Class: LlmDocsBuilder::HtmlToMarkdown::TableMarkupRenderer

Inherits:
Object
  • Object
show all
Defined in:
lib/llm_docs_builder/html_to_markdown/table_markup_renderer.rb

Overview

Handles conversion of HTML table markup to Markdown table format

Instance Method Summary collapse

Constructor Details

#initialize(inline_collapser:, block_renderer:) ⇒ TableMarkupRenderer

Initialize a new table markup renderer

Parameters:

  • inline_collapser (Proc)

    callable for collapsing inline content

  • block_renderer (Proc)

    callable for rendering block elements



11
12
13
14
# File 'lib/llm_docs_builder/html_to_markdown/table_markup_renderer.rb', line 11

def initialize(inline_collapser:, block_renderer:)
  @inline_collapser = inline_collapser
  @block_renderer = block_renderer
end

Instance Method Details

#render_table(table_node) ⇒ String

Main entry point for rendering HTML tables to Markdown

Parameters:

  • table_node (Nokogiri::XML::Node)

    the HTML table element to convert

Returns:

  • (String)

    markdown table or HTML if table cannot be converted



20
21
22
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
# File 'lib/llm_docs_builder/html_to_markdown/table_markup_renderer.rb', line 20

def render_table(table_node)
  return table_node.to_html if table_contains_nested_tables?(table_node)
  return render_table_with_rowspan_cells(table_node) if table_contains_rowspan_cells?(table_node)
  return render_table_with_colspan_cells(table_node) if table_contains_colspan_cells?(table_node)

  caption_text = caption_text_for(table_node)

  rows = table_node.css('tr').map do |row|
    cells = row.element_children.select { |child| %w[th td].include?(child.name.downcase) }
    next if cells.empty?

    header_candidate = row.ancestors('thead').any? ||
                       cells.all? { |cell| cell.name.casecmp('th').zero? }

    {
      header: header_candidate,
      values: cells.map { |cell| render_table_cell(cell) }
    }
  end.compact
  return '' if rows.empty?

  header_index = rows.find_index { |row| row[:header] }

  if header_index
    header_values = rows[header_index][:values]
    data_values = rows.each_with_index.filter_map do |row, index|
      next if index == header_index

      row[:values]
    end
  else
    header_values = rows.first[:values]
    data_values = rows.drop(1).map { |row| row[:values] }
  end

  column_count = [header_values.length, data_values.map(&:length).max || 0].max
  column_count = 1 if column_count.zero?

  header = pad_table_row(header_values, column_count)
  data_rows = data_values.map { |row| pad_table_row(row, column_count) }

  header_cells = header.map { |value| table_cell_data(value) }
  data_cells = data_rows.map { |row| row.map { |value| table_cell_data(value) } }

  column_specs = compute_table_column_specs(header_cells, data_cells)
  column_widths = column_specs.map { |spec| spec[:width] }

  lines = []
  lines.concat(format_table_row(header_cells, column_specs))
  lines << render_table_separator(column_widths)
  data_cells.each do |row_cells|
    lines.concat(format_table_row(row_cells, column_specs))
  end

  table_markdown = lines.join("\n")

  with_optional_caption(caption_text, table_markdown)
end