poppyrlots.R

library(dplyr)
library(lubridate)
library(ggplot2)
source("~/theme.R")

# skimr::skim(cycle_all)
# 
# ── Data Summary ────────────────────────
# Values   
# Name                       cycle_all
# Number of rows             5436     
# Number of columns          4        
# _______________________             
# Column type frequency:              
#   character                2        
# Date                     1        
# numeric                  1        
# ________________________            
# Group variables            None     
# 
# ── Variable type: character ─────────────────────────────
# skim_variable n_missing complete_rate min max empty
# 1 Transport             0             1  12  12     0
# 2 subdivision           0             1  13  48     0
# n_unique whitespace
# 1        1          0
# 2       18          0
# 
# ── Variable type: Date ──────────────────────────────────
# skim_variable n_missing complete_rate min       
# 1 adjusted_date         0             1 2025-01-01
# max        median     n_unique
# 1 2026-05-31 2025-09-15      302
# 
# ── Variable type: numeric ───────────────────────────────
# skim_variable  n_missing complete_rate mean   sd p0 p25
# 1 daily_subtotal         0             1 140. 122.  0  54
# p50  p75 p100 hist 
# 1 105 189. 1243 ▇▁▁▁▁

monthed <- cycle_all |> 
  summarise(.by=adjusted_date,
            totals = sum(daily_subtotal)) |> 
  mutate(monthly = floor_date(adjusted_date, unit = "month"))

densed <- function(x, datacombo = monthed){
  partial = datacombo |> filter(monthly == x)
  d = density(partial$totals)
  df <- data.frame(
    x = d$x,
    y = d$y
  )
  df$monthis = x
  return(df)
}

denser <- bind_rows(lapply(unique(monthed$monthly), densed))
densest <- denser |> 
  filter(x > 0) |> 
  mutate(Month = month(monthis),
         Year = year(monthis),
         xadjust = ifelse(Year == 2026,
                          .4 * y/max(y),
                          -0.4 * y/max(y)),
         xcord = Month + xadjust,
         ycord=x) |> 
  select(monthis,Month,Year,xcord,ycord)

top_edge = densest |> 
  summarise(.by=c(monthis,Month,Year),
            ycord=max(ycord)+.001) |> 
  mutate(xcord=Month)

bottom_edge = top_edge |> 
  mutate(ycord=0)

grfpols <- bind_rows(densest, top_edge, bottom_edge) |>
  arrange(ycord) |> 
  mutate(Year=factor(Year))

monthmeans <- monthed |> 
  summarise(.by=monthly,
            ycord=mean(totals)) |> 
  mutate(Year = factor(year(monthly)),
         Month = month(monthly),
         xcord = ifelse(Year == "2026",
         Month + 0.4, Month -0.4))
ggplot(grfpols) + 
  geom_polygon(aes(x=xcord,y=ycord, group=monthis,
                   fill=Year)) +
  theme_david() +
  geom_segment(data=monthmeans, aes(x=xcord,y=ycord,xend=Month,yend=ycord, linetype=Year)) +
  scale_linetype_discrete(name="Mean") +
  scale_fill_viridis_d(end=.85, name="Frequency") +
  labs(title="Ōtepoti/Dunedin Cycle Counts",
       y="Daily Totals", x=NULL,
       caption=make_footer("Source: DCC")) +
  scale_x_continuous(breaks=c(1,3,5),
                     labels=c("Jan","Mar","May"))
ggsave(filename="~/Desktop/ggsky.jpg", width=2016, height=1134, units="px")