Split by Variable Columns in Power Query
Вставка
- Опубліковано 11 лип 2024
- Check out our newly launched M Language course ↗️ - goodly.co.in/learn-m-powerquery/
The default Split by feature in Power Query produces a fixed number of columns. In this videos I'll share an interesting M technique to split by Variable Columns in Power Query. Enjoy!
===== ONLINE COURSES =====
✔️ Mastering DAX in Power BI -
goodly.co.in/learn-dax-powerbi/
✔️ Power Query Course-
goodly.co.in/learn-power-query/
✔️ Master Excel Step by Step-
goodly.co.in/learn-excel/
✔️ Business Intelligence Dashboards-
goodly.co.in/learn-excel-dash...
===== LINKS 🔗 =====
Blog 📰 - www.goodly.co.in/blog/
Corporate Training 👨🏫 - www.goodly.co.in/training/
Need my help on a Project 💻- www.goodly.co.in/consulting/
Split by Delimiter Blog - www.goodly.co.in/split-by-var...
===== CONTACT 🌐 =====
Twitter - / chandeep2786
LinkedIn - / chandeepchhabra
Email - goodly.wordpress@gmail.com
===== CHAPTERS =====
0:00 Intro
0:37 Split by Columns - The Problem
2:23 The Logic
4:22 The Solution
12:09 Testing the Solution
13:02 My Courses
===== WHO AM I? =====
A lot of people think that my name is Goodly, it's NOT ;)
My name is Chandeep. Goodly is my full-time venture where I share what I learn about Excel and Power BI.
Please browse around, you'd find a ton of interesting videos that I have created :) Cheers!
- - - - -
Music By: "After The Fall"
Track Name: "Tears Of Gaia"
Published by: Chill Out Records
- Source: goo.gl/fh3rEJ
Official After The Fall UA-cam Channel Below
ua-cam.com/channels/GQE.html...
License: Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)
Full license here: creativecommons.org/licenses - Наука та технологія
Check out our newly launched M Language course ↗ - goodly.co.in/learn-m-powerquery/
Fabulous! Thanks for the clear explanation.
Great solution as always. I've learned so much on your channel. Thanks for sharing
Great idea. I love the way you conserve steps and combine so much into each one. I do see an issue that will come up with steps beyond these dynamic columns. But I suppose that can be handled as the cases arise.
Thanks for demonstrating the procedure of solving the problem really amazing
mind blown. great tutorial and great explanation!!!!!
Really amazing. Dynamic columns, even outside split, remains a very challenging area - like pivoting and unpivoting a dynamic list of columns. This video definitely helps.
Always cutting-edge, Chandeep. Thanks!
Glad you like it 😊
wow, this just came just at the right time. I have a use case where I have to do a report on dynamic product category/products. Just subscribed. thanks!
Awesome Chandeep! Very clever.. as always. Thanks for walking through the steps.. well explained and easy to follow. Thumbs up!!
Glad you like it 😊
Glorious! Mastering as always Chandeep, many thanks for sharing!
Glad you like it Jose! 😊
Excelente, me ayudó el tutorial para aplicarlo en un reporte, saludos desde Perú.
Excellent way to deal with this types of queries. Thanks
Glad you like it 😊
Came across your video trying to solve this exact problem today. I needed to separate a list in a cell to separate columns prior to unpivoting. Your explanation was terrific and probably the most straightforward solution I've come across. Thanks!
Brilliant!! This is exactly what I was looking for about a year ago, but I couldn't quite work out how to do it. I can't wait to implement this. I really like the way you explained all the step. THANK YOU!!! I have just subscribed :)
Fantastic!
Thank you, this video has saved my day!
Glad it was helpful 💚
You are a wizard with this. Thanks for sharing!!!
Glad you like it 😊
Great video! Thanks for sharing!
This is great, can also be used with loading PDF, and so on, thank you!
As always, very clever. Mind blown regarding not having to to refer to the immediately proceeding applied step. Well done.
Glad you like it 😊
Great video, I'm sure these techniques can be used for many other problems.
Glad it was helful 😊
This technique blows my mind. I think M query can makes our work faster.. Thanks for sharing this knowledge. 👍
Thank you so much. You explain the concept so well.
I'm going to apply this to a problem I have. Thanks again!
Glad you like it
Fantastic! Beautiful! Thanks for sharing.
glad you like this
Thanks Chandeep for sharing this brilliant content.
Glad you liked it!
Just the tip I was looking for, Thank you! :)
Glad I could help!
Really enjoyed the video, and found the use of the functions a great learning experience.
