library(tidyverse)

main_csv <- "~/analysis_data/092925_unified_phab.csv"
main_df <- read.csv(main_csv, header = TRUE) 

date1 <-1380499199 # 9-29-2013
date2 <- 1385510399 # 11-26-2013
date3 <-1443657599 # 09-30-2015

new_outcome <- main_df |>
  mutate(
    resolution_outcome = case_when(
      comment_type == "task_description" & source == "c1" & date_closed < date1 ~ TRUE,
      comment_type == "task_description" & source == "c1" & date_closed >= date1 ~ FALSE,
      comment_type == "task_description" & source == "c2" & date_closed < date2 ~ TRUE,
      comment_type == "task_description" & source == "c2" & date_closed >= date2 ~ FALSE,
      comment_type == "task_description" & source == "c3" & date_closed < date3 ~ TRUE,
      comment_type == "task_description" & source == "c3" & date_closed >= date3 ~ FALSE,
      comment_type == "task_description" & is.na(date_closed) ~ FALSE,
      TRUE ~ NA
    )
  ) |>
  select(-closed_relevance)
anyNA(new_outcome$resolution_outcome[new_outcome$comment_type == "task_description"])

write.csv(new_outcome, "100325_unified_phab.csv", row.names = FALSE)