ggplot2 geom_mosaic with weight variable - r

Question

I am trying to make a mosaic plot with ggplot2. I am using the bladdercancer data from the HSAUR3 package. I am looking to show the relationship between tumorsize and number, but I am not sure how to weight it. I know that the number in the sample with tumorsizes<=3cm is not the same as those with tumorsize>3cm. How do I incorporate that into my mosaic plot?
Here is what I did without weighting it.
library("ggplot2")
library("ggmosaic")
ggplot(data = bladdercancer, family=poisson()) +
geom_mosaic(aes(weight= 1 , x = product(tumorsize, number),
fill=factor(tumorsize)), na.rm=TRUE) +
labs(x="Number of tumors", title='Number of tumors vs Tumorsize') +
guides(fill=guide_legend(title = "Tumor Size"))

Answer 1

This may be late nevertheless I provide my suggestions since I am trying to do a similar thing. Below are two suggestions:
library(tidyverse)
first option:
bladdercancer %>%
group_by(tumorsize, number) %>%
# get frequencies/counts for each tumor size and for each number
summarise(n.cases = n()) %>%
ggplot() +
geom_mosaic(aes(weight = n.cases, x = product(number),
fill = factor(n.cases)), offset = 0) +
guides(fill=guide_legend(title = "Tumor Size")) +
labs(x="Number of tumors", title='Number of tumors vs Tumorsize') +
# remove background colour
theme_bw() +
theme(panel.grid.major = element_blank(),
# remove major and minor grids
panel.grid.minor = element_blank(),
# push title to the middle
plot.title = element_text(size = 10, hjust = .5))
where the categories within each column represent different counts for each tumor size e.g number 1 appears 15 times for <=3cm and 5 times for >3cm. I am however not able to partitions where the frequencies are the same, in this case number 3 and 4. Hence my option 2
second option:
ggplot(bladdercancer) +
geom_bar(aes(x = number, fill = tumorsize), position = "dodge")