1
0
mw-lifecycle-analysis/121325_work/commit_attribution_count.R

67 lines
2.6 KiB
R

library(tidyverse)
library(dplyr)
library(lubridate)
c1_event_date <- as.Date("2013-07-01")
c2_event_date <- as.Date("2013-08-28")
c3_event_date <- as.Date("2015-07-02")
relative_week <- function(date, ref_date) {
as.integer(as.numeric(difftime(date, ref_date, units = "days")) %/% 7)
}
contains_http_but_not_url <- function(text) {
# Handle NA values explicitly
ifelse(
is.na(text),
FALSE,
str_detect(text, "http") &
!str_detect(text, "://") |
str_detect(text, "login") |
str_detect(text, "ssl") |
str_detect(text, "tls") |
(str_detect(text, "cert") & !str_detect(text, "certain"))
)
}
#by the time C. Scott Ananian is committing he is employed
core_csv <-"~/121325_work/121225_vd_data/core_2010-01-01_to_2024-12-31.csv"
core_df <- read.csv(core_csv, header = TRUE)
known_affil_emails <- c("krinkle@fastmail.com", "roan.kattouw@gmail.com",
"trevorparscal@gmail.com", "krinklemail@gmail.com", "moriel@gmail.com")
active_names<- c("Timo Tijhof", "Krinkle", "Roan Kattouw", "Catrope",
"Trevor Parscal", "Ed Sanders", "Moriel Schottlender", "Gabriel Wicke", "C. Scott Ananian")
core_df <- core_df |>
mutate(commit_date = ymd_hms(commit_date)) |>
mutate(isAuthorWMF = case_when(
author_name %in% active_names ~ "NAMES",
grepl("@wikimedia\\.org", author_email, ignore.case = TRUE) ~ "TRUE",
grepl("@wikimedia\\.de", author_email, ignore.case = TRUE) ~ "TRUE",
grepl("l10n-bot@translatewiki\\.net", author_email, ignore.case = TRUE) ~ "localization",
grepl("@gerrit\\.wikimedia\\.org", author_email, ignore.case = TRUE) ~ "Gerrit",
TRUE ~ "FALSE"
)) |>
mutate(isVE = case_when(
grepl("VisualEditor", message, ignore.case = TRUE) ~ TRUE,
grepl(" VE ", message, ignore.case = TRUE) ~ TRUE,
TRUE ~ FALSE
)) |>
mutate(code_location = case_when(
grepl("login", diff_info, ignore.case = TRUE) ~ "login_in_location",
grepl("auth", diff_info, ignore.case = TRUE) ~ "auth_in_location",
grepl("security", diff_info, ignore.case = TRUE) ~ "security",
TRUE ~ "Other"
))
c1_relevant <- core_df |>
mutate(week_index = relative_week(commit_date, c1_event_date)) |>
filter(week_index >= -33 & week_index <= 13) |>
filter(author_email != "jenkins-bot@gerrit.wikimedia.org") |>
filter(isVE == TRUE)
total_share <- core_df |>
filter(commit_date >= as.Date('2012-11-11') & commit_date <= as.Date("2015-10-02"))|>
group_by(isAuthorWMF)|>
summarise(count = n(), .groups = 'drop')
#moving 15 from NAMES to FALSE within this result for the 15 commits that C. Scott. Ananian made prior to his employment