Skip to contents

Formatted labels for endpoints of trend charts

Source: R/endpoint_lbls.R
endpoint_lbls.Rd

Easily put together labels for the endpoints of a chart, such as a line chart with outwardly-justified labels at each end. Labels with the formats {name}: {fun(value)} (long format) and {fun(value} (short format) will be created, such as "Connecticut: 12%" and "16%" respectively. Often one format will be used on one end and the other format on the other end, but long_side gives you some options around this. The intention around combining labels like this is so you can put labels on both ends, aligned nicely, in a single geom_text call.

Usage

endpoint_lbls(
  data,
  x,
  value,
  group,
  mult = 0.05,
  add = NULL,
  fun = NULL,
  long_side = c("right", "left", "both", "none")
)

Arguments

data

A data frame

x

Bare column name of the independent value which has the start and finish points (probably a date). If this isn't numeric, endpoint_lbls will try to coerce it to numeric, which may yield undesired results.

value

Bare column name of the value column

group

Bare column name of the name or grouping variable that should be shown in the labels.

mult

Numeric; ratio to the x variable by which labels should be offset. Default: 0.05.

add

Numeric; single value along the x variable by which labels should be offset. Defaults NULL. If both mult and add are given, mult takes precedence. Previously there was a frac argument which was supposed to act like mult but erroneously acted like add.

fun

A function, used to create value labels. scales::label_* functions will be very useful here. If NULL (the default), no formatting is done.

long_side

Character, either "right" (the default), "left", "both", or "none". For "left" or "right", this refers to whether the longer label should be on the right or the left, returning a short label on the opposite side. If "both", only long labels are returned; if "none", only short labels are returned. Regardless, labels are combined in a new column called lbl.

Value

A data frame: the original data frame passed in to data, with 3 additional columns:

  • x, the x-values with offsets added

  • just, a column of 0 or 1 giving the justification value dependent on which end the label will appear

  • lbl, a column of formatted label text

Examples

library(ggplot2)
# note that it will still be up to your judgment to set scale expansion,
# since that will depend on things outside the scope of just this function
cws_trend |>
    dplyr::filter(question == "local_govt_responsive", category == "Age") |>
    endpoint_lbls(
        value = value, x = year, group = group,
        fun = percent100
    ) |>
    ggplot(aes(x = year, y = value, color = group)) +
    geom_line() +
    geom_point(size = 3) +
    geom_text(aes(label = lbl, hjust = just, x = x)) +
    scale_x_continuous(
        expand = expansion(add = c(1, 3)),
        breaks = c(2015, 2024)
    ) +
    theme(legend.position = "none")


cws_trend |>
    dplyr::filter(question == "local_govt_responsive", category == "Age") |>
    endpoint_lbls(
        value = value, x = year, group = group, long_side = "both",
        fun = percent100, add = 0.4, mult = NULL
    ) |>
    ggplot(aes(x = year, y = value, color = group)) +
    geom_line() +
    geom_point(size = 3) +
    geom_text(aes(label = lbl, hjust = just, x = x)) +
    scale_x_continuous(
        expand = expansion(add = 3.5),
        breaks = c(2015, 2024)
    ) +
    theme(legend.position = "none")