admin/brew
1
0
Fork 0
Code Issues Packages Projects Releases 6 Wiki Activity

utils/github: use Pulls API to check open pull requests on CI

Browse Source 
We currently use the search API to check for duplicate pull requests,
but this is not very reliable. Our `autobump.yml` workflow routinely
opens duplicate pull requests [1] because the search API often returns
incorrect results.

We can make this more reliable by using the Pulls API instead.
Unfortunately, querying the Pulls API is very slow (~10s vs less than a
second for the search API), so let's limit its usage to calls made
inside CI, which should help @BrewTestBot avoid opening duplicate PRs.
(Most recent dupes were authored by @BrewTestBot.)

[1] https://github.com/Homebrew/homebrew-core/pulls?q=is%3Apr+author%3ABrewTestBot+is%3Aunmerged+in%3Acomments+Duplicate
This commit is contained in:
Carlo Cabrera 2023-05-10 13:08:32 +08:00
parent b3f7c6f7a5
commit 0376c6c5c1
No known key found for this signature in database
GPG Key ID: C74D447FC549A1D0
1 changed files with 44 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

52
Library/Homebrew/utils/github.rb
View File

@ -506,14 +506,16 @@ module GitHub
nil
end
def self.pull_request_title_regex(name, version = nil)
return /(^|\s)#{Regexp.quote(name)}(:|,|\s|$)/i.freeze if version.blank?
/(^|\s)#{Regexp.quote(name)}(:|,|\s)(.*\s)?#{Regexp.quote(version)}(:|,|\s|$)/i.freeze
end
def self.fetch_pull_requests(name, tap_remote_repo, state: nil, version: nil)
if version.present?
query = "#{name} #{version} is:pr"
regex = /(^|\s)#{Regexp.quote(name)}(:|,|\s)(.*\s)?#{Regexp.quote(version)}(:|,|\s|$)/i
else
query = "#{name} is:pr"
regex = /(^|\s)#{Regexp.quote(name)}(:|,|\s|$)/i
end
regex = pull_request_title_regex(name, version)
query = "is:pr #{name} #{version}".strip
issues_for_formula(query, tap_remote_repo: tap_remote_repo, state: state).select do |pr|
pr["html_url"].include?("/pull/") && regex.match?(pr["title"])
end
@ -522,8 +524,42 @@ module GitHub
[]
end
# WARNING: This returns results in a slightly different form from `fetch_pull_requests`.
def self.fetch_open_pull_requests(name, tap_remote_repo, version: nil)
return [] if tap_remote_repo.blank?
# Bust the cache every three minutes.
cache_expiry = 3 * 60
cache_epoch = Time.now - (Time.now.to_i % cache_expiry)
cache_key = "#{tap_remote_repo}_#{cache_epoch.to_i}"
@open_pull_requests ||= {}
@open_pull_requests[cache_key] ||= begin
owner, repo = tap_remote_repo.split("/")
endpoint = "repos/#{owner}/#{repo}/pulls"
query_parameters = ["state=open", "direction=desc"]
pull_requests = []
API.paginate_rest("#{API_URL}/#{endpoint}", additional_query_params: query_parameters.join("&")) do |page|
pull_requests.concat(page)
end
pull_requests
end
regex = pull_request_title_regex(name, version)
@open_pull_requests[cache_key].select { |pr| regex.match?(pr["title"]) }
end
def self.check_for_duplicate_pull_requests(name, tap_remote_repo, state:, file:, args:, version: nil)
pull_requests = fetch_pull_requests(name, tap_remote_repo, state: state, version: version).select do |pr|
# `fetch_open_pull_requests` is more reliable but *really* slow, so let's use it only in CI.
pull_requests = if state == "open" && ENV["CI"].present?
fetch_open_pull_requests(name, tap_remote_repo, version: version)
else
fetch_pull_requests(name, tap_remote_repo, state: state, version: version)
end
pull_requests.select! do |pr|
get_pull_request_changed_files(
tap_remote_repo, pr["number"]
).any? { |f| f["filename"] == file }