Add "Run R Script" option to Excel 2016 and Excel 2013 "Power Query" add-in, as it is for Power BI.
In July 2016, a "Run R Script" option was added to Power BI Desktop. This option would be a huge benefit to have in Excel 2016 and Excel 2013 Power Query add-in also.
Thanks for your passion around this. We’ve been discussing this item off and on for some time now and we’d love to get your input.
The reason we’ve resisted this is because the R runtime & various packages (and updating their versions) aren’t straightforward for other users to install. This limits sharing such queries across a workgroup.
What do you think?
For others reading this that’d like us to prioritize this work, please be sure to add your vote since we’re prioritizing our backlog based on these.
Lead Program Manager
Please do allow access to R from Excel. Extremely useful in Power BI but sometimes i need the same features from Excel and since Power Query seems to be "the same" it would be ideal to have access to R from both.
James Lee commented
Definitely need R.Execute functionality in Excel Power Query!
If you really really need R / Python / whatever in Excel, it's quite straightforward to roll your own. Have a look at Excel-DNA: https://github.com/Excel-DNA
There's even an example R-integration project: https://github.com/Excel-DNA/Samples/tree/master/UsingRDotNet
Kent Maxwell commented
The complexity of R didn't stop Microsoft from adding R capabilities to Power Query in Power BI. There is no reason that Power Query in Power BI should be any different than Power Query in Excel. Why?
For the love of all things holy ..... we need R scripts in excel
"The reason we’ve resisted this is because the R run-time & various packages (and updating their versions) aren’t straightforward for other users to install"
@Ashvini - All your customers are not VP's sitting in corner offices and pushing the forward button on Outlook
So don't worry about it not being straightforward for us
"The reason we’ve resisted this is because the R runtime & various packages (and updating their versions) aren’t straightforward for other users to install. This limits sharing such queries across a workgroup."
Since so many Microsoft apps support Open R, R should be installed and managed in one location on every desktop. The installation can be done by Windows, Office 365, Power BI, SQL Server, whatever. When an app is run, it checks for the install and version, and install/update as necessary (including packages). For the Microsoft apps that support R, other instances of R installed elsewhere should be ignored.
Add this option for R to Excel immediately.
Or even sooner.
pato lobos commented
Hi Excel Team,
Already more than 2 years and absolutely no progress! We need not only R but also Python support! How is it possible that power query in Power BI gets all the love and nothing in Excel?
Can we get an update ASAP?
Bruno Crotman commented
How is it going?
James Hinton commented
Imke Feldmann: "Wouldn't it be a good start to give us R in Power Query like in the query editor (!!) in Power BI (https://powerbi.microsoft.com/en-us/documentation/powerbi-desktop-r-in-query-editor/) ?
Can well imagine that the integration of plots into Excel is more complicated, but the integration into the M-scripts must be almost straightforward - or am I mistaken here?"
Will Harvey commented
It would be a huge leap for Excel if it had R and Python like Power BI Desktop.
Any progress in this?
"The reason we’ve resisted this is because the R runtime & various packages (and updating their versions) aren’t straightforward for other users to install"
Perhaps R Open and associated packages should be installed with Office 365. Updates can be part of Office 365 updates. Then you can add the R.Execute function to M in Excel, which would allow folks to build cross-platform solutions using M & R.
pato lobos commented
As many have said R (and Python) in Excel is an evolutionary step. I can't not tell you how important this is. I run here in Oslo the UseR! chapter and I work as IT consultant; this thing will solve so many problems and will open insights to upper management is so many fields and companies. This lack of R and Python is a big issue, you might have only 136 votes, but we are not "a regular vote" the R community is small with a massive impact worldwide, so please don't evaluate us on that. You have a comment from Imke Feldmann check her out and the positive impact she has over 10 of thousand of Microsoft user, maybe I can get on board Marco Russo and Alberto Ferrari on this one, maybe that can help ;-) ?
Paul Mayfield commented
I understand the resistance, because it is a global issue for an organization to solve across applications. However, I don't think Excel is the place to solve it (nor within Power BI). The scripting to validate the library versions already exists; it is up to the user or organization to keep controls in place and use the tool appropriately. Giving us the tools at all would be great!
I don't see this problem as being much different than sharing macro-enabled workbooks, sharing with people who have older versions of Excel without Power Query or Power Pivot installed/enabled, etc.
For workers who are doing the work starting from the data wrangling all the way to the report-writing, themselves, it wouldn't make a difference at all; as what they would be sharing in the end is a word document, powerpoint, or an online dashboard that doesn't really reach back to the query stage.
Bruno Crotman commented
It will be a GREAT achievement. Power Query is great for Data Wrangling. Sometimes even better than R. Its a very well built tool.
Adding the possibility to run R scritps will bring all the power of R. The tools are very complementary.
Regarding the difficulty of installing the libraries, I do not think this will be a problem. Many times the worksheets are shared but the update process is done by one of the users. Other times Excel serves as an environment to create a prototype that will be imported by Power BI. Anyway... we will manage that. Dont worry about it. This is a huge acomplishment! We will use it a lot!
I would imagine that adding package management interface, similar to what you guys have done in RTVS using timemachine and miniCRAN. Other examples are packrat and Rstudio package management interface. Options: embed a package - most packages are under 5mb and could very well reside within zipped structure of xlsx. Larger packages may leave a mandatory script header for installing them from MRAN, when needed. There can be many solutions, I am sure you guys can work out with Azure/PowerBI/RevolutionAnalytics team. We are cheerleading to the expansion of #rstats.