How to check whether all elements of a nested list are a subset of another list in R

Question

I've tried a number of different methods for this including this stack but nothing is working quite correctly.

My dataframe "SiteVisits" (a small subset dput is at bottom) is made up of columns Date (class = date), TagID (class = numeric), SiteVisits (a list of character), and NumSites (class = numeric). Each row lists all sites where an individual organism ( TagID ) is found for each Date.

I'd like to assign whether a tag spent the entire day "inside", "outside", or "transiting" based on the sites it visited. It can only be "inside" if it never visits an outside site, and it can only be "outside" if it never visits an inside site

First, I'd like to determine whether ALL the sites for a TagID for a Date are included in this list:

inside <- list(c("Release","IC1", "IC2", "IC3","RGD1"))

If TRUE SiteVisit$Location = "INSIDE" ELSE test whether ALL sites for a TagID for a Date are contained within this list:

outside <- list(c("ORS1","WC1","WC2","WC3","RGU1","ORN1","ORN2","ORS3","GL1","CVP1","CLRS"))

If TRUE SiteVisit$Location = "OUTSIDE" ELSE SiteVisit$Location = "TRANSITING"

I've tried a number of different dplyr and base versions to accomplish this, but none seem to get it right. I think it's because I'm not correctly checking through each element of SiteVisit$SiteVisits

My current attempts are:

SiteVisit <- SiteVisit %>%
  mutate(Location = ifelse(all(SiteVisits[[]] %in% inside), "INSIDE",
                           ifelse(all(SiteVisits[[]] %in% outside),"OUTSIDE","TRANSITING")))

which yields all "INSIDE"

and

SiteVisit <- SiteVisit %>%
  mutate(Location = ifelse(all(SiteVisits[] %in% inside), "INSIDE",
                           ifelse(all(SiteVisits[] %in% outside),"OUTSIDE","TRANSITING")))

which yields all "TRANSITING"

also, attempting to do this in a for loop doesn't quite work

for (i in 1: nrow(SiteVisit)) {SiteVisit$Inside <- 
all(SiteVisit$SiteVisits[[i]] %in% inside)}

yields all FALSE while

all(SiteVisit$SiteVisits[[2]] %in% inside)

is TRUE

Here's a small subset of my dataframe "SiteVisit" dput:

