Create tool for getting DCO sign off emails

This tool digs through commit messages and parses out names and email
addresses from the `Signed-off-by:` tags, collects all unique email
addresses and outputs a CSV of name/email pairs.

Signed-off-by: Andrew Ross <[email protected]>

.rubocop_todo.yml

@@ -1,6 +1,6 @@
 # This configuration was generated by
 # `rubocop --auto-gen-config`
-# on 2022-12-01 14:05:15 UTC using RuboCop version 1.36.0.
+# on 2022-12-06 18:50:08 UTC using RuboCop version 1.36.0.
 # The point is for the user to remove these configuration records
 # one by one as the offenses are removed from the code base.
 # Note that changes in the inspected code, or installation of new
@@ -113,37 +113,23 @@ RSpec/EmptyExampleGroup:
 RSpec/ExampleLength:
   Max: 6
 
-# Offense count: 14
+# Offense count: 17
 # Configuration parameters: Include, CustomTransform, IgnoreMethods, SpecSuffixOnly.
 # Include: **/*_spec*rb*, **/spec/**/*
 RSpec/FilePath:
-  Exclude:
-    - 'spec/github/contributor_spec.rb'
-    - 'spec/github/contributors_spec.rb'
-    - 'spec/github/data_spec.rb'
-    - 'spec/github/issue_spec.rb'
-    - 'spec/github/issues_spec.rb'
-    - 'spec/github/organization_spec.rb'
-    - 'spec/github/progress_spec.rb'
-    - 'spec/github/pull_request_spec.rb'
-    - 'spec/github/pull_requests_spec.rb'
-    - 'spec/github/rate_limited_spec.rb'
-    - 'spec/github/repos_spec.rb'
-    - 'spec/github/searchable_spec.rb'
-    - 'spec/github/user_spec.rb'
-    - 'spec/github/users_spec.rb'
+  Enabled: false
 
 # Offense count: 1
 # Configuration parameters: AssignmentOnly.
 RSpec/InstanceVariable:
   Exclude:
     - 'spec/github/progress_spec.rb'
 
-# Offense count: 7
+# Offense count: 12
 RSpec/MultipleExpectations:
   Max: 5
 
-# Offense count: 28
+# Offense count: 31
 # Configuration parameters: IgnoreSharedExamples.
 RSpec/NamedSubject:
   Exclude:
@@ -152,7 +138,7 @@ RSpec/NamedSubject:
     - 'spec/github/progress_spec.rb'
     - 'spec/github/pull_requests_spec.rb'
 
-# Offense count: 8
+# Offense count: 10
 # Configuration parameters: AllowedGroups.
 RSpec/NestedGroups:
   Max: 5

