tests/gt-examples/05-html-email/emailing_table.R

In rstudio/gt: Easily Create Presentation-Ready Display Tables

library(gt)
library(blastula)
library(tidyverse)
library(glue)
library(scales)

# Create short labels for months (displays
# better for small plots in email messages)
initial_months <-
  c(" ",
    "J", "F", "M", "A", "M", "J",
    "J", "A", "S", "O", "N", "D",
    " "
  )

# Get the total pizzaplace sales in the
# 2015 year; format as a currency value
total_sales_2015 <-
  pizzaplace$price %>%
  sum() %>%
  comma_format(
    big.mark = ",",
    prefix = "$"
  )(.)

# Create a plot using the `pizzaplace` dataset
pizza_plot <-
  pizzaplace %>%
  mutate(type = str_to_title(type)) %>%
  mutate(date = as.Date(date)) %>%
  group_by(date, type) %>%
  summarize(Pizzas = n(), Income = sum(price)) %>%
  ungroup() %>%
  ggplot() +
  geom_point(aes(x = date, y = Income), color = "steelblue") +
  facet_wrap(~type) +
  scale_x_date(date_breaks = "1 month", date_labels = initial_months) +
  scale_y_continuous(labels = scales::comma) +
  labs(
    title = "pizzaplace: Daily Pizza Sales in 2015",
    subtitle = "Faceted by the type of pizza",
    x = NULL,
    y = "Number of Pizzas Sold") +
  theme_minimal() +
  theme(
    axis.text = element_text(color = "grey25"),
    axis.text.x = element_text(size = 6),
    axis.text.y = element_text(size = 8),
    strip.text.x = element_text(size = 9),
    axis.title = element_text(color = "grey25"),
    legend.title = element_text(color = "grey25"),
    legend.text = element_text(color = "grey25"),
    panel.grid.major = element_line(color = "grey75"),
    plot.title = element_text(color = "grey25"),
    plot.caption = element_text(color = "grey25"),
    plot.subtitle = element_text(color = "grey25"),
    plot.margin = unit(c(20, 20, 20, 20), "points"),
    legend.box.spacing = unit(2, "points"),
    legend.position = "bottom"
  )

# Make the plot suitable for mailing by
# converting it to an HTML fragment
pizza_plot_email <- blastula::add_ggplot(plot_object = pizza_plot)

# Create `sizes_order` and `types_order` to
# support the ordering of pizza sizes and types
sizes_order <- c("S", "M", "L", "XL", "XXL")
types_order <- c("Classic", "Chicken", "Supreme", "Veggie")

# Create a gt table that uses the `pizzaplace`
# dataset; ensure that `as_raw_html()` is used
# (that gets us an HTML fragement with inlined
# CSS styles)
pizza_tab_email <-
  pizzaplace %>%
  dplyr::mutate(type = stringr::str_to_title(type)) %>%
  dplyr::mutate(size = factor(size, levels = sizes_order)) %>%
  dplyr::mutate(type = factor(type, levels = types_order)) %>%
  dplyr::group_by(type, size) %>%
  summarize(
    pies = n(),
    income = sum(price)
  ) %>%
  dplyr::arrange(type, size) %>%
  gt(rowname_col = "size") %>%
  fmt_currency(
    columns = income,
    currency = "USD"
  ) %>%
  fmt_number(
    columns = pies,
    use_seps = TRUE,
    decimals = 0
  ) %>%
  summary_rows(
    columns = "pies",
    fns = list(TOTAL = "sum"),
    formatter = fmt_number,
    use_seps = TRUE,
    decimals = 0
  ) %>%
  summary_rows(
    columns = "income",
    fns = list(TOTAL = "sum"),
    formatter = fmt_currency,
    currency = "USD"
  ) %>%
  tab_options(
    table.width = px(300),
    summary_row.background.color = "#FFFEEE",
    row_group.background.color = "#E6EFFC",
    table.font.size = "small",
    heading.title.font.size = "small",
    heading.subtitle.font.size = "x-small",
    row_group.font.size = "small",
    column_labels.font.size = "small",
    data_row.padding = "5px"
  ) %>%
  cols_label(
    pies = "Pizzas",
    income = "Income"
  ) %>%
  tab_header(
    title = paste0("My pizza sales in 2015"),
    subtitle = "Split by the type of pizza and the size"
  ) %>%
  as_raw_html()

message_body <-
  glue::glue(
"Hello,

Just wanted to let you know that pizza
sales were pretty strong in 2015. When
I look back at the numbers, it's **{total_sales_2015}**
in sales. Not too bad. All things considered.

Here's a plot of the daily pizza sales. I
faceted by the type of pizza because I know
that's your preference:

{pizza_plot_email}

Here is a table that shows a breakdown
of the 2015 results by pizza size, split into
*Pizza Type* groups:

{pizza_tab_email}

I also put all the 2015 numbers into an R
dataset called `pizzaplace`. Not sure why
I did that, but I did. It's in the `gt`
package.

Talk to you later,

Alfonso"
)

# Create an email message using the `compose_email()`
# function from the blastula package
email <- blastula::compose_email(body = md(message_body))

email