Class: Uniword::Transformation::HtmlElementBuilder

Inherits:
Object
  • Object
show all
Defined in:
lib/uniword/transformation/html_element_builder.rb

Overview

Builds OOXML element objects from parsed HTML (Nokogiri nodes).

Pure functions — no state, no side effects. Used by HtmlToOoxmlConverter to construct OOXML paragraphs, tables, runs.

Delegates formatting extraction to HtmlFormattingMapper.

Class Method Summary collapse

Class Method Details

.apply_css_style(paragraph, element) ⇒ Object

Apply CSS class → OOXML style mapping.



215
216
217
218
219
220
221
222
223
224
225
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 215

def self.apply_css_style(paragraph, element)
  css_class = element.attr("class")
  return unless css_class && !css_class.empty?

  mapped_style = HtmlFormattingMapper.map_css_class_to_style(css_class)
  return unless mapped_style

  paragraph.properties ||=
    Uniword::Wordprocessingml::ParagraphProperties.new
  paragraph.properties.style = mapped_style
end

.apply_heading_style(paragraph, element) ⇒ Object

Apply heading level style from tag name (h1-h6).



205
206
207
208
209
210
211
212
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 205

def self.apply_heading_style(paragraph, element)
  return unless element.name.match?(/^h[1-6]$/)

  paragraph.properties ||=
    Uniword::Wordprocessingml::ParagraphProperties.new
  heading_num = element.name[1]
  paragraph.properties.style = "Heading#{heading_num}"
end

.build_cell(html_cell) ⇒ Uniword::Wordprocessingml::TableCell?

Build an OOXML TableCell from an HTML td/th element.

Handles colspan → gridSpan, rowspan → vMerge restart, <th> → header.

Parameters:

  • html_cell (Nokogiri::XML::Element) —

    HTML td/th element

Returns:



114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 114

def self.build_cell(html_cell)
  cell = Uniword::Wordprocessingml::TableCell.new

  # Detect <th> → header cell
  cell.header = html_cell.name.downcase == "th"

  # Handle colspan → gridSpan, rowspan → vMerge restart
  colspan = html_cell.attr("colspan")
  rowspan = html_cell.attr("rowspan")

  if (colspan && colspan.to_i > 1) || (rowspan && rowspan.to_i > 1)
    cell.properties ||= Uniword::Wordprocessingml::TableCellProperties.new
    if colspan && colspan.to_i > 1
      cell.properties.grid_span =
        Uniword::Wordprocessingml::ValInt.new(value: colspan.to_i)
    end
    if rowspan && rowspan.to_i > 1
      cell.properties.v_merge =
        Uniword::Wordprocessingml::ValInt.new(value: 1) # restart
    end
  end

  # Convert cell content to paragraphs
  html_cell.css("p, div, h1, h2, h3, h4, h5, h6").each do |para_element|
    paragraph = build_paragraph(para_element)
    cell.paragraphs << paragraph if paragraph
  end

  # If no paragraphs found, create one from text content
  if cell.paragraphs.empty?
    text = html_cell.text.strip
    if text && !text.empty?
      para = Uniword::Wordprocessingml::Paragraph.new
      para.runs << create_run(text)
      cell.paragraphs << para
    end
  end

  # Always return a cell, even if empty (preserves table structure)
  cell.paragraphs.empty? ? ensure_empty_cell(cell) : cell
end

.build_children(paragraph, element) ⇒ Object

Build child nodes (text nodes and element nodes) into runs/SDTs/hyperlinks.

Handles:
→ Break run, <a href> → Hyperlink, <span class=“MsoFootnoteReference”> → FootnoteReference run, <span class=“MsoEndnoteReference”> → EndnoteReference run.



