Hey Naama -
This is a job better done in the DW or in some sort of an ETL process - we don't support this sort of functionality "out of the box".
Depending on your data sources and if you have SQL Server available, you might be able to make this work by using Custom SQL and leaning on SQL Server's OPENQUERY() function to hit the remote data sources. Assuming you'd configured SQL Server to connect to an Oracle data source (as "OracleDataSource") and MySQL (as "MySQLDataSource"), you could do something like this in Tableau:
SELECT * FROM OpenQuery (OracleDataSource,’SELECT F1, F2, F3 FROM FOO‘)
SELECT * FROM OpenQuery (MySQLDataSource,’SELECT F4, F5, F6 FROM BAR')
Is this a good approach? Nope :)
I've got a similar problem. I'm importing from 2 different datasets of order level sales from 2 different companies. one source is mysql, the other is mssql. the tables are all unique rows with a date/time, orderid, and sales$ as well as company name/channel. l and i want to manipulate the date/time files to graph the monthly weekly count and total$ of sales of the 2 companies and create calculated fields to sum the totals. can i do this? not sure what i can even link.
each set also has a column of State where the transaction took place, but i'm assuming that cannot be what i use to link
You should be able to use "data blending" to accomplish this. Are you familiar with that feature?
I am now. I'll spend some time learning more with the online videos. Thank you.
The challenge I have it that i want to link each dataset's date/time stamp field and they don't exactly match up. i was able to get datablending to work by adding separate new columns to each set for year, month, day and hour fields and then linking works however i cannot do date/time manipulation now against the worksheet. any ideas how to link on date/time stamp?
You'll need to do something like convert the date/time value into a string using the STR function:
....and then truncate some of the characters off the end using the MID() function.
You'll do this in both data sets, then "blend" on the result.
1 of 1 people found this helpful
It is really not required to break your date field into three parts, Change your workbook local setting Tableau automatically parse your date fields according to your country default date time standard.
For Ex.: I belongs to India and in my Country default date format is dd-mm-yyyy.
1. Excel Date Field is showing dd-mmm-yyyy
2 MSSQL Date field is yyyy-mm-dd
(In this case it will read from your computer date setting)
Database Date Time
The moment I will fetch this data into Tableau by default Tableau will parse my Date field according to my laptop date time setting.
Why it is happening because my workbook local setting is automatic
Now I will change this for my Country, it will change the date field aromatically as per India standard (dd-mm-yyyy)
As far as data union all concern, Tableau Supports Union All within the data set with same tables layout. In case of two data set you can use Full join it will behave like a Union all only.
Hope above two approach will solve your problem.
It won't really be like a union because you will have multiple fields with the same name. I have used this as a work around to this point but it is incredibly inefficient slows down Tableau considerably with large data sets.
Yes, I know that but again... I'm NOT LOOKING TO MAKE A JOIN.
I need a union and Tableau does not support cross-database unions.
I have multiple warehouse management systems' databases which are identical in their schema and tables.
Have you found a solution to this? I also need to union same table structure from different schema
I have a solution!
Background: I have four csv files in one location I want to union (no problem), but then I want to union a 5th csv file in a completely different location.
Solution: So I set up this second data connection and drag my 5th file into the window where I am prompted to join them. In the join select a field from the left and one from the right that have nothing to do with one another. Then select a full join. This gives a similar experience to a union, though your secondary file will have a separate set of columns. I created a variety of IFULL() calculated fields to essentially coerce these columns into one.