Fixes from Gitlab colleagues to make rubocop checks work (ce3607db) · Commits · Centre for Testing and Interoperability / Markdown specifications development / Specification tools

gitlabFilter/grid_table_filter.rb

+136 −117

Original line number	Diff line number	Diff line
		@@ -18,6 +18,7 @@
		module Banzai
		module Filter
		class GridTableFilter < HTML::Pipeline::TextFilter
		# rubocop:disable Lint/MixedRegexpCaptureTypes -- PoC
		MARKDOWN_GRID_TABLE_BLOCK_REGEX = %r{
		(?<code>
		# Grid table blocks:
		@@ -31,6 +32,7 @@ module Banzai

		)
		}mx
		# rubocop:enable Lint/MixedRegexpCaptureTypes

		require 'logger'

		@@ -43,7 +45,8 @@ module Banzai
		NEXT_ELEMENT_LIST_MARK = "∆"

		class Cell
		attr_accessor :content, :rowspan, :colspan, :colspan_adjusted, :alignment, :position_start, :position, :list_flag
		attr_accessor :content, :rowspan, :colspan, :colspan_adjusted, :alignment, :position_start, :position,
		:list_flag

		def initialize
		@content = nil
		@@ -56,8 +59,7 @@ module Banzai
		@list_flag = false
		end

		def calculateAndSetAlignment(header_delimiter_positions, default_alignments )

		def calculate_and_set_alignment(header_delimiter_positions, default_alignments)
		raise "Cell position must be set before calculating alignment" if @position.nil? \|\| @position_start.nil?

		header_delimiter_index = 0
		@@ -69,9 +71,8 @@ module Banzai
		raise "Invalid table formatting" unless header_delimiter_index < default_alignments.length

		@alignment = default_alignments[header_delimiter_index]

		end
		end # end of class Cell
		end

		class Row
		attr_accessor :cells
		@@ -87,7 +88,7 @@ module Banzai
		def []=(index, value)
		@cells[index] = value
		end
		end # end of class Row
		end

		class RowTracker
		attr_accessor :row_tracker
		@@ -104,55 +105,61 @@ module Banzai
		@row_tracker[index] = value
		end

		def maxValue
		def max_value
		@row_tracker.max
		end
		end # end of class RowTracker
		end

		# Helper method to detect separator lines
		def separator?(line)
		GRID_TABLE_SEPARATOR.match?(line)
		end

		# Helper method to handle content in cells
		# rubocop:disable Metrics/PerceivedComplexity -- PoC
		def handling_content(cell, content)
		_c = content.strip
		modified_content = content.strip
		if cell.content.nil?
		cell.rowspan += 1
		cell.colspan += 1
		if _c.start_with?("- ") # List
		if modified_content.start_with?("- ") # List
		cell.list_flag = true
		_c = _c.gsub(/\\\s*$/, '\n')
		cell.content = "#{_c}#{NEXT_ELEMENT_LIST_MARK}" # Add list element end mark to know when the list element ends
		modified_content = modified_content.gsub(/\\\s*$/, '\n')

		# Add list element end mark to know when the list element ends
		cell.content = "#{modified_content}#{NEXT_ELEMENT_LIST_MARK}"
		elsif cell.list_flag && !content.strip.empty?
		_c = _c.gsub(/\\\s*$/, '\n')
		cell.content = "#{_c}#{NEXT_ELEMENT_LIST_MARK}" #add the list element end mark
		elsif _c.empty?
		modified_content = modified_content.gsub(/\\\s*$/, '\n')
		cell.content = "#{modified_content}#{NEXT_ELEMENT_LIST_MARK}" # add the list element end mark
		elsif modified_content.empty?
		cell.content = "\n"
		else
		cell.content = _c.gsub(/\\\s*$/, "\n")
		cell.content = modified_content.gsub(/\\\s*$/, "\n")
		end
		elsif _c.start_with?("- ")
		elsif modified_content.start_with?("- ")
		cell.content += "\n" unless cell.list_flag
		cell.list_flag = true
		_c = _c.gsub(/\\\s*$/, '\n')
		cell.content += "#{_c}#{NEXT_ELEMENT_LIST_MARK}"
		elsif cell.list_flag && !_c.empty?
		cell.content = cell.content.strip.chomp("#{NEXT_ELEMENT_LIST_MARK}")
		_c = _c.gsub(/\\\s*$/, '\n')
		cell.content += " #{_c}#{NEXT_ELEMENT_LIST_MARK}"
		elsif _c.empty?
		modified_content = modified_content.gsub(/\\\s*$/, '\n')
		cell.content += "#{modified_content}#{NEXT_ELEMENT_LIST_MARK}"
		elsif cell.list_flag && !modified_content.empty?
		cell.content = cell.content.strip.chomp(NEXT_ELEMENT_LIST_MARK.to_s)
		modified_content = modified_content.gsub(/\\\s*$/, '\n')
		cell.content += " #{modified_content}#{NEXT_ELEMENT_LIST_MARK}"
		elsif modified_content.empty?
		if cell.list_flag
		cell.list_flag = false
		cell.content += "\n\n"
		end

