Class: Llmsherpa::Table

Inherits:
Block
  • Object
show all
Defined in:
lib/llmsherpa/blocks.rb

Overview

This conversion assumes the presence of similarly functional TableCell, Paragraph, ListItem, and Section classes or modules in Ruby.

Instance Attribute Summary

Attributes inherited from Block

#bbox, #block_idx, #block_json, #children, #left, #level, #page_idx, #parent, #sentences, #tag, #top

Instance Method Summary collapse

Methods inherited from Block

#add_child, #chunks, #iter_children, #paragraphs, #parent_chain, #parent_text, #sections, #tables, #to_context_text

Constructor Details

#initialize(table_json, _parent) ⇒ Table

Initializes a Table with child table rows and headers



304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
 
# File 'lib/llmsherpa/blocks.rb', line 304

def initialize(table_json, _parent)
  super(table_json)
  @rows = []
  @headers = []
  @name = table_json["name"]
  return unless table_json.include?("table_rows")

  table_json["table_rows"].each do |row_json|
    if row_json["type"] == "table_header"
      row = TableHeader.new(row_json)
      @headers << row
    else
      row = TableRow.new(row_json)
      @rows << row
    end
  end
end

Instance Method Details

#to_html(_include_children = false, _recurse = false) ⇒ Object

Returns html for a <table> with html from all the rows in the table as <tr>



328
329
330
331
332
333
334
 
# File 'lib/llmsherpa/blocks.rb', line 328

def to_html(_include_children = false, _recurse = false)
  html_str = "<table>"
  @headers.each { |header| html_str += header.to_html }
  @rows.each { |row| html_str += row.to_html }
  html_str += "</table>"
  html_str
end

#to_text(_include_children = false, _recurse = false) ⇒ Object

Returns text of a table with text from all the rows in the table delimited by ‘n’



323
324
325
 
# File 'lib/llmsherpa/blocks.rb', line 323

def to_text(_include_children = false, _recurse = false)
  "#{@headers.map(&:to_text).join("\n")}\n#{@rows.map(&:to_text).join("\n")}"
end