Different behaviour in Paginated and Desktop - Bug?

[BM4291]

Hi,

Noticed something odd by chance the other day that doesn't really make much sense.

I have a data model with a table that is connected to my date dimension; the relationship is active.

When I pull the data into Desktop and filter by a flag in my date dimension, I get the behaviour I expect to see. When I run an identical query in Paginated or via DAX query in SSMS the query doesn't work as expected.

Ultimately, when I filter data in Desktop, I'm seeing a different result to an identical filter in Paginated and this very nearly led to an official publication going out with errors as the wrong months were being picked up.

 The 'LastMonthActual' filter shows 1 for any date in November 2019 and 0 for any other date; the value in the fact table is for October hence I would expect to see blank.

I've attached some screenshots below to show this a bit more clearly. Any help would be greatly appreciated.

[dax]

Hi BM4291,

This might be caused by the context in powerbi. In addition, , if possible, could you please inform me more detailed information (such as your sample data and expected output)? Then I will help you more correctly.

Best Regards,
Zoe Zhi

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[BM4291]

@dax  - It's difficult to provide much more here; I have a datetime field joined to my date dimension however when I slice that data in PowerBI Desktop it behaves in one way, when I slice it in PowerBI Paginated or SSMS it behaves another.

What do you mean when you say context? I'm testing like-for-like between the two.

I cannot share my data unfortunately but see below for some equivalent tables; bearing in mind, this is the only table that seems to be affected by this; other tables have a similar set up but are filtering when flags are changed i.e. LastMonthActual

dataInsight - Redacted; from SharePoint Online list - API 14

Title	Period	Summary	Id
Private	01/10/2019 00:00	October 2019 represents the busiest …	2
Total	01/10/2019 00:00	October 2019 represents the third busiest …	3

Dates - Very condensed version; written in M

Date	Day	Year	Month	MonthName	MonthYear	LastMonthActual
01-Nov-19	Fri	2019	11	Nov	Nov-19	1
02-Nov-19	Sat	2019	11	Nov	Nov-19	1
03-Nov-19	Sun	2019	11	Nov	Nov-19	1
04-Nov-19	Mon	2019	11	Nov	Nov-19	1
05-Nov-19	Tue	2019	11	Nov	Nov-19	1
06-Nov-19	Wed	2019	11	Nov	Nov-19	1
07-Nov-19	Thu	2019	11	Nov	Nov-19	1
08-Nov-19	Fri	2019	11	Nov	Nov-19	1
09-Nov-19	Sat	2019	11	Nov	Nov-19	1
10-Nov-19	Sun	2019	11	Nov	Nov-19	1
11-Nov-19	Mon	2019	11	Nov	Nov-19	1
12-Nov-19	Tue	2019	11	Nov	Nov-19	1
13-Nov-19	Wed	2019	11	Nov	Nov-19	1
14-Nov-19	Thu	2019	11	Nov	Nov-19	1
15-Nov-19	Fri	2019	11	Nov	Nov-19	1
16-Nov-19	Sat	2019	11	Nov	Nov-19	1
17-Nov-19	Sun	2019	11	Nov	Nov-19	1
18-Nov-19	Mon	2019	11	Nov	Nov-19	1
19-Nov-19	Tue	2019	11	Nov	Nov-19	1
20-Nov-19	Wed	2019	11	Nov	Nov-19	1
21-Nov-19	Thu	2019	11	Nov	Nov-19	1
22-Nov-19	Fri	2019	11	Nov	Nov-19	1
23-Nov-19	Sat	2019	11	Nov	Nov-19	1
24-Nov-19	Sun	2019	11	Nov	Nov-19	1
25-Nov-19	Mon	2019	11	Nov	Nov-19	1
26-Nov-19	Tue	2019	11	Nov	Nov-19	1
27-Nov-19	Wed	2019	11	Nov	Nov-19	1
28-Nov-19	Thu	2019	11	Nov	Nov-19	1
29-Nov-19	Fri	2019	11	Nov	Nov-19	1
30-Nov-19	Sat	2019	11	Nov	Nov-19	1

[BM4291]

As a workaround I've altered the DAX query itself in paginated to look at the date column in my insight table rather than the Dates table; it'd still be good to have an explanation as to what's going on if anyone knows!

EVALUATE SUMMARIZECOLUMNS(
	'dataTeamInsight'[Title]
	, 'dataTeamInsight'[Summary]

	//Filter against date table not working; manual DAX filter added
	, FILTER(
		'dataTeamInsight'
		, MONTH([Period]) = MONTH(TODAY())-1
	)
)