performing date comparison with creation_date in user settings

jrmathieu63
jrmathieu63 Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Registered Posts: 26 ✭✭✭✭✭

I have a python recipe that is trapping all our users creation_date and would like to compare this date to a string date defined as following:

settings = user.get_settings()
creation_date = settings.creation_date

before_date = date.today() + relativedelta(months=-3)

if creation_date < before_date:

I know this will not work since they are not the same type but I cannot convert creation_date to string date since there none values.
Anybody know if this variable can be converted to string?

Thanks




Operating system used: AL2

Tagged:

Best Answer

  • jrmathieu63
    jrmathieu63 Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Registered Posts: 26 ✭✭✭✭✭
    Answer ✓

    Was able to find the solution myself by using chatGPT.

    Best way to handle missing dates / null/none (nan) values is to set a default date.
    Not sure why this is not done by Dataiku when creating a new administrative field for reporting.

    Example:
    last_login_time = user.activity['lastSuccessfulLogin']
    converted_last_login = date.fromtimestamp(last_login_time/1e3)


    creation_date = settings.creation_date
    if creation_date is not None:
    creation_date_str = creation_date.strftime('%Y-%m-%d')
    else:
    creation_date_str = default_date.strftime('%Y-%m-%d')
    creation_date = datetime.strptime(creation_date_str, '%Y-%m-%d').date()

    This allows me to conditional search the user list for accounts that have not logged on for at least 6 months or never logged on and account defined for at least 3 months.

    pr6mths = date.today() + relativedelta(months=-6)
    pr3mths = date.today() + relativedelta(months=-3)
    #if last_login_time == 0 and enabled == True and converted_last_login < pr6mths and np.isnat(np.datetime64(creation_date)):
    if last_login_time == 0 and enabled == True and converted_last_login < pr6mths and creation_date < pr3mths:

    Enterprise system must review account access since our security is very concern with access that is not being used.
    Access must be verified/recertified if not in use.

Answers

  • jrmathieu63
    jrmathieu63 Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Registered Posts: 26 ✭✭✭✭✭

    Here is more details on what I have attempted to do.

    #Trap Last Login Datetime
    df = []
    for user in activity:
    login = user.activity['login']
    last_login_time = user.activity['lastSuccessfulLogin']
    converted_last_login = date.fromtimestamp(last_login_time/1e3)
    user = client.get_user(login)
    settings = user.get_settings()
    displayName = settings.get_raw()["displayName"]
    email = settings.get_raw()["email"]
    enabled = settings.enabled
    creation_date = settings.creation_date
    str_date = creation_date.strftime("%Y-%m-%d")

    The attempt to convert creation_date into a string using strtime just fails with the following error.

    Job failed: Error in Python process: At line 31: AttributeError: 'NoneType' object has no attribute 'strftime'
    Not sure what this error means and what is required to resolve this?

Setup Info
    Tags
      Help me…