		cell.content += cell.content.end_with?("\n") ? "" : "\n"
		else
		_c = _c.gsub(/\\\s*$/, "\n")
		cell.content += " #{_c}"
		modified_content = modified_content.gsub(/\\\s*$/, "\n")
		cell.content += " #{modified_content}"
		end

		cell
		end
		# rubocop:enable Metrics/PerceivedComplexity

		# Helper method to adjust colspan
		def adjust_colspan(row, column_index, number_of_parts, line, number_of_columns, delimiter_positions)
		@@ -167,7 +174,7 @@ module Banzai

		delimiters = ['\|', '+']
		positions = delimiters.filter_map do \|delimiter\|
		pos = line[delimiter_start + 1..-1]&.index(delimiter)
		pos = line[delimiter_start + 1..]&.index(delimiter)
		pos ? pos + delimiter_start + 1 : nil
		end.compact

		@@ -190,37 +197,39 @@ module Banzai
		row[column_index]
		end

		def checkDelimiterAlignment(line, delimiterPositions)
		return false if line.empty? \|\| delimiterPositions.empty?
		def check_delimiter_alignment(line, delimiter_positions)
		return false if line.empty? \|\| delimiter_positions.empty?

		# puts "\nChecking line: #{line}"
		#puts "Expected delimiter positions: #{delimiterPositions}"
		# puts "Expected delimiter positions: #{delimiter_positions}"

		# For any row (only +, only \|, mix of + and \|)
		currentPositions = []
		current_positions = []
		start_pos = 1

		while start_pos < line.length
		pos = line.index(/[\|+]/, start_pos) # Find the next occurrence of \| or + starting from start_pos
		break if pos.nil? # Exit if no more delimiters are found

		currentPositions << pos
		current_positions << pos
		start_pos = pos + 1 # Move to the next character after the found delimiter
		end

		#puts "Current positions: #{currentPositions}"
		# puts "Current positions: #{current_positions}"

		# Check if the last expected delimiter position is found in currentPositions
		return currentPositions.include?(delimiterPositions[-1]) &&
		# Check if the last expected delimiter position is found in current_positions
		current_positions.include?(delimiter_positions[-1]) &&
		line.match?(/\A[\|+]/) && # Check if the line starts with \| or +
		currentPositions.all? { \|pos\| delimiterPositions.include?(pos) } # Ensure all current positions are in delimiterPositions
		# Ensure all current positions are in delimiter_positions
		current_positions.all? do \|pos\|
		delimiter_positions.include?(pos)
		end
		end

		# rubocop:disable Metrics/AbcSize -- PoC
		# rubocop:disable Metrics/CyclomaticComplexity -- PoC
		# rubocop:disable Metrics/PerceivedComplexity -- PoC
		def parse_pandoc_table_with_spans(pandoc_table)

		# Split the input into lines
		lines = pandoc_table.rstrip.split("\n").map(&:rstrip)

		@@ -237,7 +246,7 @@ module Banzai
		separator_index_max_columns = separator_indices.find { \|i\| lines[i].count("+") - 1 == number_of_columns }
		number_of_columns.times do \|j\|
		start_pos = j == 0 ? 0 : delimiter_positions[j - 1]
		pos = lines[separator_index_max_columns][start_pos + 1..-1]&.index("+")
		pos = lines[separator_index_max_columns][start_pos + 1..]&.index("+")
		delimiter_positions << (pos ? pos + start_pos + 1 : -1)
		end

