newsman 1.2.2 → 1.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: '0166813d55623364efd38ff1f3bc83ff3ae76ee9d793b1b5ac52924477122135'
-  data.tar.gz: 48d3c00284b0973851b95c2b144fe50972f801deeb826b400ab31e87102b7cfc
+  metadata.gz: 7d8f8d600c711d2570fdff045de97d9162f5173e733b164c98ec667ad01e1158
+  data.tar.gz: 427198b752758475747275768a3db990f563d90d0853653ba81b1b03a1664257
 SHA512:
-  metadata.gz: 193cf28ea41b994cab278212cc255dcf0d1f7b78570e631f68e17003d489d505fd0e12a8b3244d576fa1e0826d992ead7af9c1febefd80074dc8260174294dae
-  data.tar.gz: 4e3a9c5920d471c3c5a1dcabe16692a6154ea50d44adaf775d263ce04186e814068391555d3e90682b043dbb2b3c4fb0372da2f8fa7e91abf331a594bfefe4e7
+  metadata.gz: 4f1855fcef29439b336ef10eef275a927a5a4c47574c67275fd35e10bbe9a64b731081bcc80c0525b17d1c1b46c68759c14f05564de251f9e097b3708edebfd5
+  data.tar.gz: 3fde1f89ade004ccf2e3a77a86e94489590a933d3cf36e824a43d7ba314ddef53ff70235bee91e1a20edb3950804b1604caa72317aac099d95081b1941f1f7f1

data/lib/newsman/assistant.rb

@@ -45,7 +45,8 @@ class Assistant
     example = "repository-name:\n
       - To publish ABC package draft [#27]\n
       - To review first draft of the report [#56]\n
-      - To implement optimization for the class X [#125]"
+      - To implement optimization for the class X [#125]\n
+      - Updated dependencies: rubocop, ruby [#197, #199, #204]"
     prompt = <<~PROMPT
       Please compile a summary of the plans for the next week using the GitHub Issues.
       Each issue should be summarized in a single sentence.
@@ -53,6 +54,8 @@ class Assistant
       Pay attention, that you didn't loose any issue.
       Ensure that each sentence includes the corresponding issue number as an integer value. If an issue doesn't mention an issue number, just print [#chore].
       If several issues have the same number, combine them into a single sentence.
+      Dependency version-bump issues (e.g. titles like "chore(deps): update dependency X to vY" or similar renovate/dependabot style bumps) must all be grouped into a single combined bullet listing the dependency names and all associated issue numbers, instead of one bullet per issue.
+      Other substantive issues should keep their own individual bullet as usual.
       Please strictly adhere to the example template provided: "#{example}".
       List of GitHub Issues in JSON format: ```json #{issues}```.
     PROMPT
@@ -65,13 +68,16 @@ class Assistant
     example = "repository-name:\n
       - Added 100 new files to the Dataset [#168]\n
       - Fixed the deployment of XYZ [#169]\n
-      - Refined the requirements [#177]\n"
+      - Refined the requirements [#177]\n
+      - Updated dependencies: rubocop, ruby, nokogiri [#197, #199, #200, #202, #203, #204]\n"
     prompt = <<~PROMPT
       Please compile a summary of the work completed in the following Pull Requests (PRs).
       Each PR should be summarized in a single sentence, focusing on the PR title rather than the implementation details.
       Ensure no PR is omitted.
       Each sentence must include the corresponding issue number as an integer value. If a PR does not mention an issue number, use [#chore].
       If several PRs have the same number, combine them into a single sentence.
+      Dependency version-bump PRs (e.g. titles like "chore(deps): update dependency X to vY" or similar renovate/dependabot style bumps) must all be grouped into a single combined bullet listing the dependency names and all associated PR numbers, instead of one bullet per PR.
+      Other substantive PRs should keep their own individual bullet as usual.
       Combine the information from each PR into a concise and fluent sentence.
       Follow the provided example template strictly: "#{example}".
       List of Pull Requests in JSON format: ```json #{prs}```.

data/lib/newsman/docx_output.rb

@@ -30,7 +30,7 @@ class Docxout
   # Content types part, required by every OOXML package.
   CONTENT_TYPES = <<~XML
     <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
-    <Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types"><Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/><Default Extension="xml" ContentType="application/xml"/><Override PartName="/word/document.xml" ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml"/></Types>
+    <Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types"><Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/><Default Extension="xml" ContentType="application/xml"/><Override PartName="/word/document.xml" ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml"/><Override PartName="/word/numbering.xml" ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.numbering+xml"/></Types>
   XML
 