231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 231

def self.build_children(paragraph, element)
  element.children.each do |child|
    case child.type
    when Nokogiri::XML::Node::TEXT_NODE
      text = child.text.strip
      next if text.empty?

      paragraph.runs << create_run(text)
    when Nokogiri::XML::Node::ELEMENT_NODE
      case child.name.downcase
      when "w:sdt", "sdt"
        sdt = create_sdt_from_element(child)
        paragraph.sdts << sdt if sdt
      when "br"
        paragraph.runs << create_break_run(child)
      when "a"
        hyperlink = create_hyperlink(child)
        paragraph.hyperlinks << hyperlink if hyperlink
      when "img"
        # Images require binary data not available from HTML parsing; skip
      when "span"
        if footnote_reference_span?(child)
          paragraph.runs << create_footnote_reference_run(child)
        elsif endnote_reference_span?(child)
          paragraph.runs << create_endnote_reference_run(child)
        else
          run = create_run_from_element(child)
          paragraph.runs << run if run
        end
      else
        run = create_run_from_element(child)
        paragraph.runs << run if run
      end
    end
  end
end

.build_paragraph(element) ⇒ Uniword::Wordprocessingml::Paragraph?

Build an OOXML Paragraph from an HTML element.

Parameters:

  • element (Nokogiri::XML::Element) —

    HTML element

Returns:



16
17
18
19
20
21
22
23
24
25
26
27
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 16

def self.build_paragraph(element)
  paragraph = Uniword::Wordprocessingml::Paragraph.new

  apply_heading_style(paragraph, element)
  apply_css_style(paragraph, element)
  build_children(paragraph, element)

  has_content = paragraph.runs.any? ||
    paragraph.hyperlinks.any? ||
    paragraph.sdts.any?
  has_content ? paragraph : nil
end

.build_row(html_row) ⇒ Uniword::Wordprocessingml::TableRow?

Build an OOXML TableRow from an HTML tr element.

Parameters:

  • html_row (Nokogiri::XML::Element) —

    HTML tr element

Returns:



93
94
95
96
97
98
99
100
101
102
103
104
105
106
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 93

def self.build_row(html_row)
  row = Uniword::Wordprocessingml::TableRow.new
  cells = []

  html_row.css("td, th").each do |html_cell|
    cell = build_cell(html_cell)
    cells << cell if cell
  end

  return nil if cells.empty?

  cells.each { |c| row.cells << c }
  row
end

.build_table(html_table) ⇒ Uniword::Wordprocessingml::Table?

Build an OOXML Table from an HTML table element.

Handles colspan → gridSpan and rowspan → vMerge with continuation cell insertion. The grid layout is computed row-by-row to correctly place vMerge continuation cells where HTML rowspan omits cells.

Parameters:

  • html_table (Nokogiri::XML::Element) —

    HTML table element

Returns:



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
84
85
86
87
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 37

def self.build_table(html_table)
  table = Uniword::Wordprocessingml::Table.new
  html_rows = html_table.css("tr")
  return nil if html_rows.empty?

  # Track columns occupied by rowspan: col_idx → remaining continuation rows
  occupied = {}
  rows = []

  html_rows.each do |html_row|
    row = Uniword::Wordprocessingml::TableRow.new
    col_idx = 0

    html_row.css("td, th").each do |html_cell|
      # Insert vMerge continuation cells for columns occupied by rowspan
      while occupied.key?(col_idx)
        row.cells << create_vmerge_continuation_cell
        occupied[col_idx] -= 1
        occupied.delete(col_idx) if occupied[col_idx] <= 0
        col_idx += 1
      end

      cell = build_cell(html_cell)
      row.cells << cell

      # Track rowspan for continuation cell insertion in subsequent rows
      rowspan = html_cell.attr("rowspan")&.to_i
      if rowspan && rowspan > 1
        colspan = html_cell.attr("colspan")&.to_i || 1
        colspan.times { |c| occupied[col_idx + c] = rowspan - 1 }
      end

      col_idx += html_cell.attr("colspan")&.to_i || 1
    end

    # Handle remaining occupied columns at end of row
    while occupied.key?(col_idx)
      row.cells << create_vmerge_continuation_cell
      occupied[col_idx] -= 1
      occupied.delete(col_idx) if occupied[col_idx] <= 0
      col_idx += 1
    end

    rows << row unless row.cells.empty?
  end

  return nil if rows.empty?

  rows.each { |r| table.rows << r }
  table
end

.create_break_run(element) ⇒ Uniword::Wordprocessingml::Run

Create a Run with Break from an HTML
element.

