates before detecting dates

two new methods

from date_extractor import detect_format

data = [None, "", "10/31/23", "1/2/23"]

detect_format(data)
"%m/%d/%Y"


from date_extractor import prepare

extract_date = prepare(data)
for date in data:
    extract_date(date)

[DanielJDufour]

Have to also handle situations where the date format doesn't fit with the traditional date formats, in which case might need to return a special format or regex like
"(?PJanu|Febr|Marc|)/..."

[DanielJDufour]

I think I like the word train more like

from date_extractor import train

data = [None, "", "10/31/23", "1/2/23"]

# train will basically randomly select 100 text from the data set
extract_date = train(data, limit=100)



for item in data:
    date = extract_date(item)