Module: Aircana::Contexts::ConfluenceContent

Included in:
Confluence
Defined in:
lib/aircana/contexts/confluence_content.rb

Instance Method Summary collapse

Instance Method Details

#convert_to_markdown(html_content) ⇒ Object 



14
15
16
17
18
19
20
 
# File 'lib/aircana/contexts/confluence_content.rb', line 14

def convert_to_markdown(html_content)
  return "" if html_content.nil? || html_content.empty?

  # Preprocess Confluence macros before converting to Markdown
  cleaned_html = preprocess_confluence_macros(html_content)
  ReverseMarkdown.convert(cleaned_html, github_flavored: true)
end

#fetch_page_content(page_id) ⇒ Object 



6
7
8
9
10
11
12
 
# File 'lib/aircana/contexts/confluence_content.rb', line 6

def fetch_page_content(page_id)
  Aircana.human_logger.info("Looking for page with ID `#{page_id}`")
  response = get_page_content(page_id)
  response.dig("body", "storage", "value") || ""
rescue HTTParty::Error, StandardError => e
  handle_api_error("fetch content for page #{page_id}", e, "Failed to fetch page content")
end

#log_pages_found(count, kb_name) ⇒ Object

rubocop:enable Layout/LineLength, Metrics/MethodLength



86
87
88
 
# File 'lib/aircana/contexts/confluence_content.rb', line 86

def log_pages_found(count, kb_name)
  Aircana.human_logger.info "Found #{count} pages for KB '#{kb_name}'"
end

#preprocess_confluence_macros(html) ⇒ Object

rubocop:disable Layout/LineLength, Metrics/MethodLength



23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
 
# File 'lib/aircana/contexts/confluence_content.rb', line 23

def preprocess_confluence_macros(html)
  # Process Confluence structured macros to make them compatible with Markdown conversion
  cleaned = html.dup

  # Convert code blocks with CDATA content to <pre><code> tags
  # Capture parameters separately and extract language in the replacement block
  # to avoid pathological backtracking when optional language group spans multiple code blocks
  cleaned.gsub!(
    %r{<ac:structured-macro[^>]*ac:name="code"[^>]*>(.*?)<ac:plain-text-body>\s*<!\[CDATA\[(.*?)\]\]>\s*</ac:plain-text-body>.*?</ac:structured-macro>}m
  ) do
    params = Regexp.last_match(1)
    code = Regexp.last_match(2) || ""
    language = params[/ac:name="language"[^>]*>([^<]*)/, 1]&.strip || ""
    "<pre><code class=\"language-#{language}\">#{code}</code></pre>"
  end

  # Remove empty code blocks (common issue with Confluence API)
  cleaned.gsub!(
    %r{<ac:structured-macro[^>]*ac:name="code"[^>]*>.*?<ac:plain-text-body>\s*</ac:plain-text-body>.*?</ac:structured-macro>}m, ""
  )

  # Convert panel macros to blockquotes, preserving inner content
  cleaned.gsub!(
    %r{<ac:structured-macro[^>]*ac:name="panel"[^>]*>.*?<ac:rich-text-body>(.*?)</ac:rich-text-body>.*?</ac:structured-macro>}m, '<blockquote>\1</blockquote>'
  )

  # Convert info/note/warning macros to blockquotes with indicators
  cleaned.gsub!(
    %r{<ac:structured-macro[^>]*ac:name="info"[^>]*>.*?<ac:rich-text-body>(.*?)</ac:rich-text-body>.*?</ac:structured-macro>}m, '<blockquote><strong>ℹī¸ Info:</strong> \1</blockquote>'
  )
  cleaned.gsub!(
    %r{<ac:structured-macro[^>]*ac:name="note"[^>]*>.*?<ac:rich-text-body>(.*?)</ac:rich-text-body>.*?</ac:structured-macro>}m, '<blockquote><strong>📝 Note:</strong> \1</blockquote>'
  )
  cleaned.gsub!(
    %r{<ac:structured-macro[^>]*ac:name="warning"[^>]*>.*?<ac:rich-text-body>(.*?)</ac:rich-text-body>.*?</ac:structured-macro>}m, '<blockquote><strong>⚠ī¸ Warning:</strong> \1</blockquote>'
  )

  # Remove toc macros outright: they have no rich-text-body of their own,
  # so the generic strip-and-preserve step below can't match them and
  # would otherwise span into an unrelated later macro's body, deleting
  # everything in between.
  cleaned.gsub!(
    %r{<ac:structured-macro[^>]*ac:name="toc"[^>]*>.*?</ac:structured-macro>}m, ""
  )

  # Strip other structured macros but preserve rich text body content.
  # The gaps around <ac:rich-text-body> are barred from crossing into
  # another <ac:structured-macro> tag, so a macro without its own body
  # (e.g. jira) fails this match instead of swallowing unrelated content.
  cleaned.gsub!(
    %r{<ac:structured-macro(?![^>]*ac:name="code")[^>]*>(?:(?!<ac:structured-macro).)*?<ac:rich-text-body>(.*?)</ac:rich-text-body>(?:(?!<ac:structured-macro).)*?</ac:structured-macro>}m, '\1'
  )

  # Remove any remaining Confluence-specific tags
  cleaned.gsub!(%r{</?ac:[^>]*>}m, "")

  # Clean up Confluence parameter tags
  cleaned.gsub!(%r{<ac:parameter[^>]*>.*?</ac:parameter>}m, "")

  cleaned
end

#store_page_as_markdown(page, kb_name) ⇒ Object 



90
91
92
93
94
95
96
97
98
99
 
# File 'lib/aircana/contexts/confluence_content.rb', line 90

def store_page_as_markdown(page, kb_name)
  content = page&.dig("body", "storage", "value") || fetch_page_content(page&.[]("id"))
  markdown_content = convert_to_markdown(content)

  @local_storage.store_content(
    title: page&.[]("title"),
    content: markdown_content,
    kb_name: kb_name
  )
end