   # Package-level relationships, pointing at the main document part.
@@ -39,6 +39,27 @@ class Docxout
     <Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="word/document.xml"/></Relationships>
   XML
 
+  # Document-level relationships, pointing at the numbering part used by lists.
+  DOCUMENT_RELS = <<~XML
+    <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+    <Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/numbering" Target="numbering.xml"/></Relationships>
+  XML
+
+  # Numbering definitions: numId 1 is a bulleted list, numId 2 is a decimal (numbered) list.
+  NUMBERING = <<~XML
+    <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+    <w:numbering xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:abstractNum w:abstractNumId="0"><w:lvl w:ilvl="0"><w:numFmt w:val="bullet"/><w:lvlText w:val="&#8226;"/><w:pPr><w:ind w:left="720" w:hanging="360"/></w:pPr></w:lvl></w:abstractNum><w:abstractNum w:abstractNumId="1"><w:lvl w:ilvl="0"><w:numFmt w:val="decimal"/><w:lvlText w:val="%1."/><w:pPr><w:ind w:left="720" w:hanging="360"/></w:pPr></w:lvl></w:abstractNum><w:num w:numId="1"><w:abstractNumId w:val="0"/></w:num><w:num w:numId="2"><w:abstractNumId w:val="1"/></w:num></w:numbering>
+  XML
+
+  # Matches markdown headers, e.g. "## Title" (levels 1-6).
+  HEADING = /^\s{0,3}(\#{1,6})\s+(.+?)\s*$/
+  # Matches markdown bullet list items, e.g. "- item" or "* item".
+  BULLET = /^\s{0,3}[-*]\s+(.+?)\s*$/
+  # Matches markdown numbered list items, e.g. "1. item".
+  NUMBERED = /^\s{0,3}\d+\.\s+(.+?)\s*$/
+  # Matches markdown bold markers, e.g. "**bold**" or "__bold__".
+  BOLD = /(\*\*|__)(.+?)\1/
+
   FONT = 'Times New Roman'
   # Word expresses font size in half-points, so 12pt is 24.
   SIZE = '24'
@@ -50,11 +71,7 @@ class Docxout
   def print(report, reporter, model)
     puts "Create a docx file in a directory #{@root}"
     path = File.join(@root, filename(reporter, model))
-    Zip::File.open(path, create: true) do |zip|
-      zip.get_output_stream('[Content_Types].xml') { |f| f.write(CONTENT_TYPES) }
-      zip.get_output_stream('_rels/.rels') { |f| f.write(RELS) }
-      zip.get_output_stream('word/document.xml') { |f| f.write(document_xml(report)) }
-    end
+    write_package(path, report)
     puts "Report was successfully printed to a #{path}"
     File.basename(path)
   end
@@ -67,16 +84,63 @@ class Docxout
 
   private
 
+  def write_package(path, report)
+    Zip::File.open(path, create: true) do |zip|
+      zip.get_output_stream('[Content_Types].xml') { |f| f.write(CONTENT_TYPES) }
+      zip.get_output_stream('_rels/.rels') { |f| f.write(RELS) }
+      zip.get_output_stream('word/_rels/document.xml.rels') { |f| f.write(DOCUMENT_RELS) }
+      zip.get_output_stream('word/numbering.xml') { |f| f.write(NUMBERING) }
+      zip.get_output_stream('word/document.xml') { |f| f.write(document_xml(report)) }
+    end
+  end
+
   def document_xml(report)
-    body = paragraphs(report).map { |lines| paragraph_xml(lines) }.join
+    body = blocks(report).map { |block| block_xml(block) }.join
     <<~XML
       <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
       <w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"><w:body>#{body}<w:sectPr/></w:body></w:document>
     XML
   end
 
