# End of Month Calculation

Solved!
Level 1
###### End of Month Calculation

I have a date column and I want to show that date as end of month. For example, if one date is 08/01/21, I am looking to show this as 08/31/21. Something similar to End Of Month function in Excel.

Thanks

2 Solutions
Dataiker

You may want to consider trying the "Increment date" processor within a Prepare recipe. I've included a screenshot below demonstrating how it works. Note that you may make the changes in place or add an output column.

You may also find the tutorial on Filtering and Processing Dates helpful!

If that does not work for you, would you please provide more detailed information on your use case?

Thanks so much,

Jordan

Dataiker

You could use a prepare recipe with custom Python functions. An example of a custom Python function that calculates the end of month is below. The input for this is a column called line  that contains the dates as Strings (format MM/DD/YYY).

``````import datetime

def process(row):

current = datetime.datetime.strptime(row["line"],'%m/%d/%y')
EndOfMonth = (datetime.datetime(current.year + int(current.month / 12), ((current.month % 12) + 1), 1) - datetime.timedelta(days=1))
row["line"]= datetime.datetime.strftime(EndOfMonth,"%m/%d/%y")
return row``````

18 Replies
Dataiker

You may want to consider trying the "Increment date" processor within a Prepare recipe. I've included a screenshot below demonstrating how it works. Note that you may make the changes in place or add an output column.

You may also find the tutorial on Filtering and Processing Dates helpful!

If that does not work for you, would you please provide more detailed information on your use case?

Thanks so much,

Jordan

Level 1
Author

Thanks Jordan. This is helpful and will work for what I am looking for now as I have the same date. However, if I have different dates (i.e. 8/1/21 & 8/3/21). If i do the "increment date" here, the first date will be end of month but the second will shift to Sep. What's the solution here?

Dataiker

You could use a prepare recipe with custom Python functions. An example of a custom Python function that calculates the end of month is below. The input for this is a column called line  that contains the dates as Strings (format MM/DD/YYY).

``````import datetime

def process(row):

current = datetime.datetime.strptime(row["line"],'%m/%d/%y')
EndOfMonth = (datetime.datetime(current.year + int(current.month / 12), ((current.month % 12) + 1), 1) - datetime.timedelta(days=1))
row["line"]= datetime.datetime.strftime(EndOfMonth,"%m/%d/%y")
return row``````

Level 1

@Alhamdan1995 were you able to figure this out? I am also in need of this solution. Please advise. Thanks

Dataiker Alumni

Hi @hemsharma82 and welcome to the Dataiku Community. Do either of the two accepted solutions here work for your use case? If not can you describe a little bit about how your use case may differ?

Looking for more resources to help you use Dataiku effectively and upskill your knowledge? Check out these great resources: Dataiku Academy | Documentation | Knowledge Base

Level 1

@CoreyS , thank you for responding. None of the two solutions are working for my use case. My goal is to calculate "Last Day of the Current Month" and "Last Day of the Previous Month". I have use now() function to get the "Current Date" and trying to calculate the above two values from "Current Date Column". I hope this clarifies the requirement. Best, Hem

Dataiker

Hi @hemsharma82,

We would like to know why the accepted solutions are not working for your use case. If you are getting any error messages, please share them.

Can you share also the Python code that you are using? Please confirm the input column type and date format.

Dataiker

Hi @hemsharma82,

Below is the Python code that calculates last day of the current month based on the current date returned by the now() function:

``````import datetime

current = datetime.datetime.now()
print(current)

EndOfMonth = (datetime.datetime(current.year + int(current.month / 12), ((current.month % 12) + 1), 1) - datetime.timedelta(days=1))
print(EndOfMonth)``````

The output of this code is:

``````2022-08-04 14:51:54.232576
2022-08-31 00:00:00``````

Level 1

Hi @CatalinaS ,

Granted, my knowledge of Python is very limited. I have tried to use this code in a prepare recipe (both with and without the "def process(row):" and "return len(row)" that are inserted automatically) and keep getting errors.

Level 1

Thank you Catalina. Where exactly in the recipe interface do I write this code?

Dataiker Alumni

Hi @hemsharma82 as mentioned in @CatalinaS original solution, you can use a prepare recipe with custom Python functions.  I hope this helps!

Looking for more resources to help you use Dataiku effectively and upskill your knowledge? Check out these great resources: Dataiku Academy | Documentation | Knowledge Base

Level 2

if you don't want to use python, you can do it in a formula step

inc(concat(datePart(audt_start_dt_max,"year"),"-",mod(datePart(audt_start_dt_max,"Month"),12)+1,"-01"),-1,"day")

Level 1

This formula works perfectly until we are looking at December. A minor modification to to include December:
if(
datePart(audt_start_dt_max,"Month") == 12 ,
inc(concat(datePart(audt_start_dt_max,"year")+1,"-",mod(datePart(audt_start_dt_max,"Month"),12)+1,"-01"),-1,"day"),
inc(concat(datePart(audt_start_dt_max,"year"),"-",mod(datePart(audt_start_dt_max,"Month"),12)+1,"-01"),-1,"day")
)

Level 1

This formula brings the last day of the month, is there a way that we could get the last working/business day of the month?

Dataiker

You can get the last working day of the month using this custom Python function:

``````import datetime
from datetime import timedelta

def process(row):
current = datetime.datetime.strptime(row["line"],'%m/%d/%y')
EndOfMonth = (datetime.datetime(current.year + int(current.month / 12), ((current.month % 12) + 1), 1) - datetime.timedelta(days=1))
if EndOfMonth.isoweekday() >5:
EndOfMonth = EndOfMonth - datetime.timedelta(days=EndOfMonth.isoweekday()-5)
row["EndWeek"]= datetime.datetime.strftime(EndOfMonth,"%m/%d/%y")
return row``````

Level 1

@CatalinaSthanks very much, that works perfectly for the weekends but, besides the weekends, I would like to consider the holidays too. I tried the "Flag holidays" processor but it doesn't includes brazilian holidays, do you have any ideas that I could adopt?

PS: I'm from Brazil.

if you are comfortable with python recipes you might try using the python library https://pypi.org/project/holidays/

or if you can find or create a table of Brazilian holidays you could join this to your dataset.

just a couple of ideas.  Let us know how you get on with this.

--Tom
Level 1

@tgb417  Thank you! I guess I will go with the table of Brazilian holidays option.