I have an executable C program (./saucy) that I get from running the makefile of Saucy( source code here)
Running ./saucy graphfile
Gets me as output
(0 2)(3 4)
(0 1)(2 4)
How can I call this from an R script? I'm scratching my head even after reading this post.
I've only got the saucy command line wrapped into an R function of the same name and have only tested it on macOS and Ubuntu but it tracks with your sample output.
By "wrapped" I mean it does not call a binary, it's a C function called from R. It wasn't to difficult to strip away the cmdline cruft. The authors did a pretty decent job.
devtools::install_github("hrbrmstr/saucy")
library(saucy)
First example file:
graph1 <- saucy::saucy(system.file("extdata", "graphfile", package="saucy"))
graph1
## (0 2)(3 4)
## (0 1)(2 4)
##
## input_file: graphfile
## vertices: 5
## edges: 5
## group_size_base: 1
## group_size_exp: 1
## levels: 3
## nodes: 7
## generators: 2
## total_support: 8
## average_support: 4
## nodes_per_generator: 3.5
## bad_nodes: 0
str(graph1)
## List of 13
## $ input_file : chr "/Library/Frameworks/R.framework/Versions/3.4/Resources/library/saucy/extdata/graphfile"
## $ vertices : int 5
## $ edges : int 5
## $ group_size_base : num 1
## $ group_size_exp : int 1
## $ levels : int 3
## $ nodes : int 7
## $ generators : int 2
## $ total_support : int 8
## $ average_support : num 4
## $ nodes_per_generator: num 3.5
## $ bad_nodes : int 0
## $ printed_output : chr [1:2] "(0 2)(3 4)" "(0 1)(2 4)"
## - attr(*, "class")= chr [1:2] "saucy" "list"
Second example file:
graph2 <- saucy::saucy(system.file("extdata", "graphfile2", package="saucy"))
graph2
## (3 5)
## (3 6)(4 5)
## (0 1)
## (0 2)
##
## input_file: graphfile2
## vertices: 7
## edges: 7
## group_size_base: 4.8
## group_size_exp: 1
## levels: 5
## nodes: 13
## generators: 4
## total_support: 10
## average_support: 2.5
## nodes_per_generator: 3.25
## bad_nodes: 0
str(graph2)
## List of 13
## $ input_file : chr "/Library/Frameworks/R.framework/Versions/3.4/Resources/library/saucy/extdata/graphfile2"
## $ vertices : int 7
## $ edges : int 7
## $ group_size_base : num 4.8
## $ group_size_exp : int 1
## $ levels : int 5
## $ nodes : int 13
## $ generators : int 4
## $ total_support : int 10
## $ average_support : num 2.5
## $ nodes_per_generator: num 3.25
## $ bad_nodes : int 0
## $ printed_output : chr [1:4] "(3 5)" "(3 6)(4 5)" "(0 1)" "(0 2)"
## - attr(*, "class")= chr [1:2] "saucy" "list"
shatter is wrapped as well.
I can also make it take real graph structures at some point vs read serialized ones, but I don't know how popular this library/tools is so can't commit to a time-frame.