-  def paragraphs(report)
-    report.to_s.split(/\n{2,}/).map { |paragraph| paragraph.split("\n") }
+  # Splits the report into blocks: markdown headers and list items each become
+  # their own block, and consecutive plain-text lines are grouped together.
+  def blocks(report)
+    classified = report.to_s.split("\n", -1).map { |line| classify(line) }
+    classified.chunk_while { |prev, curr| prev[:type] == :text && curr[:type] == :text }
+              .reject { |group| group.first[:type] == :blank }
+              .map { |group| merge_text(group) }
+  end
+
+  def merge_text(group)
+    return group.first unless group.first[:type] == :text
+
+    { type: :text, lines: group.map { |block| block[:line] } }
+  end
+
+  def classify(line)
+    return { type: :blank } if line.strip.empty?
+    return { type: :heading, level: ::Regexp.last_match(1).length, text: ::Regexp.last_match(2) } if line =~ HEADING
+    return { type: :bullet, text: ::Regexp.last_match(1) } if line =~ BULLET
+    return { type: :numbered, text: ::Regexp.last_match(1) } if line =~ NUMBERED
+
+    { type: :text, line: line }
+  end
+
+  def block_xml(block)
+    case block[:type]
+    when :heading
+      "<w:p><w:pPr><w:pStyle w:val=\"Heading#{block[:level]}\"/></w:pPr>#{run_xml(block[:text])}</w:p>"
+    when :bullet
+      list_item_xml(block[:text], '1')
+    when :numbered
+      list_item_xml(block[:text], '2')
+    else
+      paragraph_xml(block[:lines])
+    end
+  end
+
+  def list_item_xml(text, num_id)
+    "<w:p><w:pPr><w:numPr><w:ilvl w:val=\"0\"/><w:numId w:val=\"#{num_id}\"/></w:numPr></w:pPr>#{run_xml(text)}</w:p>"
   end
 
   def paragraph_xml(lines)
@@ -89,6 +153,6 @@ class Docxout
   def run_xml(line)
     "<w:r><w:rPr><w:rFonts w:ascii=\"#{FONT}\" w:hAnsi=\"#{FONT}\" w:cs=\"#{FONT}\"/>" \
       "<w:sz w:val=\"#{SIZE}\"/><w:szCs w:val=\"#{SIZE}\"/></w:rPr>" \
-      "<w:t xml:space=\"preserve\">#{CGI.escapeHTML(line)}</w:t></w:r>"
+      "<w:t xml:space=\"preserve\">#{CGI.escapeHTML(line.gsub(BOLD, '\2'))}</w:t></w:r>"
   end
 end

data/lib/newsman/github.rb

@@ -24,8 +24,8 @@
 
 # This class represents a useful abstraction over Github API.
 class Github
-  def initialize(token)
-    @client = Octokit::Client.new(github_token: token)
+  def initialize(token, client: nil)
+    @client = client || Octokit::Client.new(github_token: token)
   end
 
   def pull_requests(username, repositories)
@@ -48,6 +48,24 @@ class Github
     end.select(&:important?)
   end
 
+  def issues_reviewed(username, repositories)
+    since = date_one_week_ago(Date.today)
+    reviewed_query = "is:issue involves:#{username} updated:>=#{since} #{repositories}"
+    closed_query = "is:issue is:closed involves:#{username} closed:>=#{since} #{repositories}"
+    puts "Searching issues reviewed using the following query: '#{reviewed_query}'"
+    reviewed = @client.search_issues(reviewed_query).total_count
+    puts "Searching issues closed using the following query: '#{closed_query}'"
+    closed = @client.search_issues(closed_query).total_count
+    IssueActivity.new(reviewed, closed)
+  end
+
+  def pull_requests_reviewed(username, repositories)
+    since = date_one_week_ago(Date.today)
+    query = "is:pr reviewed-by:#{username} updated:>=#{since} #{repositories}"
+    puts "Searching pull requests reviewed using the following query: '#{query}'"
+    @client.search_issues(query).total_count
+  end
+
   def parse_pr(pull_request)
     title = pull_request.title.to_s
     repository = pull_request.repository_url.split('/').last

data/lib/newsman/issues.rb

@@ -25,6 +25,21 @@ require 'json'
 
 IMPORTANT_ISSUE = 'soon'
 
+# Represents how many issues a user reviewed (labeled, assigned, commented, etc.)
+# and how many of those were closed, during a given period.
+class IssueActivity
+  attr_reader :reviewed, :closed
+
+  def initialize(reviewed, closed)
+    @reviewed = reviewed
+    @closed = closed
+  end
+
+  def to_s
+    "Issues reviewed: #{@reviewed} (labeled/assigned), #{@closed} closed"
+  end
+end
+
 # This class represents a GitHub Issue abstraction created by a user.
 class Issue
   attr_accessor :title, :body, :repo, :number

data/lib/newsman/report.rb

@@ -65,9 +65,11 @@ end
 
 # Report items inner class.
 class ReportItems
-  def initialize(prs, issues)
+  def initialize(prs, issues, issues_reviewed: nil, prs_reviewed: nil)
     @prs = prs || []
     @issues = issues || []
+    @issues_reviewed = issues_reviewed
+    @prs_reviewed = prs_reviewed
   end
 