structure(list(Date = structure(c(15828, 15828, 15847, 15847, 
15847, 15847, 15847, 15847, 15848, 15848, 15848, 15848, 15848, 
15848, 15848, 15848, 15849, 15849, 15849, 15849, 15849, 15849, 
15849, 15850, 15850, 15850, 15850, 15850, 15850, 15850, 15851, 
15851, 15851, 15851, 15851, 15851, 15851, 15851, 15852, 15852, 
15852, 15852, 15852, 15852, 15852, 15853, 15853, 15853, 15853, 
15853, 15853, 15853, 15853, 15853, 15854, 15854, 15854, 15854, 
15854, 15854, 15854, 15854, 15855, 15855, 15855, 15855, 15855, 
15855, 15855, 15855, 15855, 15855, 15855, 15855, 15855, 15855, 
15856, 15856, 15856, 15856, 15856, 15856, 15856, 15856, 15856, 
15856, 15856, 15856, 15856, 15857, 15857, 15857, 15857, 15857, 
15857, 15857, 15857, 15857, 15857, 15857), class = "Date"), TagID = c(5717.06, 
6277.06, 5073.06, 5717.06, 11121.1, 11191.1, 11387.1, 11415.1, 
5717.06, 6277.06, 11121.1, 11191.1, 11219.1, 11289.1, 11387.1, 
11415.1, 5717.06, 11121.1, 11191.1, 11219.1, 11289.1, 11387.1, 
11415.1, 5717.06, 11121.1, 11191.1, 11219.1, 11289.1, 11387.1, 
11415.1, 5717.06, 11121.1, 11191.1, 11219.1, 11289.1, 11317.1, 
11387.1, 11415.1, 5717.06, 6277.06, 11191.1, 11219.1, 11289.1, 
11387.1, 11415.1, 5717.06, 6277.06, 9015.01, 9833.06, 11191.1, 
11219.1, 11289.1, 11387.1, 11415.1, 5717.06, 6277.06, 9015.01, 
11191.1, 11219.1, 11289.1, 11387.1, 11415.1, 5641.22, 5717.06, 
6221.06, 6277.06, 7909.22, 9015.01, 9833.06, 11121.1, 11191.1, 
11219.1, 11289.1, 11317.1, 11387.1, 11415.1, 5717.06, 6277.06, 
6529.06, 8119.01, 8545.06, 9015.01, 9497.06, 9833.06, 11191.1, 
11219.1, 11289.1, 11387.1, 11415.1, 5717.06, 6277.06, 6529.06, 
9015.01, 9497.06, 9833.06, 11191.1, 11219.1, 11289.1, 11387.1, 
11415.1), SiteVisits = list("Release", "Release", c("IC2", "IC1", 
"Release"), "IC3", "WC2", "RGD1", c("WC1", "WC3"), "WC3", "IC3", 
    "IC3", "WC2", "RGD1", "IC2", "IC1", "WC1", "WC3", "IC3", 
    "WC2", "RGD1", c("IC2", "IC1"), "IC1", "WC1", "WC3", "IC3", 
    "WC2", "RGD1", "IC2", "IC1", "WC1", "WC3", "IC3", "WC2", 
    "RGD1", "IC2", "IC1", "WC1", "WC1", "WC3", "IC3", "IC3", 
    "RGD1", "IC2", "IC1", "WC1", "WC3", "IC3", "IC3", c("IC3", 
    "Release"), c("IC3", "IC2", "IC1", "Release"), "RGD1", "IC2", 
    "IC1", "WC1", "WC3", "IC3", "IC3", c("IC3", "IC2"), "RGD1", 
    "IC2", "IC1", "WC1", "WC3", "Release", "IC3", "Release", 
    "IC3", c("RGD1", "Release"), c("IC3", "IC2"), c("IC3", "IC1"
    ), "WC2", "RGD1", "IC2", "IC1", "WC1", "WC1", "WC3", "IC3", 
    "IC3", c("RGD1", "Release"), c("RGD1", "Release"), "Release", 
    c("IC3", "IC2", "IC1"), "Release", c("IC3", "IC2", "IC1", 
    "RGD1"), "RGD1", "IC2", "IC1", "WC1", "WC3", "IC3", "IC3", 
    "RGD1", c("IC3", "IC2", "IC1"), "RGD1", c("IC3", "IC1", "RGD1"
    ), "RGD1", "IC2", c("IC2", "IC1"), "WC1", "WC3"), NumSites = c(1L, 
1L, 3L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 4L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 1L, 
3L, 1L, 4L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L, 1L, 3L, 1L, 1L, 
2L, 1L, 1L)), row.names = c(NA, -100L), groups = structure(list(
    Date = structure(c(15828, 15847, 15848, 15849, 15850, 15851, 
    15852, 15853, 15854, 15855, 15856, 15857), class = "Date"), 
    .rows = list(1:2, 3:8, 9:16, 17:23, 24:30, 31:38, 39:45, 
        46:54, 55:62, 63:76, 77:89, 90:100)), row.names = c(NA, 
-12L), class = c("tbl_df", "tbl", "data.frame"), .drop = TRUE), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"))

Answer 1

The following works once inside and outside are stored as an array and not a list

inside <- c("Release", "IC1", "IC2", "IC3", "RGD1")
outside <- c("ORS1", "WC1", "WC2", "WC3", "RGU1", "ORN1", "ORN2", "ORS3", "GL1", "CVP1", "CLRS")

df1$Location <- lapply(df1$SiteVisits, function(x) ifelse(all(x %in% inside), "INSIDE", ifelse(all(x %in% outside), "OUTSIDE", "TRANSIT")))

Answer 2

Want an answer that's about 1/100th as fast? (Not a typo*, this is way worse than manotheshark's answer, but it works on your data structured as it was). *it was a typo! 1/100th not 1/10th

for (i in 1:nrow(SiteVisit)) {
  SiteVisit_test$Location[i] <- if (all(unlist(SiteVisit[i, ]$SiteVisits) %in% unlist(inside))) {
    "INSIDE"
  } else if (all(unlist(SiteVisit[i, ]$SiteVisits) %in% unlist(outside))) {
    "OUTSIDE"
  } else {"TRANSITIONING"}
}

Benchmarks for the 2 approaches:

microbenchmark(
  for_statement = for (i in 1:nrow(SiteVisit)) {
    SiteVisit_test$Location[i] <- if (all(unlist(SiteVisit[i, ]$SiteVisits) %in% unlist(inside))) {
      "INSIDE"
    } else if (all(unlist(SiteVisit[i, ]$SiteVisits) %in% unlist(outside))) {
      "OUTSIDE"
    } else {"TRANSITIONING"}
  },
  lapply_statemnt = lapply(SiteVisit$SiteVisits, function(x) ifelse(all(x %in% inside2), "INSIDE", ifelse(all(x %in% outside2), "OUTSIDE", "TRANSIT")))
)

Unit: microseconds
            expr     min      lq      mean  median       uq     max neval
   for_statement 28874.4 30082.0 32411.968 31008.3 33108.90 48878.1   100
 lapply_statemnt   268.4   284.2   346.201   295.5   310.85  4114.9   100

I don't really get why the lapply approach is so much faster here... probably because I'm unlisting for every i in the loop.