Forum Discussion
DAX makes me crazy : CONVERT
Hello again 🙂
Finally I made all work this morning at 4AM (CET). The
"attribute" = CONVERT(column, STRING)
has been replaced by a
"attribute" = column & ""
...Something I use to do when I do not have the choice to use a var typed variable. Not clean but it works.
I had a second issue coming from 2 other columns (one in each SharePoint list) that are search typed, with the same source. As I explained, it works until I edit the DAX script but I needed to change something in it, so I took 2 attributes with a different name (instead of the same one to be taken into account by the NATURALLEFTOUTERJOIN), and I added a FILTER on [attribute1] = [attribute2].
Now I can finish the v2 of my dashboards.
Many thanks, again, for your advice about Power Query. Be sure this is something I will work on once the reports will be published (even if I do not eat caviar anymore since I discovered the female fishs were opened alive to get the eggs, then thrown back into the sea still alive but dying, thanks to humanity) !
Thanks again !
B.R.
Eric
Hello,
Many thanks for your answer. Why I clinged to DAX ? Just because it exists and I wanted to try. It worked with no real issue until today. To say the truth, it still works until I edit the piece of script (and if I do that, as I explained, both Convert() and Naturalleftouterjoin go to errors).
For the "join", I find it really like in SQL requests.
I did not know Power Query. I see it is a low / no code interface... I do not like that because it is full of traps (at least if one could code in C# inside !!!) and so complex to configure and slow to execute when you have something a little bit complex to do (it often take hours to do something new in Power Automate), but I will try it, thanks for the advice !
UPDATE : OK, I did not know that Power Query was the thing I also already use in Power BI, so that will not help.
Just one thing I have not said, it is that the source is a SharePoint list and not a database, I do not know if this is something that may have an impact or not... But never mind, I will try something else.
Eric
Community ChampionI once did what you do now, considering DAX as a cure-all until I found PQ (the Mashup language to be exact) is another caviar in PBI.
Like what you mentioned, PQ desguises with a low/no code UI😀; but its real power lies in its script editor. At least 90% of complicated missions of data ETL are to accomplish there.
Hello again 🙂
Finally I made all work this morning at 4AM (CET). The
"attribute" = CONVERT(column, STRING)
has been replaced by a
"attribute" = column & ""
...Something I use to do when I do not have the choice to use a var typed variable. Not clean but it works.
I had a second issue coming from 2 other columns (one in each SharePoint list) that are search typed, with the same source. As I explained, it works until I edit the DAX script but I needed to change something in it, so I took 2 attributes with a different name (instead of the same one to be taken into account by the NATURALLEFTOUTERJOIN), and I added a FILTER on [attribute1] = [attribute2].
Now I can finish the v2 of my dashboards.
Many thanks, again, for your advice about Power Query. Be sure this is something I will work on once the reports will be published (even if I do not eat caviar anymore since I discovered the female fishs were opened alive to get the eggs, then thrown back into the sea still alive but dying, thanks to humanity) !
Thanks again !
B.R.
Eric