1
0
mw-lifecycle-analysis/commit_analysis/plotting/ww-plot-script.R
2025-05-01 20:51:23 -07:00

84 lines
4.2 KiB
R

library(tidyverse)
c1_count <-"/mmfs1/gscratch/comdata/users/mjilg/mw-repo-lifecycles/case1/event_0421_extensions_ve_weekly_count.csv"
c1_input_df <- read.csv(c1_count , header = TRUE)
c2_count <-"/mmfs1/gscratch/comdata/users/mjilg/mw-repo-lifecycles/case2/event_0430_mediawiki_core_weekly_count.csv"
c2_input_df <- read.csv(c2_count , header = TRUE)
c3_count <-"/mmfs1/gscratch/comdata/users/mjilg/mw-repo-lifecycles/case3/event_0430_mediawiki_core_weekly_count.csv"
c3_input_df <- read.csv(c3_count , header = TRUE)
# Add a column to each dataframe to label them
c1_input_df <- c1_input_df %>% mutate(source = "c1")
c2_input_df <- c2_input_df %>% mutate(source = "c2")
c3_input_df <- c3_input_df %>% mutate(source = "c3")
# Combine the dataframes into one
combined_df <- bind_rows(c1_input_df, c2_input_df, c3_input_df)
combined_df$nonbot_commit_count <- combined_df$commit_count - combined_df$bot_commit_count
combined_df$unaff_commit_count <- combined_df$nonbot_commit_count - combined_df$wikimedia_commit_count
library(scales)
library(ggplot2)
long_df <- combined_df |>
tidyr::pivot_longer(cols = c(unaff_commit_count, wikimedia_commit_count),
names_to = "commit_type",
values_to = "lengthened_commit_count")
affiliationColors <-
setNames( c('#5da2d8', '#c7756a')
,c("unaff_commit_count", "wikimedia_commit_count"))
commit_authors <- long_df |>
ggplot(aes(x=relative_week,
y=lengthened_commit_count,
fill=factor(commit_type))) +
geom_col(position='dodge') +
labs(x = "Relative Week", y = "Commits", fill="Commit Author") +
geom_vline(data = long_df |> filter(source == "c1"),
aes(xintercept = -29),
linetype = "dotted", color = "black", linewidth = 1) +
geom_vline(data = long_df |> filter(source == "c1"),
aes(xintercept = -9),
linetype = "dotted", color = "black", linewidth = 1) +
geom_vline(data = long_df |> filter(source == "c2"),
aes(xintercept = -99),
linetype = "dotted", color = "black", linewidth = 1) +
geom_vline(data = long_df |> filter(source == "c3"),
aes(xintercept = -97),
linetype = "dotted", color = "black", linewidth = 1) +
geom_text(data = data.frame(source = "c1", relative_week = -40, lengthened_commit_count = 50),
aes(x = relative_week, y = lengthened_commit_count, label = "Opt-In Testing Deployment"),
inherit.aes = FALSE, color = "black", size = 4) +
geom_vline(xintercept = 0, linetype = "dashed", color = "black", linewidth = 1) + # Add vertical line at week 0
geom_text(data = data.frame(source = "c1", relative_week = 7, lengthened_commit_count = 50),
aes(x = relative_week, y = lengthened_commit_count, label = "Wide Deployment"),
inherit.aes = FALSE, color = "black", size = 4) +
scale_fill_manual(values = affiliationColors,
labels = c("unaff_commit_count" = "Unaffiliated",
"wikimedia_commit_count" = "WMF-affiliated")) +
ggtitle("Feature Commits During Deployment Process") +
theme_bw()+
theme(
legend.position = "top",
plot.title = element_text(size = 18, face = "bold"), # Increase title font size
axis.title.x = element_text(size = 16), # Increase x-axis title font size
axis.title.y = element_text(size = 16), # Increase y-axis title font size
axis.text.x = element_text(size = 16), # Increase x-axis text font size
axis.text.y = element_text(size = 16), # Increase y-axis text font size
legend.text = element_text(size = 16), # Increase legend text font size
legend.title = element_text(size = 16),
strip.text = element_text(size = 14)# Increase legend title font size
) +
facet_wrap(~source, nrow = 3, labeller = labeller(source = c(
"c1" = "VisualEditor (commits to extensions/visualeditor)",
"c2" = "HTTPS-as-default (relevant commits to mediawiki/core)",
"c3" = "HTTP-deprecation (relevant commits to mediawiki/core)"
)))
commit_authors
ggsave(filename = "ww-0501-commits-faceted.png", plot = commit_authors, width = 15, height = 9, dpi = 800)