Openning a ROOT file in R

[Javier_Galan]

I have found that there is a library named ROOTR that allows to use R code inside ROOT.

I am looking for something that allows me to do the opposite.

There is any R library that allows me to open a ROOT File, extract a TTree, and associate to dataframe, or data table, in R?

Thanks!

[ksmith]

Duplicates this posts: http://stackoverflow.com/questions/43988982/a-library-to-open-root-files-in-r

(Sorry I don’t know of a package, just wanted to link to your other posts in case it is answered there.)

[Graipher]

Yes, there exists such a library, https://github.com/rmatev/RootTreeToR (which has some fixes on top of the official repository).

There exists also a package that allows reading a ROOT tree into a dplyr data_frame, https://github.com/rmatev/dplyr.root

You can install them via:

#Interface between ROOT and R
devtools::install_github("rmatev/RootTreeToR")
#Read ROOT ntuple into dplyr data_frame
devtools::install_github("rmatev/dplyr.root")

Here is some example usage using the hsimple.root file from the tutorials folder:

library(dplyr)
library(dplyr.root)

ntuple <- tbl_rootchain("hsimple.root", "ntuple")
events <- ntuple %>%
    mutate(P=px**2 + py**2 + pz**2) %>%
    filter(random > 0.2) %>%
    select(i, P) %>%
    collect()

> ntuple
Source: root file [25,000 x 5]

# S3: tbl_rootchain
            px          py           pz    random     i
         <dbl>       <dbl>        <dbl>     <dbl> <dbl>
1   0.01940945  0.01182548 0.0005165689 0.2826178     0
2   0.32718953  0.03787821 0.1084877476 0.4849736     1
3  -0.29461166 -0.01054588 0.0869072452 0.5400437     2
4  -0.77458960  0.04845844 0.6023373008 0.6586366     3
5  -1.43079531  0.50906366 2.3063209057 0.5196721     4
6   0.28898728 -1.86474288 3.5607798100 0.3923140     5
7   1.68962383  0.39789826 3.0131516457 0.0303352     6
8   1.39157462  0.50942063 2.1959893703 0.9437168     7
9  -0.62253845  0.83793545 1.0896899700 0.6655639     8
10 -0.43862954 -1.09520090 1.3918609619 0.1822846     9
# ... with 24,990 more rows

> events
# A tibble: 20,073 × 2
       i            P
*  <dbl>        <dbl>
1      0 5.168357e-04
2      1 1.202573e-01
3      2 9.446012e-02
4      3 9.651475e-01
5      4 7.625437e+00
6      5 1.623993e+01
7      7 7.018359e+00
8      8 2.277114e+00
9     11 1.619623e+00
10    12 1.128433e+00
# ... with 20,063 more rows

The tree will only be read after the call to collect. You need to use collect(protect=F) if the resulting dataframe is larger than 2GB, if I remember correctly.

As you can see, you can use the normal mutate, filter and select syntax of dplyr.

[Graipher]

The question on SO was closed, because it was asking for a recommendation of a software library…

[ksmith]

I sometimes see people answer those kind of posts in comments. Thought it may be useful for others who find it. Your answer here was the most useful though.

[Javier_Galan]

Yes. Thats it! Thank you!