newsman 1.2.2 → 1.2.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: '0166813d55623364efd38ff1f3bc83ff3ae76ee9d793b1b5ac52924477122135'
-  data.tar.gz: 48d3c00284b0973851b95c2b144fe50972f801deeb826b400ab31e87102b7cfc
+  metadata.gz: 8bff52319f7db052d667bb38fa31764a4204ddab745eeaaa1daf090d36a73789
+  data.tar.gz: a7ccce39053631d627d210fea0897988b29ac24a62d9a17f98d83055777fbe40
 SHA512:
-  metadata.gz: 193cf28ea41b994cab278212cc255dcf0d1f7b78570e631f68e17003d489d505fd0e12a8b3244d576fa1e0826d992ead7af9c1febefd80074dc8260174294dae
-  data.tar.gz: 4e3a9c5920d471c3c5a1dcabe16692a6154ea50d44adaf775d263ce04186e814068391555d3e90682b043dbb2b3c4fb0372da2f8fa7e91abf331a594bfefe4e7
+  metadata.gz: 638afa442dd92022d7966ea99405fea3838a6a92edd12c71f32d9a7cb8392c59287e647d8376524e9a97f61ea9ce3b74b5160b85b27940627d478528d76d3eef
+  data.tar.gz: c8ee4f4a28435a66132f865046f222a2370361884d81d29536e1748778429cfd1a60581e6027448c0bbf5a491d55fda13e0d3f45af293e25b5fbe9d9fff5d5c2

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 +39,25 @@ 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*$/
+
   FONT = 'Times New Roman'
   # Word expresses font size in half-points, so 12pt is 24.
   SIZE = '24'
@@ -50,11 +69,7 @@ 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 +82,63 @@ 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)

data/test/test_docxout.rb

@@ -50,4 +50,50 @@ 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_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

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: newsman
 version: !ruby/object:Gem::Version
-  version: 1.2.2
+  version: 1.2.3
 platform: ruby
 authors:
 - Volodya Lombrozo