ggplot piecharts on a ggmap: labels destroy the small plots - r

Question

My ggmap on which I would like small piecharts with labels is generated with the code:
p <-
get_googlemap(
"Poland",
maptype = "roadmap",
zoom = 6,
color = "bw",
crop = T,
style = 'feature:all|element:labels|visibility:off' #'feature:administrative.country|element:labels|visibility:off' or 'feature:all|element:labels|visibility:off'
) %>%
ggmap() + coord_cartesian() +
scale_x_continuous(limits = c(14, 24.3), expand = c(0, 0)) +
scale_y_continuous(limits = c(48.8, 55.5), expand = c(0, 0))
I am trying to plot my small ggplot piecharts on a ggmap following the answer
R::ggplot2::geom_points: how to swap points with pie charts?
I prepare data as follows:
df <-
df %>% mutate(Ours = Potential * MS, Others = Potential - Ours) %>%
na.omit() %>% filter(Potential > 0) %>%
select(-L.p., -MS) %>%
group_by(Miasto) %>%
summarise_each_(vars = c("Potential", "Ours", "Others"),
funs = funs(Sum = "sum")) %>%
left_join(coordinatesTowns, by = c("Miasto" = "address")) %>%
distinct(Miasto, .keep_all = T) %>%
select(-X) %>% ungroup()
df <-df %>% gather(key=component, value=sales, c(Ours_Sum,Others_Sum)) %>%
group_by(lon, lat,Potential_Sum)
My data looks then like
tibble::tribble(
~Miasto, ~Potential_Sum, ~lon, ~lat, ~component, ~sales,
"Bialystok", 100, 23.16433, 53.13333, "Ours_Sum", 70,
"Bialystok", 100, 23.16433, 53.13333, "Others_Sum", 30,
"Bydgoszcz", 70, 18.00762, 53.1235, "Ours_Sum", 0,
"Bydgoszcz", 70, 18.00762, 53.1235, "Others_Sum", 70,
"Gdansk", 50, 18.64637, 54.35205, "Ours_Sum", 25,
"Gdansk", 50, 18.64637, 54.35205, "Others_Sum", 75,
"Katowice", 60, 19.02754, 50.25842, "Ours_Sum", 20,
"Katowice", 60, 19.02754, 50.25842, "Others_Sum", 40
)
The last line group_by is essential for generating plots that will be pasted into my map. (I suspected maybe here is the reason of my problems described below).
Instead of totals, I would like to provide labels for each share in a piechart
In this answer I found the syntax, that should add labels to the piecharts https://stackoverflow.com/a/22804400/3480717
Below is the syntax in my script the line with geom_text (commented with hash) if uncommented, causes my plots to disappear and a long list (16 entries) for all small plots, of warnings:
1: Removed 1 rows containing missing values (geom_col).
I presume the reason can be in the last line of preparing the data, grouping it for the plotting.
The line I mark with a hash is a problem. If I put the hash plots are correct, if I include it, trying to get the desired labels on the slices, plots disappear or are very narrow vertical slices.
df.grobs <- df %>%
do(subplots = ggplot(., aes(1, sales, fill = component)) +
geom_bar(position = "fill", alpha = 0.5, colour = "white", stat="identity") +
# geom_text( aes(label = round(sales), y=sales), position = position_stack(vjust = 0.5), size = 2.5) +
coord_polar(theta = "y") +
scale_fill_manual(values = c("green", "red"))+
theme_void()+ guides(fill = F)) %>%
mutate(subgrobs = list(annotation_custom(ggplotGrob(subplots),
x = lon-Potential_Sum/300, y = lat-Potential_Sum/300,
xmax = lon+Potential_Sum/300, ymax = lat+Potential_Sum/300)))
df.grobs
df.grobs %>%
{p +
.$subgrobs +
geom_col(data = df,
aes(0,0, fill = component),
colour = "white")+ geom_text(data=df, aes(label = Miasto),nudge_y = -0.15, size=2.5)}
Why is the line marked with a hash (if uncommented) destroying the plot instead of adding labels? It seems to completely redefine aesthetics.
EDIT: I modified the marked line, now label=sales and y=sales. Now if I comment the line, the plots are generated, if I uncomment it, the labels are generated in correct position but without plots. Why I cannot have both?

Answer 1

Short answer:
I think the problem is actually in your earlier line:
geom_bar(position = "fill", alpha = 0.5, colour = "white", stat="identity") +
If you change the position from fill to stack (i.e. the default), it should work properly (at least it did on mine).
Long(-winded) explanation:
Let's use a summarised version of the mtcars dataset to reproduce the problem:
dfm <- mtcars %>% group_by(cyl) %>% summarise(disp = mean(disp)) %>% ungroup()
# correct pie chart
ggplot(dfm, aes(x = 1, y = disp, label = factor(cyl), fill = factor(cyl))) +
geom_bar(stat = "identity", position = "stack") +
geom_text(position = position_stack(vjust = 0.5)) +
coord_polar(theta = "y") + theme_void()
# "empty" pie chart
ggplot(dfm, aes(x = 1, y = disp, label = factor(cyl), fill = factor(cyl))) +
geom_bar(stat = "identity", position = "fill") +
geom_text(position = position_stack(vjust = 0.5)) +
coord_polar(theta = "y") + theme_void()
Why does changing geom_bar's position affect this? If we look at the plot before the coord_polar step, things may become clearer:
ggplot(dfm, aes(x = 1, y = disp, label = factor(cyl), fill = factor(cyl))) +
geom_bar(stat = "identity", position = "stack") +
geom_text(position = position_stack(vjust = 0.5))
Check the bar chart's y-axis. The bars & the labels are correctly positioned.
Now the version with position = "fill":
ggplot(dfm, aes(x = 1, y = disp, label = factor(cyl), fill = factor(cyl))) +
geom_bar(stat = "identity", position = "fill") +
geom_text(position = position_stack(vjust = 0.5))
Your bar chart now occupies the range 0-1 on the y-axis, while your labels continue to occupy the original full range, which is much larger. Thus when you convert the chart to polar coordinates, the bar chart is squeezed to a tiny slice that becomes practically invisible.