For this example I think the list of names can be replaced with the number of columns, so you could have used the max number straight in the argument, although the video was more informative using the other way
= let
_max = List.Max(Table.AddColumn(Source,"Custom",each List.Count(Text.PositionOfAny([Hobbies],{","},Occurrence.All)))[Custom])+1
in
Table.SplitColumn(Source, "Hobbies",Splitter.SplitTextByDelimiter(",", QuoteStyle.Csv),_max)
You are... just unbelivable 🙂 Thank you!
Thanks Ales!
Thx from Brazil! 💚💛
Welcome!
Excellent. Thank you very much!
Glad you like it 😊
Excellent Chandeep 👍👍
Extraordinary result, thank you very much for sharing this type of solutions, personally it has helped me a lot.
Greetings from Chile
Glad it was helpful 💚
Excellent video...great explanation....subscribed.
Thank you!
Great content!
This great content, cheer for your good work! 👍🎉
Glad you like! 💚
Very, very cool and handy tip!!
Glad you liked it
As always amazing stuff Chandeep
Thank you 😊
Thanks, Chandeep, I liked the part which you use Text.PositionOf for AllOccurance but for split by variable columns why do you make it really complicated?!!!
Just remove the list in the last argument of Table.SplitColumn function. (then it will be dynamic for new columns)😉
Didn't know that.. thanks
Wow-- That works really well.. makes me wonder if lots of hardcoded lists in M are Optional Arguments
This only works if the first row happens to have the max delimiters required, I have tried, and lets say if the first record has 2 delimiters, then you get 3 columns and no more.
@@martyc5674 Of course, I reached to this point and the solution is making Custom column as =List.Count(List.Split([Hobbies],","))
And before split, we must sort z to a this custom column.
Still easier.
Very well put together video
Glad you enjoyed!
Awesome! 👍
Great presentation - excellent walkthrough of the critical thinking and problem-solving technique. I wonder if there's a way to do it in DAX, as well...
Also... I, too, enjoy Hulking. Excellent choice, Rehet.
Just amezing❤❤
Amazing....The only guy on UA-cam who can teach power query transformation like bread & butter... Kudus Chandeep for this awesome playlist. ❤❤
Thanks a ton 😉
Great Video!!
Thank you 💚
Thank you
Finished watching
Amazing ❤
Very Goodly,my good sir! 🔥👊🏽🔥
Thanks Oz. Glad to see you here!
Great 💯👍
Great video, thanks! Your son sounds cool 😀
Thanks Man!
Golden!
Thanks!
Wow just wow ...
Thanks 😊
Wow 😲 Chandeep Amazing!, what a technique you taught. Please solve my problem also which I commented in another's video. How can I pass dynamic list into table.combine function.
Great!
Thanks 😊
Great video
💚
Bravo
that's amazing
Thanks 😊
Cheers!
Cheers! Thanks!
You are genius
Thank you! 😊
I did this solution some time ago more for a "sport" activity. However, the easier way is to use to Text.Split which will convert the text into list, and then expand the list.
Very good
Thanks!
Great lesson once again.... Sir pls help to understand how get total of if we have numbers in place of Hobbies. Pls help sir...
The BEST TEATCHER EVER about Power Query (M) language. Loving all videos about it. Thank you a lot for your great and amazing commitment to help world community. I wish all the best to you and your family!
Glad you are loving the videos 😊
It is nice
Thanks!
Thanks for the explanation.
I've shorten it a bit.
This avoids me having to clean up added columns in the end. The 'n' column, that contains the number of occurrences for ';' exists only inside the MaxAnswers step.
I could also add it and then refer to a previous step, true...
MaxAnswers = List.Max(Table.AddColumn(Source, "n", each List.Count(Text.PositionOf([Answers], ";", Occurrence.All)) + 1)[n]),
HeadersAnswers = List.Transform({1..MaxAnswers}, each "Answer"&Text.From(_)),
The red color code is hard code,to modify them is a good job.
It is awesome chandeep sir,
Can you please help me how can we add multiple custom columns in single step at power query
Love it! but is there also a way to covert it each in a rows rather than columns?
This is amazing, exactly what I needed.
Can I ask, how do you come up with this logic/idea? Does it come naturally or..
thanks for great tutorial! what if there would be more columns like Hobbies_1, Hobbies_2 and few items in each of them?
Thank you so much! I am having a problem in that I need to split a column up to 15 characters for example. The column has characters up to 15 and then a space and then another set of characters but within those first 15 characters there can also be spaces, So I can’t split by the delimiter of a space. And I can’t split by 15 characters because sometimes it is less than 15 characters. I need to make the split dynamic but I am very lost on what to tell it to do.
Excellent video once again. Does Rehet means Happiness?
it means calm
Hi Chandeep,
Did you do a video on how to remove duplicates when column values can be swapped like (1,2) and (2,1) would be a duplicate. Thanks
💯👍
Very interesting tutorial.
