Match Apache Arrow Jira user accounts with GitHub user accounts

clean_github_jira_ids.R

# run this script second

library(dplyr)

df <- read.csv("dirty.csv")

agg <- df %>%
  group_by(jira, github) %>%
  summarise(n = n(), .groups = "keep") %>%
  ungroup() %>%
  arrange(desc(n))

jira_users <- unique(agg$jira)

final <- data.frame(jira = jira_users, github = character(length(jira_users)))

for (jira_user in jira_users) {
  temp <- agg %>% filter(jira == jira_user)
  if (nrow(temp) > 1) {
    print(temp)
    i <- as.integer(readline(prompt="Which row matches? "))
  } else {
    i <- 1
  }
  final[final$jira == jira_user, "github"] <- temp[i, "github", drop = TRUE]
}

final %>% write.csv("clean.csv")

get_github_jira_ids.py

# run this script first

from jira import JIRA
from github import Github
import re
import pandas as pd

jira = JIRA('https://issues.apache.org/jira')
github = Github("ENTER_YOUR_GITHUB_PAT_HERE")
repo = github.get_repo("apache/arrow")

jira_issues = []
i = 0
chunk_size = 200
while True:
    chunk = jira.search_issues(f'project = ARROW AND status = Resolved ORDER BY key', startAt=i, maxResults=chunk_size, fields='assignee,comment')
    i += chunk_size
    jira_issues += chunk.iterable
    if i >= chunk.total:
        break

jira_users = []
gh_users = []

for jira_issue in jira_issues:
    print(jira_issue.key)
    if jira_issue.fields.assignee is None:
        continue
    n = len(jira_issue.fields.comment.comments)
    for j in reversed(range(0, n)):
        t = jira_issue.fields.comment.comments[j].body
        s = re.search(r"^Issue resolved by pull request ([0-9]{1,5})\n\[https://github.com/apache/arrow/pull/\1]$", t)
        if s:
            gh_issue_id = int(s.group(1))
            gh_issue = repo.get_issue(number=gh_issue_id)
            print('\t' + jira_issue.fields.assignee.key)
            jira_users.append(jira_issue.fields.assignee.key)
            print('\t' + gh_issue.user.login)
            gh_users.append(gh_issue.user.login)
            break

df = pd.DataFrame({'jira': jira_users, 'github': gh_users})
df.to_csv('dirty.csv')

This was created to help with apache/arrow#14510.

Replace assignee.key with assignee.name if needed.