SQL Start! 2021 - The hidden treasure inside power bi
0 likes46 views
This document contains an agenda for a session on using the M query language in Power BI. The agenda includes introductions and then covers topics like: the benefits of Power Query and M for transforming data; best practices for writing clean M code; how M can be used for complex data mashups and transformations; and a demo of using M to solve a challenging scenario of transforming web page source code data. References for further reading on M and Power BI are also provided at the end.
SQL Start! 2021 - The hidden treasure inside power bi
- 1. SQL START! 2021 – 10° ANNIVERSARY The Hidden Treasure inside Power BI Lorenzo Vercellati
- 2. Lorenzo Vercellati BI & Power BI Architect @ 20 years burning on “Data Realm” My little contribution to the community
- 3. Partners
- 4. Session Agenda •Data is Good, PowerQuery is Better, M is Best •PowerQuery & M Best Practices •M is for MashUp •M ission (Im)possible
- 5. M is Best
- 6. Data is Good Being able to access data sources of many different types is good Good data are rarities
- 7. PowerQuery is Better PowerQuery User Interface is •Office like •Easy-to-use •Powerful Make your data better
- 8. M is Best PowerQuery covers only 40% of M power M can simplify step sequence M makes possible really complex transformations With M you can manage errors
- 9. PowerQuery & M Best Practices
- 10. Nomen - Omen
- 11. No Space, please! let Source = ……., #”Removed Non-Key Columns” = Table.SelectColumns(Source,{… let Source = ……., Removed_NonKey_Columns = Table.SelectColumns(Source,{…
- 13. Formula Bar Always on Train yourself to understand the code behind the User Interface Verify what you are really doing
- 14. M is for MashUp
- 15. M is a functional Language Between let and in One step, One function (at least) StepName = Function(PreviousStepName, Parameters)
- 16. M is case sensitive Keep in mind! Text <> text DAX is case insensitive COD1 <> cod1 for M COD1 = cod1 for DAX
- 17. M Fundmentals
- 19. Lists {“Pippo”, “Pluto”, “Paperino”} {1..10} {{"Squadra.1", "Squadra"}, {"Squadra.2", "Schema"}}
- 20. Records
- 23. Scenario • Web page source Code • Over 3000 rows of complicated XML • Data of the same match in different tables • It’s mandatory to have Team Name as column not as general header • We need to intercept and interpretate also icons
- 24. Scenario
- 25. Scenario
- 26. Time to jump into the Demo
- 27. Reference
- 28. Reference • Chris Webb's BI Blog Chris Webb's BI Blog (crossjoin.co.uk) • The Excelguru Blog - More geeky stuff from the author of www.excelguru.ca...The Excelguru Blog | More geeky stuff from the author of www.excelguru.ca…
- 29. Any Questions?
- 30. Thank You For Attending lvercellati@lucient.com @supergimi LorenzoVercellati Data Pied Piper – Medium