Plotly stacked bar chart add_trace loop issue

Question

I am building a stacked bar chart in plotly and whenever I run my loop for add_trace, there is an issue as data of the previous trace seems to dissapear while names remain.

        p = plot_ly( x = rownames(dist_data), y = as.numeric(dist_data[,1]), type = 'bar', name = colnames(dist_data)[1])%>%
      layout(legend = list(x = 0.1, y = 0.9))

    for ( j in 2:length(colnames(dist_data)))
    {
      p =  add_trace(p, y = ~as.numeric(dist_data[,j]), type = 'bar',name = colnames(dist_data)[j]) %>%
        layout( barmode = 'stack')
    }
    p

I am wondering whether there is something wrong in the loop. When trying manually for adding trace1, works fine. When adding trace 2 (j=3), trace1's values become automatically equal to trace2's value.

UPD: When using dplyr, I have issues with some other parts of my code. Is there a solution without it?

Thank you for the help,

Answer 1

You are overwriting p at each loop iteration. Try:

p =  p %>% add_trace(y = ~as.numeric(dist_data[,j]), type = 'bar',name = colnames(dist_data)[j]) %>%
        layout( barmode = 'stack')

2nd Edit: not sure why this is happening. It works if you reshape to a long format dataframe which is better practice than using a for loop:

mtcars %>% 
  mutate(id = rownames(.)) %>%
  gather(key = "variable",value = "value",-id) %>% 
  plot_ly(x = ~id, y=~value, type="bar", color=~variable) %>%
  layout(barmode = "stack")

Edit:

dist_data=mtcars
p = plot_ly( x = rownames(dist_data), 
             y = as.numeric(dist_data[,1]), 
             type = 'bar', name = colnames(dist_data)[1]) %>%
  layout(legend = list(x = 0.1, y = 0.9))

for ( j in 2:length(colnames(dist_data))){
  p =  add_trace(p, y = ~as.numeric(dist_data[,j]), 
type = 'bar',name = colnames(dist_data)[j]) %>%
    layout( barmode = 'stack')
}
p