But you didn't foresee that in the future you might list a little friend of your friends and that he will have other hobbies, for example football and Crying, just two hobbies, but football would deserve a new separate column and the current formula does not provide for this case .... :) :)
Greetings.
Is it possible to have unique values only in the columns?
In your example, the "sleeping" value would have its own column.
SAVING LIIIIIIVESSSSSSS
Thank You!
Super duber
Thanks!
I found a bug in the split function while working with a large dataset. I was manually applying a sequence of commands, with the split function in the middle. I was splitting a column using the delimiter ";;;". The columns had a varying number of delimiter occurrences, but I expected them to consistently generate 4 or 5 columns each.
Only later did I realize that the split function was only generating 2 columns. I had to re-run the split operation, and only then did it generate the correct number of columns. Now, whenever I apply the split command, I wait for it to show the detected number of columns before applying it again.
Now I want to trim (or do some other function) to all of the columns.. Can you think of a good way to do that?
This tutorial is great but it's not meeting my needs exactly. My issue is I have a list of names and I want to split them into columns by First Name, Middle Name or Initial, Last Name, and Suffix name. Not all have a suffix, middle int/name or period. when I split them last names go into the middle name column and I don't know how to bring it to the right column. When I split the delimiter one step at a time by the first name it's in the right column but when I split the second time I run into problems. How do I fix this?
Can you please send me some dummy data and describe the problem clearly - goodly.wordpress@gmail.com
This solved my main issue to learn m code how to modify but couldn't we split the hobies by delimiter by row and then we get a unique list and refer that list this would solve issue if you have duplicate entries...
Yes there could be many ways of solving such problems
Yes, And you know it all... You explain things very well..and i think far better than anyone else when it comes to power query.....
my report have an uneven space length.. is there a way to split them?
Hello where to find your fundamental course.
goodly.podia.com/power-bi-beginner-course
Awesome trick however is there a way to sort the information.
Let's say you asked 100 people and they chose a hobby but it appears any of them columns in the order it was split.
Maybe you want a column based on the hobby type.
Other functions, e.g. Text.Combine uses a list to reference the columns to be combined. So far I was unable to define a list similar to this example to get this function working. Was anybody more successful with Text.Combine? Seb
Sorry what is the issue that you're facing.
Text.Combine works on a list (and not a column references)
I have a table with three columns Col1 … Col3. I merge the columns into a new column via add column / merge column with gives this M code:
= Table.AddColumn(Source, "ColCombine", each Text.Combine({[Col1], [Col2], [Col3]}, "|"), type text)
I wanted to do something similar to your video that I have a formula like
= Table.AddColumn(Source, "ColCombine", each Text.Combine(ListColHeading, "|"), type text)
So far I did not manage to create a List that the content of [Col1], [Col2], [Col3] is referenced - did I explain okay?
A possible solution is using Table.CombineColumns instead of merging the columns into a new column. The steps to perform are
- Creating the list “ColList” from the tables header
- Transforming every column into type text using “ColList“ in combination with Table.TransformColumnTypes
- Applying Table.CombineColumns in combination with “ColList“
Code could look like this:
Source …
IntoText = Table.TransformColumnTypes(Source, List.Transform(ColList, each {_, type text})),
CombineCols = Table.CombineColumns(IntoText,ColList,Combiner.CombineTextByDelimiter(":", QuoteStyle.None),"ColsCombined")
I got the inspiration for this from the video 😉 ua-cam.com/video/1fn8fXYw6M4/v-deo.html
Apart from your video, I have a question, is there any way to fetch data from Facebook by using a power query?
www.goodly.co.in/facebook-dashboard-in-excel/
Hi I have some need in data
What is the difference between
each & _ ??
When use this or this ????
each is they keyword that let's you access each row of a table or a list.
_ is the keyword to reference each item in the list or table.
You made the step look fancy but it happens to me not remember the syntax.
Chandeep, how do i dynamically split up to the second last value in PBI for each record?
For e.g. I have a table with EE hierarchy for each user-
For John, hierarchy is User1.User2.User3.User4
If John is User4, I want to split his hierarchy till User3.
For Jane, hierarchy is User1.User2.User3.User4.User5
If Jane is User5, I want to split her hierarchy till User4
With the current setup, the columns are split into 5 columns by default, as there are 5 users in the table.
So even though John is User4, the columns are split into 5 columns, where the 4th column consists of John's name.
This is how the columns are split:
User 1 | User 2 | User 3 | User 4 | User 5
M1 | M2 | M3 | John | NULL
M1 | M2 | M3 | M4 | Jane
I want the columns to stop at User 3 for John, so column USER 4 should be NULL as well
Any way to split the columns based on each record, rather than looking at the maximum number of splits in the table?
#powerbinareal