README.md

@@ -16,6 +16,7 @@
     - [Pull Request Stats](#pull-request-stats)
     - [Issues](#issues)
     - [Member Bios](#member-bios)
+    - [DCO Signers](#dco-signers)
 - [Contributing](#contributing)
 - [Code of Conduct](#code-of-conduct)
 - [Security](#security)
@@ -288,6 +289,14 @@ Shows users in [data/users/members.txt](data/users/members.txt) that do not have
 ./bin/project members check
 ```
 
+#### DCO Signers
+
+Shows name and email address from all contributors that have signed a developer certificate of origin on any commit.
+
+```
+./bin/project dco-signers --from=2022-01-01 --to=2022-01-31 --org=opensearch-project --repo=OpenSearch
+```
+
 ## Contributing
 
 See [how to contribute to this project](CONTRIBUTING.md).

bin/commands/contributors.rb

@@ -32,4 +32,15 @@
       end
     end
   end
+
+  g.desc 'Create a list of all DCO signers'
+  g.command 'dco-signers' do |c|
+    c.action do |_global_options, options, _args|
+      org = GitHub::Organization.new(options)
+      signers = org.commits(options).dco_signers
+      signers.sort_for_display.each do |signer|
+        puts signer.to_s
+      end
+    end
+  end
 end

lib/github/commit.rb

@@ -0,0 +1,13 @@
+# frozen_string_literal: true
+
+module GitHub
+  class Commit < Item
+    # Creates an array of Signers from all 'Signed-off-by' tags included in the
+    # commit message
+    def dco_signers
+      commit.message.scan(/Signed-off-by: (.+) <([A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Za-z]+)>/).map do |signer|
+        Signer.new(signer[0], signer[1])
+      end
+    end
+  end
+end

lib/github/commits.rb

@@ -0,0 +1,31 @@
+# frozen_string_literal: true
+
+module GitHub
+  class Commits < Items
+    def initialize(arr_or_options)
+      super arr_or_options, GitHub::Commit
+    end
+
+    # Gets all unique DCO signers (by email address) from all commits
+    def dco_signers
+      Signers.new(each.map(&:dco_signers).flatten)
+    end
+
+    def page(options)
+      data = $github.search_commits(query(options), per_page: 1000).items
+      raise 'There are 1000+ commits returned from a single query, reduce --page.' if data.size >= 1000
+
+      data.reject do |commit|
+        commit.commit.author.email.include?('[bot]')
+      end
+    end
+
+    def query(options = {})
+      GitHub::Searchables.new(options).to_a.concat(
+        [
+          "committer-date:#{options[:from]}..#{options[:to]}"
+        ]
+      ).compact.join(' ')
+    end
+  end
+end

lib/github/organization.rb

@@ -28,6 +28,10 @@ def pull_requests(options = {})
       @pull_requests ||= GitHub::PullRequests.new({ org: name, status: :merged }.merge(options))
     end
 
+    def commits(options = {})
+      @commits ||= GitHub::Commits.new({ org: name }.merge(options))
+    end
+
     def issues(options = {})
       @issues ||= GitHub::Issues.new({ org: name }.merge(options))
     end

lib/github/signer.rb

@@ -0,0 +1,16 @@
+# frozen_string_literal: true
+
+module GitHub
+  class Signer
+    attr_reader :email, :name
+
+    def initialize(name, email)
+      @name = name
+      @email = email
+    end
+
+    def to_s
+      "#{name},#{email}"
+    end
+  end
+end

lib/github/signers.rb

@@ -0,0 +1,32 @@
+# frozen_string_literal: true
+
+module GitHub
+  class Signers < Array
+    def initialize(arr)
+      # De-dupe by email address, choosing the "best" name
+      by_email = {}
+      arr.each do |signer|
+        by_email[signer.email] = best_signer(by_email[signer.email], signer)
+      end
+      super by_email.values
+    end
+
+    # Sort all "noreply" email addresses to the bottom (for manual curation), then sort by name
+    def sort_for_display
+      Signers.new(sort_by { |signer| [signer.email.include?('noreply') ? 1 : 0, signer.name.downcase] })
+    end
+
+    private
+
+    def best_signer(left, right)
+      # The "best" name is defined by the name with the most words. For example,
+      # if both "dblock" and "Daniel (dB.) Doubrovkine" are encountered, then
+      # "Daniel (dB.) Doubrovkine" will be chosen.
+      if left.nil? || right.name.split.length > left.name.split.length
+        right
+      else
+        left
+      end
+    end
+  end
+end

lib/tools.rb

@@ -19,9 +19,13 @@
 require_relative 'github/repos'
 require_relative 'github/pull_requests'
 require_relative 'github/pull_request'
+require_relative 'github/commits'
+require_relative 'github/commit'
 require_relative 'github/contributors'
 require_relative 'github/contributor'
 require_relative 'github/maintainers'
+require_relative 'github/signers'
