R script not reacting to change in selection

[andrnev]

Hi, I have created two R scripts script1 and script2

script1 saves the dataframe using saveRDS function into a local directory

script2 reads the saved dataframe and displays the data.

The problem is that script1 reacts to a selection I make via a "slicer" dropdown but script2 does not. Only after manually running script2 by clicking "run script" button do I get a reaction. Any idea how I could trigger script2 to react to the changed script1's saved data?

Code of script1

library(gridExtra) # Displaying data onto Power BI Desktop
library(dplyr) # For data munging

pathToProject <-"F:\\PowerBiDocuments\\"
df1 <- group_by(dataset, manufacturer) %>% summarise('avg_cty_mileage' = mean(cty), 'avg_hwy_mileage' = mean(hwy))
saveRDS(df1, paste0(pathToProject,"df1"))

Code of script2

require(gridExtra)
require(ggplot2)
pathToProject <- "F:\\PowerBiDocuments\\"
df2 <- readRDS(paste0(pathToProject,"df1"))
grid.table(df2)

[TomMartens]

Hey,

I just curious, why can't you just combine both scripts into one like so

library(gridExtra) # Displaying data onto Power BI Desktop
library(dplyr) # For data munging

df1 <- group_by(dataset, manufacturer) %>% summarise('avg_cty_mileage' = mean(cty), 'avg_hwy_mileage' = mean(hwy))
grid.table(df1)

Regards

Tom

[andrnev]

Well if I wanted to modify the data within script 2 and display both script 1 original data and script2 modified data then script 2 should understand that it needs to 'react' only after script 1 completes. Running script 1 first and then manually running script2 ensures the hierarchy is maintained I.e. script1 has to run first and then script2 will pull in the modified script1 data and further modifies it. If there was a way in power bi to maintainthis hierarchial triggering with reactivity it would be super.

[TomMartens]

Hey,

I'm not sure if this will help, but maybe you are not aware of the function "grid.arrange"

This little script

library(ggplot2)
library(gridExtra)

df <- data.frame(cat = c("A", "B"), amount = c(110, 120))


p1 <- ggplot()
p1 <- p1 + geom_col(data = df, aes(x = cat, y = amount))

# here starts the modification of the data.frame df that than can used with the function tableGrob 

# create a grob (graphical object) from a data frame
p2 <- tableGrob(df)

# this creates a single plot that contains "two different plots" that also can be arranged in almost
# any way you want use ?grid.arrange to the documentation of grid.arrange
grid.arrange(p1, p2)

creates this output

Maybe this is what you are looking for

Regards Tom

[andrnev]

Tom many thanks. Yes I did look at the grid.arrange. it definitely is one workaround solution. I guess I was looking for a more azure way of doing things where we can just plug in models. Would be a great feature to have.

[v-sihou-msft · Accepted solution]

@andrnev

No, it's not possible. You must click the "run script" or click "Refresh" to make the R visual take the slicer selection and execute the script again. You may create a feature request on idea.

Regards,

[andrnev]

ThanksSimon. I asked for the feature on the ideas section.