diff --git a/.sh_history b/.sh_history index 255c7f4..0d31a35 100644 --- a/.sh_history +++ b/.sh_history @@ -82,3 +82,29 @@ mv core_2010-01-01_to_2024-12-31.csv mediawiki_core.csv ls mv mediawiki_core.csv mediawiki_core_commits.csv ls +cd .. +ls +cd case1 +l;s +ls +mkdir 042125_stale_counts +mv en-testing_0312_extensions_ve_weekly_commit_count_data.csv 042125_stale_counts/ +mv en-testing_0316_extensions_ve_weekly_commit_count_data.csv 042125_stale_counts/ +mv wide-testing_0316_extensions_ve_weekly_commit_count_data.csv 042125_stale_counts/ +mv wide-testing_0312_extensions_ve_weekly_commit_count_data.csv 042125_stale_counts/ +mv event__0312_extensions_ve_weekly_commit_count_data.csv 042125_stale_counts/ +mv event_0312_extensions_ve_weekly_commit_count_data.csv 042125_stale_counts/ +mv event_0316_extensions_ve_weekly_commit_count_data.csv 042125_stale_counts/ +mv event_0316_mediawiki_core_weekly_commit_count_data.csv 042125_stale_counts/ +mv event_0314_mediawiki_core_weekly_commit_count_data.csv 042125_stale_counts/ +mv event_0312_mediawiki_core_weekly_commit_count_data.csv 042125_stale_counts/ +mv wide-testing_0312_mediawiki_core_weekly_commit_count_data.csv 042125_stale_counts/ +mv wide-testing_0316_mediawiki_core_weekly_commit_count_data.csv 042125_stale_counts/ +mv en-testing_0316_mediawiki_core_weekly_commit_count_data.csv 042125_stale_counts/ +mv en-testing_0312_mediawiki_core_weekly_commit_count_data.csv 042125_stale_counts/ +l +ls +mv event_0401_extensions_ve_weekly_commit_count_data.csv 042125_stale_commits/ +mv event_0401_extensions_ve_weekly_commit_count_data.csv 042125_stale_counts/ +mv event_0401_mediawiki_core_weekly_commit_count_data.csv 042125_stale_counts/ +ls diff --git a/0421-ve-commits.png b/0421-ve-commits.png new file mode 100644 index 0000000..4bc55f9 Binary files /dev/null and b/0421-ve-commits.png differ diff --git a/0421-ve-spike-commits.png b/0421-ve-spike-commits.png new file mode 100644 index 0000000..96f0e5e Binary files /dev/null and b/0421-ve-spike-commits.png differ diff --git a/commit_analysis/commit_count_collation.R b/commit_analysis/commit_count_collation.R index 597e775..5139edb 100644 --- a/commit_analysis/commit_count_collation.R +++ b/commit_analysis/commit_count_collation.R @@ -5,7 +5,7 @@ library(tidyr) library(purrr) library(stringr) -https_commit_fp <- "/mmfs1/gscratch/comdata/users/mjilg/mw-repo-lifecycles/case3/mediawiki_core_commits.csv" +https_commit_fp <- "/mmfs1/gscratch/comdata/users/mjilg/mw-repo-lifecycles/case1/extensions_visualeditor_commits.csv" contains_http_but_not_url <- function(text) { if (is.na(text)) { @@ -45,8 +45,8 @@ transform_commit_data <- function(filepath){ # TODO: this is project/event specific - event_date <- as.Date("2015-07-02") - #event_date <- as.Date("2013-07-01") + #event_date <- as.Date("2015-07-02") + event_date <- as.Date("2013-07-01") #event_date <- as.Date("2013-04-25") #event_date <- as.Date("2012-12-11") @@ -69,7 +69,7 @@ transform_commit_data <- function(filepath){ #drop out data from ''before'' the release process df <- df |> - filter(commit_date >= as.Date("2015-04-01")) + filter(commit_date >= as.Date("2012-11-11")) #we are looking at weekly data, 6m before and 6m after #start_date <- event_date %m-% months(6) @@ -96,7 +96,7 @@ transform_commit_data <- function(filepath){ # list all author_emails with >5 commits # for big df: if author not in the list, 'new' author old_author_list <- df |> - filter(commit_date < as.Date("2015-06-01"))|> + filter(commit_date < as.Date("2013-06-06"))|> group_by(author_email) |> summarise(commit_count = n()) |> filter(commit_count > 5) |> @@ -192,7 +192,7 @@ transform_commit_data <- function(filepath){ # ) |> weekly_commits <- weekly_commits |> - filter(relative_week >= (-14) & relative_week <= 52 ) + filter(relative_week >= (-34) & relative_week <= 13 ) #gracefully exit return(weekly_commits) @@ -364,7 +364,7 @@ transform_relevant_commit_data <- function(filepath){ } transformed <- transform_commit_data(https_commit_fp) -output_filepath <-"/mmfs1/gscratch/comdata/users/mjilg/mw-repo-lifecycles/case3/event_0415_mediawiki_core_weekly_commit_count_data.csv" +output_filepath <-"/mmfs1/gscratch/comdata/users/mjilg/mw-repo-lifecycles/case1/event_0421_extensions_ve_weekly_count.csv" write.csv(transformed, output_filepath, row.names = FALSE) diff --git a/commit_analysis/commit_plotting.R b/commit_analysis/commit_plotting.R index 38eecb4..d796201 100644 --- a/commit_analysis/commit_plotting.R +++ b/commit_analysis/commit_plotting.R @@ -1,20 +1,74 @@ library(tidyverse) -count_data_fp <-"/mmfs1/gscratch/comdata/users/mjilg/mw-repo-lifecycles/case3/event_0415_mediawiki_core_weekly_commit_count_data.csv" +count_data_fp <-"/mmfs1/gscratch/comdata/users/mjilg/mw-repo-lifecycles/case1/event_0421_extensions_ve_weekly_count.csv" input_df <- read.csv(count_data_fp, header = TRUE) input_df$nonbot_commit_count <- input_df$commit_count - input_df$bot_commit_count +input_df$unaff_commit_count <- input_df$nonbot_commit_count - input_df$wikimedia_commit_count library(scales) library(ggplot2) long_df <- input_df |> - tidyr::pivot_longer(cols = c(nonbot_commit_count, unaff_new_commit_count, wmf_new_commit_count), + tidyr::pivot_longer(cols = c(nonbot_commit_count, unaff_commit_count, wikimedia_commit_count), names_to = "commit_type", values_to = "lengthened_commit_count") affiliationColors <- setNames( c('black','#5da2d8', '#c7756a') - ,c("nonbot_commit_count","unaff_new_commit_count", "wmf_new_commit_count")) + ,c("nonbot_commit_count","unaff_commit_count", "wikimedia_commit_count")) + +commit_authors <- long_df |> + ggplot(aes(x=relative_week, + y=lengthened_commit_count, + color=factor(commit_type))) + + geom_point() + + geom_line() + + labs(x = "Relative Week", y = "Commits", color="Commit Type") + + scale_color_manual(values = affiliationColors, + labels = c("nonbot_commit_count" = "Total Nonbot Commits", + "unaff_commit_count" = "Unaffiliated Commits", + "wikimedia_commit_count" = "WMF Commits")) + + ggtitle("Total VE Commits Around Opt-out Deployment By Affiliation") + + theme_bw() + + theme(legend.position = "top") +commit_authors + +ggsave(filename = "0421-ve-commits.png", plot = commit_authors, width = 12, height = 9, dpi = 800) + + +# new affiliation things +# relative week >= -4 +# new unaffiliated +# returning unaffiliated +input_df$returning_unaff_commit_count = input_df$unaff_commit_count - input_df$unaff_new_commit_count + + +new_authors_long_df <- input_df |> + filter(relative_week >= -4) |> + tidyr::pivot_longer(cols = c(unaff_new_commit_count, returning_unaff_commit_count), + names_to = "commit_seniority", + values_to = "lengthened_commit_count") + +new_unaff_authors <- new_authors_long_df |> + ggplot(aes(x=relative_week, + y=lengthened_commit_count, + linetype=commit_seniority)) + + geom_point(color = '#5da2d8') + + geom_line(color='#5da2d8') + + labs(x = "Relative Week", y = "Commits", linetype="Commit Seniority ('New' contributors <= 5 commits before 06-06-2013)") + + scale_linetype_manual( + values = c("returning_unaff_commit_count" = "solid", + "unaff_new_commit_count" = "dotted"), + labels = c("returning_unaff_commit_count" = "Returning Contributors", + "unaff_new_commit_count" = "New Contributors") + ) + + ggtitle("Unaffiliated VE Commits Surrounding Opt-out Deployment") + + theme_bw() + + theme(legend.position = "top") + +new_unaff_authors + +ggsave(filename = "0421-ve-spike-commits.png", plot = new_unaff_authors, width = 12, height = 9, dpi = 800) new_authors <- long_df |> ggplot(aes(x=relative_week, @@ -28,8 +82,9 @@ new_authors <- long_df |> "unaff_new_commit_count" = "New Unaffiliated Commits", "wmf_new_commit_count" = "New WMF Commits")) + ggtitle("Total MW-core Commits Around HTTP-deprecation ('New' contributors <= 5 commits 04-01-2015 and 06-01-2015)") + - theme_bw() + - theme(legend.position = "top") + theme_bw() + #+ + #theme(legend.position = "top") new_authors ggsave(filename = "0403-https-core-event-new-commits.png", plot = new_authors, width = 12, height = 9, dpi = 800)