		@@ -266,7 +275,7 @@ module Banzai
		end

		start_pos = part_index == 0 ? 0 : header_delimiter_positions[part_index - 1]
		pos = lines[index][start_pos + 1..-1]&.index("+")
		pos = lines[index][start_pos + 1..]&.index("+")
		header_delimiter_positions << (pos ? pos + start_pos + 1 : -1)
		end
		break
		@@ -287,14 +296,14 @@ module Banzai
		end

		start_pos = part_index == 0 ? 0 : header_delimiter_positions[part_index - 1]
		pos = lines[0][start_pos + 1..-1]&.index("+")
		pos = lines[0][start_pos + 1..]&.index("+")
		header_delimiter_positions << (pos ? pos + start_pos + 1 : -1)
		end
		end

		# Check end table delimiter alignment (not checked during the lines processing)
		raise "Misaligned delimiters in table separators: #{lines[-1]}" unless checkDelimiterAlignment(lines[-1], delimiter_positions)

		raise "Misaligned delimiters in table separators: #{lines[-1]}" unless check_delimiter_alignment(lines[-1],
		delimiter_positions)

		# Process table body (including rows belonging to header as they are processed in the same way)
		data_rows = []
		@@ -315,7 +324,8 @@ module Banzai
		if separator?(line) && !in_data_row
		in_data_row = true
		# Check end table delimiter alignment (not checked during the lines processing)
		raise "Misaligned delimiters in separator row: #{line}" unless checkDelimiterAlignment(line, delimiter_positions)
		raise "Misaligned delimiters in separator row: #{line}" unless check_delimiter_alignment(line,
		delimiter_positions)

		parts = line.strip.delete_prefix("+").split("+")
		delimiter_index = 0
		@@ -329,7 +339,7 @@ module Banzai
		delimiter_index += parts[j].length + 1
		rows[-1][i].position_start = delimiter_index - parts[j].length
		rows[-1][i].position = delimiter_index
		rows[-1][i].calculateAndSetAlignment(header_delimiter_positions, default_alignments )
		rows[-1][i].calculate_and_set_alignment(header_delimiter_positions, default_alignments)

		i += 1 while delimiter_index > delimiter_positions[i]
		i += 1
		@@ -339,7 +349,8 @@ module Banzai
		# Regular data row or partial separator
		if GRID_TABLE_BODY_SEPARATOR.match?(line) # Partial separator
		# Check end table delimiter alignment (not checked during the lines processing)
		raise "Misaligned delimiters in partial separator: #{line}" unless checkDelimiterAlignment(line, delimiter_positions)
		raise "Misaligned delimiters in partial separator: #{line}" unless check_delimiter_alignment(line,
		delimiter_positions)

		cells_content = line.strip.gsub(/^(\+\|\\|)/, '').split(/[\\|\+]/)

		@@ -353,7 +364,8 @@ module Banzai
		aux_delimiter_index += cells_content[i].length + 1
		rows[-1][auxiliar_cell_index].position_start = aux_delimiter_index - cells_content[i].length
		rows[-1][auxiliar_cell_index].position = aux_delimiter_index
		rows[-1][auxiliar_cell_index].calculateAndSetAlignment(header_delimiter_positions, default_alignments )
		rows[-1][auxiliar_cell_index].calculate_and_set_alignment(header_delimiter_positions,
		default_alignments)

		auxiliar_cell_index += 1 while aux_delimiter_index > delimiter_positions[auxiliar_cell_index]

		@@ -363,11 +375,11 @@ module Banzai
		raise "More cells than columns found" unless cells_content.length <= number_of_columns

		column_index = 0
		maxRowTracker = rows_tracker.maxValue
		max_row_tracker = rows_tracker.max_value

		cells_content.each_with_index do \|content, _i\|
		if GRID_TABLE_BODY_SEPARATOR_LINE.match?(content) # Separator - split row
		rows_tracker[column_index] = maxRowTracker + 1
		rows_tracker[column_index] = max_row_tracker + 1
		rows[rows_tracker[column_index]][column_index].list_flag = false

		column_forward = 0
		@@ -401,9 +413,10 @@ module Banzai
		end
		end
		else # Data row
		cells_content = line.strip.delete_prefix("\|").split(/\\|/)
		cells_content = line.strip.delete_prefix("\|").split("\|")
		# Check end table delimiter alignment (not checked during the lines processing)
		raise "Misaligned delimiters in row: #{line}" unless checkDelimiterAlignment(line, delimiter_positions)
		raise "Misaligned delimiters in row: #{line}" unless check_delimiter_alignment(
		line, delimiter_positions)