Parameters:

  • element (Nokogiri::XML::Element) —


    element

Returns:



325
326
327
328
329
330
331
332
333
334
335
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 325

def self.create_break_run(element)
  run = Uniword::Wordprocessingml::Run.new
  brk = Uniword::Wordprocessingml::Break.new
  style = element.attr("style").to_s
  clear = element.attr("clear").to_s
  if style.include?("page-break-before") || style.include?("page-break-after") || clear == "all"
    brk.type = "page"
  end
  run.break = brk
  run
end

.create_endnote_reference_run(element) ⇒ Uniword::Wordprocessingml::Run

Create a Run with EndnoteReference from an HTML endnote reference span.

Parameters:

  • element (Nokogiri::XML::Element) —

    Span element with MsoEndnoteReference class

Returns:



373
374
375
376
377
378
379
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 373

def self.create_endnote_reference_run(element)
  run = Uniword::Wordprocessingml::Run.new
  id = extract_note_id(element) || "1"
  run.endnote_reference =
    Uniword::Wordprocessingml::EndnoteReference.new(id: id)
  run
end

.create_footnote_reference_run(element) ⇒ Uniword::Wordprocessingml::Run

Create a Run with FootnoteReference from an HTML footnote reference span.

Parameters:

  • element (Nokogiri::XML::Element) —

    Span element with MsoFootnoteReference class

Returns:



361
362
363
364
365
366
367
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 361

def self.create_footnote_reference_run(element)
  run = Uniword::Wordprocessingml::Run.new
  id = extract_note_id(element) || "1"
  run.footnote_reference =
    Uniword::Wordprocessingml::FootnoteReference.new(id: id)
  run
end

.create_hyperlink(element) ⇒ Uniword::Wordprocessingml::Hyperlink?

Create a Hyperlink from an HTML <a href> element.

Parameters:

  • element (Nokogiri::XML::Element) —

    <a> element

Returns:



410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 410

def self.create_hyperlink(element)
  href = element.attr("href")
  return nil unless href

  hyperlink = Uniword::Wordprocessingml::Hyperlink.new
  hyperlink.target = href

  element.children.each do |child|
    case child.type
    when Nokogiri::XML::Node::TEXT_NODE
      text = child.text.strip
      next if text.empty?

      hyperlink.runs << create_run(text)
    when Nokogiri::XML::Node::ELEMENT_NODE
      run = create_run_from_element(child)
      hyperlink.runs << run if run
    end
  end

  hyperlink.runs.any? ? hyperlink : nil
end

.create_run(text) ⇒ Uniword::Wordprocessingml::Run

Create a simple OOXML Run from text.

Parameters:

  • text (String) —

    Run text

Returns:



160
161
162
163
164
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 160

def self.create_run(text)
  run = Uniword::Wordprocessingml::Run.new
  run.text = HtmlFormattingMapper.decode_entities(text)
  run
end

.create_run_from_element(element) ⇒ Uniword::Wordprocessingml::Run?

Create an OOXML Run from an HTML element with inline formatting.

Parameters:

  • element (Nokogiri::XML::Element) —

    HTML element

Returns:



170
171
172
173
174
175
176
177
178
179
180
181
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 170

def self.create_run_from_element(element)
  text = element.text.strip
  return nil if text.empty?

  decoded_text = HtmlFormattingMapper.decode_entities(text)
  props = HtmlFormattingMapper.collect_formatting(element)

  run = Uniword::Wordprocessingml::Run.new
  run.text = decoded_text
  run.properties = props if HtmlFormattingMapper.has_formatting?(props)
  run
end

.create_sdt_from_element(element) ⇒ Uniword::Wordprocessingml::StructuredDocumentTag?

Create an OOXML SDT from an HTML w:sdt element.

Parameters:

  • element (Nokogiri::XML::Element) —

    HTML w:sdt element

Returns:



187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 187

def self.create_sdt_from_element(element)
  text = element.text.strip
  return nil if text.empty?

  sdt = Uniword::Wordprocessingml::StructuredDocumentTag.new
  sdt_props = parse_sdt_attributes(element)
  sdt.properties = sdt_props if sdt_props

  content = Uniword::Wordprocessingml::StructuredDocumentTag::Content.new
  run = Uniword::Wordprocessingml::Run.new
  run.text = HtmlFormattingMapper.decode_entities(text)
  content.runs = [run]
  sdt.content = content

  sdt
end

.create_vmerge_continuation_cell ⇒ Uniword::Wordprocessingml::TableCell

Create a vMerge continuation cell (empty cell with vMerge, no val).

