Recreating the Chaos of Taco Hat Cake Gift Pizza in R

🌮🎩🍰🎁🍕 Taco Hat Cake Gift Pizza with Monkey, Ninja, and Unicorn

Over the Christmas holiday, my niece and nephew introduced me to Taco Cat Goat Cheese Pizza, a game that’s equal parts chaotic and hilarious. Imagine trying to slap a card while screaming “TACO!” but your hand freezes like it forgot its job. 🖐️🌮

My niece and nephew absolutely crushed me — I like to blame that I’ve had one too many cookies and was too slow 🍪🍫🍷.

But even in defeat, I had so much fun that I decided I needed a version of this game at home. So, I went ahead and bought Taco Hat Cake Gift Pizza (because, let’s be honest, who can resist a ninja 🥷 and unicorn 🦄?).

Crafting the Card Deck

Pakcages Used in This Blog Post

library(tidyverse) # Easily Install and Load the 'Tidyverse'
library(ggforce) # Accelerating 'ggplot2'
library(ggtext) # Improved Text Rendering Support for 'ggplot2'
library(ggimage) # Use Image in 'ggplot2'
library(fs) # Cross-Platform File System Operations Based on 'libuv'

When I opened the game, I found myself fascinated by the simple yet clever design of the card deck. It’s a perfect balance of “base” cards (like Taco, Hat, Cake, Gift, and Pizza) and “special” cards (Monkey, Unicorn and Ninja). Naturally, my data-nerd instincts kicked in, and I thought: Why not recreate this deck as a dataset in R?

Using the tidyverse, I created a small dataset to represent the deck. Instead of manually duplicating rows to match the card counts, I used one of my favorite hidden gems in tidyverse: uncount() in tidyr package.

Creation of the Card Deck

# Define the palette
pal5 <- c(taco="#2f4858",hat="#33658a",cake="#86bbd8ff",pizza="#f6ae2dff",gift="#f26419")
base_img_path <- path(here::here(),"posts","2024_12","images")

# Sample card composition data
cards_comp <- tibble(
  card = c("Taco", "Hat", "Cake", "Gift", "Pizza", "Monkey", "Ninja", "Unicorn"),
  emoji = c("🌮", "🎩", "🍰", "🎁", "🍕", "🐵", "🥷", "🦄"),
  emoji_img = c(taco=path(base_img_path,"taco.png"),
                hat=path(base_img_path,"hat.png"),
                cake=path(base_img_path,"cake.png"),
                gift=path(base_img_path,"gift.png"),
                pizza=path(base_img_path,"pizza.png"),
                monkey=path(base_img_path,"monkey.png"),
                ninja=path(base_img_path,"ninja.png"),
                unicorn=path(base_img_path,"unicorn.png")),
  type = c(rep("base",times=5),rep("special",times=3)),
  count = c(11,11,11,11,11,3,3,3)
)

###. Here's the power of Uncount function, opposite of dplyr::count()
cards <- cards_comp |> uncount(count)

### After cards have been unnested, assigning background color
cards$color <- c(rep(pal5["taco"],times=8),c(pal5[c("pizza","gift","hat")]),
          rep(pal5["hat"],times=9),c(pal5[c("taco","cake")]),
          rep(pal5["cake"],times=8),c(pal5[c("taco","gift","hat")]),
          rep(pal5["gift"],times=8),c(pal5[c("pizza","taco","pizza")]),
          rep(pal5["pizza"],times=8),c(pal5[c("hat","taco","gift")]),
          c(pal5[c("gift","pizza","hat")]),
          c(pal5[c("hat","pizza","taco")]),
          c(pal5[c("taco","cake","hat")]))

Spotlight on uncount()

If you’ve ever struggled to expand a dataset by repeating rows based on a count column, uncount() is a lifesaver. In just one line, it takes a column like count and automatically duplicates rows the specified number of times. I actually didn’t know this function till fairly recently.

Visualizing the Card Composition

With the deck dataset ready, I couldn’t resist visualizing it. Using ggplot2 and ggimage, I created a colorful layout that shows the composition of cards in Taco Hat Cake Gift Pizza.

Here’s the plot:

Layout Out Base & Special Card

cards |>
  arrange(type) |>
  #slice_sample(n = nrow(cards)) |>
  #mutate(player=c(rep(c("A","B","C","D","E"),times=12),"X","X","X","X")) |>
  mutate(grp=if_else(type=="base",card,type)) |>
  mutate(grp=fct_inorder(grp)) |>
  group_by(grp) |>
  mutate(idx=row_number()) |>
  ggplot(aes(x=idx,y=grp)) +
  geom_tile(aes(fill=I(color)), width=0.75, height=0.97) +
  geom_image(aes(image=emoji_img, by="height"), size=0.1) +
  coord_fixed() +
  theme_nothing(base_family="Roboto Condensed") +
  geom_richtext(aes(x=0,y=grp,
                    label=str_glue("<b>{n}</b><br>{grp}<br>cards")),
                data = . %>% group_by(grp) %>% summarise(n=n(),det=paste(unique(card),collapse="<br>")),
                family="Roboto Condensed",
                fill="transparent",
                label.colour="transparent") +
  labs(title=".    Composition of Cards inside of Taco Hat Cake Gift Pizza")

Shuffling the Deck and Dealing to Players 🎲

Once the deck was ready, it was time to shuffle and deal. After all, what’s a card game without a bit of randomness? Using slice_sample(), I shuffled the cards, then distributed them to five imaginary players.

Here’s how the shuffled cards looked when dealt out, with any extra cards forming a “Left Over Pile”:

Shuffling and Dealing

# Shuffle the cards
shuffled_cards <- cards |> 
  slice_sample(n = nrow(cards))

player_num <- 5
shuffled_cards |>
  mutate(idx=row_number()-1,
         player = (idx%%player_num)+1) |>
  mutate(is_left_over=idx>=player_num*12) |>
  mutate(player=if_else(is_left_over,"Left Over Pile",str_c("Player ",player))) |>
  ggplot(aes(x=idx%%4,y=floor((idx%%12)/4))) +
  geom_tile(aes(fill=I(color)),width=0.97, height=0.97) +
  geom_image(aes(image=emoji_img, by="height"), size=0.12) +
  theme_nothing(base_family="Roboto Condensed") +
  facet_wrap(~player)

Final Thoughts

Recreating Taco Hat Cake Gift Pizza in R was a fun way to blend coding with a bit of chaos. From shuffling cards to imagining new twists, this exercise now sparked few different ideas on how to utilize this mini dataset — and maybe a few new rules for game night. 🌮🎩🍰