   # Returns true if there are no pull requests or issues, false otherwise
@@ -78,6 +80,15 @@ class ReportItems
   def to_s
     prs_list = @prs.map(&:detailed_title).map { |obj| " - #{obj}\n" }.join
     issues_list = @issues.map(&:detailed_title).map { |obj| " - #{obj}\n" }.join
-    "Closed Pull Requests:\n#{prs_list}\nOpen Issues:\n#{issues_list}"
+    "Closed Pull Requests:\n#{prs_list}\nOpen Issues:\n#{issues_list}#{activity}"
+  end
+
+  private
+
+  def activity
+    return '' unless @issues_reviewed || @prs_reviewed
+
+    lines = [@issues_reviewed&.to_s, (@prs_reviewed ? "PRs reviewed: #{@prs_reviewed}" : nil)].compact
+    "\n#{lines.join("\n")}\n"
   end
 end

data/lib/newsman.rb

@@ -99,6 +99,8 @@ def generate
   github = Github.new(github_token)
   prs = github.pull_requests(github_username, github_repositories)
   issues = github.issues(github_username, github_repositories)
+  issues_reviewed = github.issues_reviewed(github_username, github_repositories)
+  prs_reviewed = github.pull_requests_reviewed(github_username, github_repositories)
   puts "\nNow lets test some aggregation using OpenAI\n\n"
   assistant = Assistant.new(openai_token, model: options[:model])
   # Build previous results
@@ -118,7 +120,7 @@ def generate
     reporter,
     reporter_position,
     options[:title],
-    additional: ReportItems.new(prs, issues)
+    additional: ReportItems.new(prs, issues, issues_reviewed: issues_reviewed, prs_reviewed: prs_reviewed)
   )
   full_answer = assistant.format(report.build(
                                    answer,

data/test/test_docxout.rb

@@ -50,4 +50,62 @@ class TestDocxout < Minitest::Test
       assert_includes(document_xml, 'Issue description')
     end
   end
+
+  def test_renders_markdown_headers_as_word_headings
+    Dir.mktmpdir do |temp_dir|
+      output = Docxout.new(temp_dir)
+      report = "# Title\n\n## Subtitle\n\n### Section"
+      path = File.join(temp_dir, output.print(report, 'volodya-lombrozo', 'gpt-3.5-turbo'))
+      document_xml = Zip::File.open(path) { |zip| zip.read('word/document.xml') }
+      %w[Heading1 Heading2 Heading3].each { |style| assert_includes(document_xml, %(<w:pStyle w:val="#{style}"/>)) }
+      assert_includes(document_xml, '>Title<')
+      refute_includes(document_xml, '# Title')
+    end
+  end
+
+  def test_renders_markdown_bullet_lists_as_word_lists
+    Dir.mktmpdir do |temp_dir|
+      output = Docxout.new(temp_dir)
+      report = "List is here:\n\n- one\n- two\n* three"
+      path = File.join(temp_dir, output.print(report, 'volodya-lombrozo', 'gpt-3.5-turbo'))
+      document_xml = Zip::File.open(path) { |zip| zip.read('word/document.xml') }
+      assert_includes(document_xml, '<w:numId w:val="1"/>')
+      assert_includes(document_xml, '>one<')
+      refute_includes(document_xml, '>- one<')
+    end
+  end
+
+  def test_renders_markdown_numbered_lists_as_word_lists
+    Dir.mktmpdir do |temp_dir|
+      output = Docxout.new(temp_dir)
+      report = "1. first\n2. second"
+      path = File.join(temp_dir, output.print(report, 'volodya-lombrozo', 'gpt-3.5-turbo'))
+      document_xml = Zip::File.open(path) { |zip| zip.read('word/document.xml') }
+      assert_includes(document_xml, '<w:numId w:val="2"/>')
+      assert_includes(document_xml, '>first<')
+    end
+  end
+
+  def test_strips_markdown_bold_markers
+    Dir.mktmpdir do |temp_dir|
+      output = Docxout.new(temp_dir)
+      report = 'Deploys are **blocked** until review, __really__ blocked.'
+      path = File.join(temp_dir, output.print(report, 'volodya-lombrozo', 'gpt-3.5-turbo'))
+      document_xml = Zip::File.open(path) { |zip| zip.read('word/document.xml') }
+      assert_includes(document_xml, 'Deploys are blocked until review, really blocked.')
+      refute_includes(document_xml, '**')
+      refute_includes(document_xml, '__')
+    end
+  end
+
+  def test_docx_package_includes_a_valid_numbering_part
+    Dir.mktmpdir do |temp_dir|
+      output = Docxout.new(temp_dir)
+      path = File.join(temp_dir, output.print("- one\n- two", 'volodya-lombrozo', 'gpt-3.5-turbo'))
+      Zip::File.open(path) do |zip|
+        assert(zip.find_entry('word/numbering.xml'))
+        assert(zip.find_entry('word/_rels/document.xml.rels'))
+      end
+    end
+  end
 end