Returns:



436
437
438
439
440
441
442
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 436

def self.create_vmerge_continuation_cell
  cell = Uniword::Wordprocessingml::TableCell.new
  cell.properties = Uniword::Wordprocessingml::TableCellProperties.new
  cell.properties.v_merge = Uniword::Wordprocessingml::ValInt.new
  cell.paragraphs << Uniword::Wordprocessingml::Paragraph.new
  cell
end

.endnote_reference_span?(element) ⇒ Boolean

Check if element is an endnote reference span.

Parameters:

  • element (Nokogiri::XML::Element)

Returns:

  • (Boolean) 


351
352
353
354
355
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 351

def self.endnote_reference_span?(element)
  return false unless element.name.downcase == "span"

  element.attr("class").to_s.split.include?("MsoEndnoteReference")
end

.ensure_empty_cell(cell) ⇒ Object

Ensure cell has at least one empty paragraph to preserve structure.



269
270
271
272
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 269

def self.ensure_empty_cell(cell)
  cell.paragraphs << Uniword::Wordprocessingml::Paragraph.new
  cell
end

.extract_note_id(element) ⇒ String?

Extract footnote/endnote ID from reference span element.

Tries: nested <a href=“#_ftnN”>, then text content digits.

Parameters:

  • element (Nokogiri::XML::Element)

Returns:

  • (String, nil) 


387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 387

def self.extract_note_id(element)
  # Try nested <a href="#_ftnN"> or <a href="#_ednN">
  anchor = element.at_css("a[href^='#']")
  if anchor
    href = anchor.attr("href").sub(/^#/, "")
    if href =~ /(\d+)\s*$/
      return Regexp.last_match(1)
    end
  end

  # Fall back to digit in text content
  text = element.text.strip
  if text =~ /(\d+)/
    return Regexp.last_match(1)
  end

  nil
end

.footnote_reference_span?(element) ⇒ Boolean

Check if element is a footnote reference span.

Parameters:

  • element (Nokogiri::XML::Element)

Returns:

  • (Boolean) 


341
342
343
344
345
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 341

def self.footnote_reference_span?(element)
  return false unless element.name.downcase == "span"

  element.attr("class").to_s.split.include?("MsoFootnoteReference")
end

.parse_sdt_attributes(element) ⇒ Object

Parse SDT attributes from HTML element.



275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
 
# File 'lib/uniword/transformation/html_element_builder.rb', line 275

def self.parse_sdt_attributes(element)
  attrs = element.attributes
  return nil if attrs.empty?

  sdt_props =
    Uniword::Wordprocessingml::StructuredDocumentTagProperties.new

  if attrs["showingplchdr"] || attrs["ShowingPlcHdr"]
    sdt_props.showing_placeholder_header =
      Uniword::Wordprocessingml::StructuredDocumentTag::ShowingPlaceholderHeader.new
  end

  if attrs["temporary"] || attrs["Temporary"]
    sdt_props.temporary =
      Uniword::Wordprocessingml::StructuredDocumentTag::Temporary.new
  end

  doc_part = attrs["docpart"] || attrs["DocPart"]
  if doc_part
    placeholder =
      Uniword::Wordprocessingml::StructuredDocumentTag::Placeholder.new
    doc_part_ref =
      Uniword::Wordprocessingml::StructuredDocumentTag::DocPartReference.new(value: doc_part.value)
    placeholder.doc_part = doc_part_ref
    sdt_props.placeholder = placeholder
  end

  if attrs["text"] || attrs["Text"]
    sdt_props.text =
      Uniword::Wordprocessingml::StructuredDocumentTag::Text.new(value: "whole")
  end

  id_attr = attrs["id"] || attrs["ID"]
  if id_attr
    sdt_props.id =
      Uniword::Wordprocessingml::StructuredDocumentTag::Id.new(value: id_attr.value.to_i)
  end

  if attrs["bibliography"] || attrs["Bibliography"]
    sdt_props.bibliography =
      Uniword::Wordprocessingml::StructuredDocumentTag::Bibliography.new
  end

  sdt_props
end