Plotly: dynamicTicks = TRUE for categorical axis

[stanstrup]

See the following example.
When you add dynamicTicks = TRUE The date is not converted properly on the x-axis.

library("reshape2")
library("ggplot2")
library("plotly")

test_data <-
  data.frame(
    var0 = 100 + c(0, cumsum(runif(49, -20, 20))),
    var1 = 150 + c(0, cumsum(runif(49, -10, 10))),
    date = seq(as.Date("2002-01-01"), by="1 month", length.out=100)
  )


test_data_long <- melt(test_data, id="date")  # convert to long format

ggplot(data=test_data_long,
       aes(x=date, y=value, colour=variable)) +
       geom_line()

ggplotly(dynamicTicks = TRUE)

[lgaliano]

The issue is also seen with categorical x axis:

library("ggplot2")
library("plotly")
ggplotly(ggplot(iris, aes(x=Species, y=Sepal.Length))+geom_point(), dynamicTicks = T)

[stanstrup]

And modified axis labels e.g.:
scale_y_continuous(labels = percent)

[cpsievert]

This is now fixed for dates, but still somewhat broken for categorical axes (in which case, I'm sure dynamicTicks are useful?)

[stanstrup]

This was solved for dates but not for POSIXct.

library("reshape2")
library("ggplot2")
library("plotly")

test_data <-
  data.frame(
    var0 = 100 + c(0, cumsum(runif(49, -20, 20))),
    var1 = 150 + c(0, cumsum(runif(49, -10, 10))),
    date = seq(as.Date("2002-01-01"), by="1 month", length.out=100) %>% as.POSIXct()
  )


test_data_long <- melt(test_data, id="date")  # convert to long format

ggplot(data=test_data_long,
       aes(x=date, y=value, colour=variable)) +
       geom_line()

ggplotly(dynamicTicks = TRUE)

Do you want a separate issue opened?

[cpsievert]

Ahh, thanks I see the issue...fix coming up

[stanstrup]

I am sorry but it is still not exactly right. We now start from 1970 somehow. But that is regardless of dynamicTicks.

library("reshape2")
library("ggplot2")
library("plotly")

test_data <-
  data.frame(
    var0 = 100 + c(0, cumsum(runif(49, -20, 20))),
    var1 = 150 + c(0, cumsum(runif(49, -10, 10))),
    date = seq(as.Date("2002-01-01"), by="1 month", length.out=100) %>% as.POSIXct()
  )


test_data_long <- melt(test_data, id="date")  # convert to long format

p <- ggplot(data=test_data_long,
       aes(x=date, y=value, colour=variable)) +
       geom_line()

pp <- ggplotly(p, dynamicTicks = TRUE)

pp$x$layout$margin$l <- pp$x$layout$margin$l + 30 # avoid cut axis titles
pp$x$layout$margin$b <- pp$x$layout$margin$b + 30

[cpsievert]

Thanks; good thing someone is paying close attention 😉

[fruce-ki]

Error in tickvals - val : non-numeric argument to binary operator

Is it possible to have dynamic ticks for the (numeric) Y-axis when the X-axis is categorical (meaning string/factor values, not dates -- ticks should not be changed on the X-axis)?