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(
    jsonfixed = gsub("(?<=\\{|\\[|, )'(\\w+?)'(?=\\s*:)", '"\\1"', selected_gerrit_results, perl = TRUE),
    jsonfixed = gsub(":(\\s*)'(.*?)'(?=[}\\],])", ':\\1"\\2"', jsonfixed, perl = TRUE),
    jsonfixed = gsub("(?<=[:\\[,\\{])\\s*'([^']*)'\\s*(?=[,\\}\\]])", '"\\1"', jsonfixed, perl = TRUE),
    expandedjsonlist = lapply(jsonfixed, fromJSON)
    )

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