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)
}

core_csv <-"~/121325_work/121225_vd_data/core_2010-01-01_to_2024-12-31.csv"
core_df <- read.csv(core_csv, header = TRUE) 
core_df <- core_df |>
  mutate(commit_date = ymd_hms(commit_date)) |>
  mutate(isAuthorWMF = case_when(
    grepl("krinkle@fastmail\\.com", author_email, ignore.case = TRUE) ~ "TRUE",
    grepl("@wikimedia\\.org", author_email, ignore.case = TRUE) ~ "TRUE",
    grepl("@wikimedia\\.de", author_email, ignore.case = TRUE) ~ "TRUE",
    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
  ))

c1_core_weekly <- core_df |>
  mutate(week_index = relative_week(commit_date, c1_event_date)) |>
  group_by(week_index, isVE)|>
  summarise(count = n(), .groups = 'drop')|>
  filter(week_index >= -33 & week_index <= 13) |>
  mutate(source = 'c1')

true_ <- c1_core_weekly |>
  filter(isVE == TRUE)