Color specific datatable cells based on their factor values

I don't necessarily recommend this approach because I am fairly against embedding one language in another, but I think it solves your ask. You can probably save the JavaScript to a file and load that into a variable, and that would be a better approach I think, but, in the interest of a self-contained solution, I have inlined it.

Based on the DT documentation here, https://rstudio.github.io/DT/functions.html, it is clear that the desired approach is to use formatStyle in some way.

DT provides a few convenience methods to be used with formatStyle, whose source is here: https://github.com/rstudio/DT/blob/0b9710f5a9391c634a3865961083740f1cbf657b/R/format.R, which I have based my solution on.

Basically, we need to pass some JavaScript to formatStyle which will do all of the table styling based on a variable called value. We will need to do something like this:

datatable(store2) %>% formatStyle(colnames(store2), backgroundColor=JS(jsFunc))

Where the variable jsFunc is some JavaScript string. Since the source indicates this string must be an expression rather than a statement, and because this will be somewhat complicated, we will use an anonymous function which is immediately evaluated to perform the logic. This function must take a value and return a color based on that value. Here is the function we need.

function(value){
  // find a number preceeded by an open parenthesis with an optional minus sign
  var matches = /\((-?\d+)/.exec(value);
  // ignore values which do not match our pattern, returning white as the background color
  if(!matches || matches.length < 2) { 
    return 'white'; 
  }
  // attempt to convert the match we found into a number
  var int = parseInt(matches[1]); 
  // if we can't ignore it and return a white color
  if(isNaN(int)) { 
    return 'white';
  } 
  // if the value is negative, return red
  if(int < 0) { 
    return 'red' 
  }
  // otherwise, by default, return green
  return 'green';
}

We want to call this function immediately, so we wrap it in parenthesis, and pass the value argument to it.

(function(value){
  // find a number preceeded by an open parenthesis with an optional minus sign
  var matches = /\((-?\d+)/.exec(value);
  // ignore values which do not match our pattern, returning white as the background color
  if(!matches || matches.length < 2) { 
    return 'white'; 
  }
  // attempt to convert the match we found into a number
  var int = parseInt(matches[1]); 
  // if we can't ignore it and return a white color
  if(isNaN(int)) { 
    return 'white';
  } 
  // if the value is negative, return red
  if(int < 0) { 
    return 'red';
  }
  // otherwise, by default, return green
  return 'green';
})(value)

We wrap this value in a multi-line R string, escape any backslashes and double-quotes (I avoided using them), and assign it to the value jsFunc.

jsFunc <- "(function(value){
  // find a number preceeded by an open parenthesis with an optional minus sign
  var matches = /\\((-?\\d+)/.exec(value);
  // ignore values which do not match our pattern, returning white as the background color
  if(!matches || matches.length < 2) { 
    return 'white'; 
  }
  // attempt to convert the match we found into a number
  var int = parseInt(matches[1]); 
  // if we can't ignore it and return a white color
  if(isNaN(int)) { 
    return 'white';
  } 
  // if the value is negative, return red
  if(int < 0) { 
    return 'red' 
  }
  // otherwise, by default, return green
  return 'green';
})(value)"

Finally, we can call formatStyle using this variable

datatable(store2) %>% formatStyle(colnames(store2), backgroundColor=JS(jsFunc))

That should give us a result like:

I have an answer using data.table and DT, its a bit tricky. Here it is:

library(data.table)
library(DT)

store <- setDT(store)
store[,plouf := paste0(tempr,"(",value,")")]
store[,color := ifelse(value > 0,1,0)]

table1 <- dcast(store[,.SD[1],.SDcols = c("product","shop"),by = plouf],product ~ shop,value.var = "plouf")
table2 <- dcast(store[,.SD[1],.SDcols = c("product","shop","color"),by = plouf],product ~ shop,value.var = "color")
table2[,names(table2)[-1] :=  lapply(.SD,function(x){ifelse(is.na(x),0,x)}),.SDcols = names(table2)[-1] ]
setnames(table2,paste0(names(table1),"_col"))

plouf <- cbind(table1,table2[,-1])

datatable(plouf) %>% 
  formatStyle(names(table1)[-1],paste0(names(table1)[-1],"_col"), backgroundColor = styleEqual(c(0, 1), c('red', 'green')))

The explanation: first, i redo your store2 table using dacst (passing to large format), which is more convenient, and that i ll reuse to define the colors:

store <- setDT(store)
store[,plouf := paste0(tempr,"(",value,")")]
table1 <- dcast(store[,.SD[1],.SDcols = c("product","shop"),by = plouf],product ~ shop,value.var = "plouf")

   product  fghj    sad sadas xzzv
1:      ab    NA  35(0)    NA   NA
2:      ac 14(4) 14(-6) 24(8) 5(0)

Here store[,.SD[1],.SDcols = c("product","shop"),by = plouf] allow to have only one line per data you want, to avoid replica

Then I do the same, but with a variable that gives the color (1 for green, 0 for red):

store[,color := ifelse(value > 0,1,0)]
table2 <- dcast(store[,.SD[1],.SDcols = c("product","shop","color"),by = plouf],product ~ shop,value.var = "color")

   product fghj sad sadas xzzv
1:      ab   NA   0    NA   NA
2:      ac    1   0     1    0

I tranform Nas into 0 in all column exept the first one:

table2[,names(table2)[-1] :=  lapply(.SD,function(x){ifelse(is.na(x),0,x)}),.SDcols = names(table2)[-1] ]

   product fghj sad sadas xzzv
1:      ab    0   0     0    0
2:      ac    1   0     1    0

and modify the names of the second table (giving the colors):

setnames(table2,paste0(names(table1),"_col"))

you then bind the two

plouf <- cbind(table1,table2[,-1])

and use DT where you specify different column : one for the target, one for the background color

datatable(plouf) %>% 
  formatStyle(names(table1)[-1],paste0(names(table1)[-1],"_col"), backgroundColor = styleEqual(c(0, 1), c('red', 'green')))

I had to change your data, because there wasn't an unique set for each shop. Here are the data:

product<-c("ab","ab","ac","ac","ac","ac")
shop<-c("sad","sad","sad","sadas","fghj","xzzv")
category<-c("a","a","a","b","b","b")
tempr<-c(35,35,14,24,14,5)
value<-c(0,0,-6,8,4,0)
store<-data.frame(product,shop,category,tempr,value)