+require_relative 'github/signer'
 require_relative 'github/users'
 require_relative 'github/user'
 require_relative 'github/issues'

spec/github/commit_spec.rb

@@ -0,0 +1,51 @@
+# frozen_string_literal: true
+
+describe GitHub::Commit do
+  subject(:commit) do
+    message = %{Bump opencensus-contrib-http-util from 0.18.0 to 0.31.1 in /plugins/repository-gcs (#3633)
+
+            * Bump opencensus-contrib-http-util in /plugins/repository-gcs
+
+            Bumps [opencensus-contrib-http-util](https://github.com/census-instrumentation/opencensus-java) from 0.18.0 to 0.31.1.
+            - [Release notes](https://github.com/census-instrumentation/opencensus-java/releases)
+            - [Changelog](https://github.com/census-instrumentation/opencensus-java/blob/master/CHANGELOG.md)
+            - [Commits](census-instrumentation/[email protected])
+
+            ---
+            updated-dependencies:
+            - dependency-name: io.opencensus:opencensus-contrib-http-util
+              dependency-type: direct:production
+              update-type: version-update:semver-minor
+            ...
+
+            Signed-off-by: dependabot[bot] <[email protected]>
+
+            * Updating SHAs
+
+            Signed-off-by: dependabot[bot] <[email protected]>
+
+            * Adding missing classes
+
+            Signed-off-by: Vacha Shah <[email protected]>
+
+            * changelog change
+
+            Signed-off-by: Poojita Raj <[email protected]>
+
+            Signed-off-by: dependabot[bot] <[email protected]>
+            Signed-off-by: Vacha Shah <[email protected]>
+            Signed-off-by: Poojita Raj <[email protected]>
+            Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
+            Co-authored-by: dependabot[bot] <dependabot[bot]@users.noreply.github.com>
+            Co-authored-by: Vacha Shah <[email protected]>
+            Co-authored-by: Poojita Raj <[email protected]>
+        }
+    resource = Sawyer::Resource.new(Sawyer::Agent.new('fake'), { commit: { message: message } })
+    described_class.new(resource)
+  end
+
+  it 'parses signers from a commit message' do
+    expect(commit.dco_signers.count).to eq 7
+    expect(commit.dco_signers.map(&:name)).to eq ['dependabot[bot]', 'dependabot[bot]', 'Vacha Shah', 'Poojita Raj', 'dependabot[bot]', 'Vacha Shah', 'Poojita Raj']
+  end
+end

spec/github/commits_spec.rb

@@ -0,0 +1,25 @@
+# frozen_string_literal: true
+
+describe GitHub::Commits do
+  context 'with contributors' do
+    context 'with org' do
+      context 'with january 2022' do
+        context 'with OpenSearch commits', vcr: { cassette_name: 'search/opensearch-project/commits_2022-01-01_2022-01-31' } do
+          subject(:commits) do
+            described_class.new(org: 'opensearch-project', repo: 'OpenSearch', from: Date.new(2022, 1, 1), to: Date.new(2022, 1, 31), page: 7)
+          end
+
+          it 'fetches commits between two dates' do
+            expect(commits.count).to eq 62
+            expect(commits.first['sha']).to eq 'db23f72a2a5da1f21d674bde3a9d1cbe4fb74b19'
+          end
+
+          it 'collects DCO signers from commits' do
+            expect(commits.dco_signers.count).to eq 25
+            expect(commits.dco_signers.first.name).to eq 'Tianli Feng'
+          end
+        end
+      end
+    end
+  end
+end

spec/github/signers_spec.rb

@@ -0,0 +1,33 @@
+# frozen_string_literal: true
+
+describe GitHub::Signers do
+  context 'with duplicate emails' do
+    subject(:signers) do
+      described_class.new([
+                            GitHub::Signer.new('dev', '[email protected]'),
+                            GitHub::Signer.new('Dev Eloper', '[email protected]'),
+                            GitHub::Signer.new('Dev Eloper, Esq.', '[email protected]')
+                          ])
+    end
+
+    it 'chooses the best name' do
+      expect(signers.size).to eq(1)
+      expect(signers.first.name).to eq('Dev Eloper, Esq.')
+    end
+  end
+
+  context 'with noreply email addresses' do
+    subject(:signers) do
+      described_class.new([
+                            GitHub::Signer.new('Mis Configurer', '[email protected]'),
+                            GitHub::Signer.new('dev', '[email protected]')
+                          ])
+    end
+
+    it 'sorts noreply email addresses to the end' do
+      expect(signers.size).to eq(2)
+      expect(signers.sort_for_display.first.email).to eq('[email protected]')
+      expect(signers.sort_for_display.last.email).to eq('[email protected]')
+    end
+  end
+end