data/test/test_github.rb

@@ -0,0 +1,77 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+# Copyright (c) 2024 Volodya Lombrozo
+#
+# Permission is hereby granted, free of charge, to any person obtaining a copy
+# of this software and associated documentation files (the 'Software'), to deal
+# in the Software without restriction, including without limitation the rights
+# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+# copies of the Software, and to permit persons to whom the Software is
+# furnished to do so, subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+# SOFTWARE.
+
+require 'minitest/autorun'
+require_relative '../lib/newsman/issues'
+require_relative '../lib/newsman/github'
+
+# A fake Octokit client that records every query it receives and answers
+# with a canned total_count, so we can assert on the query strings we build
+# without touching the network.
+class FakeOctokit
+  Result = Struct.new(:total_count, :items)
+
+  def initialize(counts)
+    @counts = counts
+    @queries = []
+  end
+
+  attr_reader :queries
+
+  def search_issues(query)
+    @queries << query
+    Result.new(@counts.fetch(query, 0), [])
+  end
+end
+
+class TestGithub < Minitest::Test
+  def test_issues_reviewed_builds_involves_and_closed_queries
+    client = FakeOctokit.new({})
+    github = Github.new('token', client: client)
+    github.issues_reviewed('volodya-lombrozo', 'repo:objectionary/eo')
+    assert(client.queries.any? { |q| q.include?('is:issue') && q.include?('involves:volodya-lombrozo') })
+    assert(client.queries.any? do |q|
+      q.include?('is:issue') && q.include?('is:closed') && q.include?('involves:volodya-lombrozo')
+    end)
+  end
+
+  def test_issues_reviewed_returns_reviewed_and_closed_counts
+    since = date_one_week_ago(Date.today)
+    reviewed_query = "is:issue involves:volodya-lombrozo updated:>=#{since} repo:objectionary/eo"
+    closed_query = "is:issue is:closed involves:volodya-lombrozo closed:>=#{since} repo:objectionary/eo"
+    client = FakeOctokit.new(reviewed_query => 7, closed_query => 3)
+    github = Github.new('token', client: client)
+    activity = github.issues_reviewed('volodya-lombrozo', 'repo:objectionary/eo')
+    assert_equal 7, activity.reviewed
+    assert_equal 3, activity.closed
+    assert_equal 'Issues reviewed: 7 (labeled/assigned), 3 closed', activity.to_s
+  end
+
+  def test_pull_requests_reviewed_builds_reviewed_by_query
+    since = date_one_week_ago(Date.today)
+    query = "is:pr reviewed-by:volodya-lombrozo updated:>=#{since} repo:objectionary/eo"
+    client = FakeOctokit.new(query => 5)
+    github = Github.new('token', client: client)
+    assert_equal 5, github.pull_requests_reviewed('volodya-lombrozo', 'repo:objectionary/eo')
+  end
+end

data/test/test_report.rb

@@ -73,4 +73,17 @@ class TestReport < Minitest::Test
     actual = ReportItems.new(prs, issues).to_s
     assert_equal expected, actual
   end
+
+  def test_report_items_with_activity_metrics
+    expected = <<~EXPECTED
+      Closed Pull Requests:
+
+      Open Issues:
+
+      Issues reviewed: 7 (labeled/assigned), 3 closed
+      PRs reviewed: 5
+    EXPECTED
+    actual = ReportItems.new([], [], issues_reviewed: IssueActivity.new(7, 3), prs_reviewed: 5).to_s
+    assert_equal expected, actual
+  end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: newsman
 version: !ruby/object:Gem::Version
-  version: 1.2.2
+  version: 1.3.0
 platform: ruby
 authors:
 - Volodya Lombrozo
@@ -201,6 +201,7 @@ files:
 - lib/newsman/txt_output.rb
 - test/test_assistant.rb
 - test/test_docxout.rb
+- test/test_github.rb
 - test/test_htmlout.rb
 - test/test_issue.rb
 - test/test_pdd_issue.rb