library(tidyverse)
library(jsonlite)

main_csv <-"~/p2/071425_master_discussion_data.csv"
main_df <- read.csv(main_csv, header = TRUE) 

gerrit_csv <-"~/p2/080425_gerrit_filled_df.csv"
gerrit_df <- read.csv(gerrit_csv, header = TRUE) 

library(dplyr)

gerrit_extracted_df <- gerrit_df |>
  mutate(
    gerrit_status = str_match(selected_gerrit_results, "'status':\\s*'([^']*)',\\s*'reviewer")[,2],
    owner_email = str_match(selected_gerrit_results, "'owner_email':\\s*'([^']*)'")[,2],
    written_url_in_message = str_match(selected_gerrit_results, "'written_url_in_message':\\s*'([^']*)'")[,2],
    code_insertions = as.integer(str_match(selected_gerrit_results, "'code_insertions':\\s*(\\d+)")[,2]),
    code_deletions = as.integer(str_match(selected_gerrit_results, "'code_deletions':\\s*(\\d+)")[,2]),
    reviewer_count = as.integer(str_match(selected_gerrit_results, "'reviewer_count':\\s*(\\d+)")[,2])
  )

# TODO select
# gerrit status: categorical
# owner_wikimedia: BOOL
# LOCa: #
# LOCd: #
# count_reviewers: # 
# url: url