max or min per column by factor

Question

I have a dataframe A with n columns. I need to find max value (but not 0 when other values are less than 0) per column by factor B.

dataframe(A)

B   a    b
1   0    0
2   0    0
3   0    0
1  -0.1 0.1
2   0.2-0.3
3   0   1
1  -0.3 0.4
2  -0.5 0.2
3   0.1 0.2

The output im looking for looks like this

B  a     b
1  -0.3  0.4
2  0.2   0.2
3  0.1   1

I know that i can use aggregate function but it only works for one column at a time.
The algorithm for each column is:
1. neglect all 0
2. if all values<0 then take the min of the values, else take the max

Answer 1

Here is a solution with base R:

f1 <- function(x) { x1 <- x[x!=0]; if(all(x1<0)) min(x1) else max(x1) }
aggregate(cbind(a,b) ~ B, data=A, FUN=f1)

(The function f1() is taken from the answer of @akrun)
result:

#> aggregate(cbind(a,b) ~ B, data=A, FUN=f1)
#  B    a   b
#1 1 -0.3 0.4
#2 2  0.2 0.2
#3 3  0.1 1.0

data:

A <- structure(list(B = c(1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L), a = c(0, 
0, 0, -0.1, 0.2, 0, -0.3, -0.5, 0.1), b = c(0, 0, 0, 0.1, -0.3, 
 1, 0.4, 0.2, 0.2)), .Names = c("B", "a", "b"), class = "data.frame", 
row.names = c(NA, -9L))

Answer 2

We can try with data.table

library(data.table)
f1 <- function(x) {x1 <- x[x!=0];
                   if(all(x1<0)) min(x1) else max(x1)}
setDT(A)[, lapply(.SD, f1), by = B]
#   B    a   b
#1: 1 -0.3 0.4
#2: 2  0.2 0.2
#3: 3  0.1 1.0

---

Or with dplyr

library(dplyr)
A %>% 
    group_by(B) %>% 
    summarise_each(funs(f1))
# A tibble: 3 × 3
#      B     a     b
#   <int> <dbl> <dbl>
#1     1  -0.3   0.4
#2     2   0.2   0.2
#3     3   0.1   1.0

data

A <- structure(list(B = c(1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L), a = c(0, 
0, 0, -0.1, 0.2, 0, -0.3, -0.5, 0.1), b = c(0, 0, 0, 0.1, -0.3, 
 1, 0.4, 0.2, 0.2)), .Names = c("B", "a", "b"), class = "data.frame", 
row.names = c(NA, -9L))