		column_index = 0

		@@ -440,9 +453,7 @@ module Banzai
		end
		end

		if has_header && start >= header_separator_index
		rows.each { \|body_row\| data_rows << body_row.cells }
		elsif has_header && start < header_separator_index
		if has_header && start < header_separator_index
		rows.each { \|header_row\| header_rows << header_row.cells }
		else
		rows.each { \|body_row\| data_rows << body_row.cells }
		@@ -459,9 +470,11 @@ module Banzai

		cell.content = cell.content.gsub(/</, "<")

		cell.content = cell.content.gsub(/(?<espace>^\|\s)(?<bold>\\\|__)(?<text>.+?)\g<bold>(?!\w)/, "\\k<espace><strong>\\k<text></strong>")
		cell.content = cell.content.gsub(/(?<espace>^\|\s)(?<bold>\\\|__)(?<text>.+?)\g<bold>(?!\w)/,
		"\\k<espace><strong>\\k<text></strong>")

		cell.content = cell.content.gsub(/(?<espace>^\|\s)(?<italic>\*\|_)(?<text>.+?)\g<italic>(?!\w)/, "\\k<espace><i>\\k<text></i>")
		cell.content = cell.content.gsub(/(?<espace>^\|\s)(?<italic>\*\|_)(?<text>.+?)\g<italic>(?!\w)/,
		"\\k<espace><i>\\k<text></i>")
		# Convert newlines to HTML breaks
		cell.content = cell.content&.gsub("\n", "<br />")
		end
		@@ -491,7 +504,7 @@ module Banzai
		end

		[header_rows, data_rows]
		end # end of parse_pandoc_table_with_spans
		end

		def generate_html_table_with_spans(pandoc_table)
		begin
		@@ -522,14 +535,17 @@ module Banzai
		next if cell.rowspan == 0 \|\| cell.colspan == 0

		# Prepare content, in case there's a list
		matches = cell.content&.scan(/\s([-+]\|\d+\.)\s+([^#{NEXT_ELEMENT_LIST_MARK}]+?)#{NEXT_ELEMENT_LIST_MARK}\n?/)
		matches = cell.content&.scan(
		/\s([-+]\|\d+\.)\s+([^#{NEXT_ELEMENT_LIST_MARK}]+?)#{NEXT_ELEMENT_LIST_MARK}\n?/o)

		if matches
		list = "<ul>"
		matches.each do \|match\|
		list += "<li>#{match[1]}</li>"
		end
		list += "</ul>"
		cell.content = cell.content.gsub(/(\s([-+]\|\d+\.)\s+([^#{NEXT_ELEMENT_LIST_MARK}]+#{NEXT_ELEMENT_LIST_MARK}\n?))+/, list)
		cell.content = cell.content.gsub(
		/(\s([-+]\|\d+\.)\s+([^#{NEXT_ELEMENT_LIST_MARK}]+#{NEXT_ELEMENT_LIST_MARK}\n?))+/o, list)
		# Enforce left alignment if cell contains a list
		cell.alignment = 'align="left"'
		end
		@@ -549,13 +565,16 @@ module Banzai
		row.each do \|cell\|
		next if cell.rowspan == 0 \|\| cell.colspan == 0

		matches = cell.content&.scan(/\s([-+]\|\d+\.)\s+([^#{NEXT_ELEMENT_LIST_MARK}]+?)#{NEXT_ELEMENT_LIST_MARK}\n?/)
		matches = cell.content&.scan(
		/\s([-+]\|\d+\.)\s+([^#{NEXT_ELEMENT_LIST_MARK}]+?)#{NEXT_ELEMENT_LIST_MARK}\n?/o)

		if matches
		list = "<ul>"
		matches.each do \|match\|
		list += "<li>#{match[1]}</li>"
		end
		cell.content = cell.content.gsub(/(\s([-+]\|\d+\.)\s+([^#{NEXT_ELEMENT_LIST_MARK}]+#{NEXT_ELEMENT_LIST_MARK}\n?))+/, list)
		cell.content = cell.content.gsub(
		/(\s([-+]\|\d+\.)\s+([^#{NEXT_ELEMENT_LIST_MARK}]+#{NEXT_ELEMENT_LIST_MARK}\n?))+/o, list)
		# Enforce left alignment if cell contains a list
		cell.alignment = 'align="left"'
		end
		@@ -571,7 +590,7 @@ module Banzai
		html += "</table>"
		html
		end
		end # end of def generate_html_table_with_spans
		end
		# rubocop:enable Metrics/PerceivedComplexity
		# rubocop:enable Metrics/CyclomaticComplexity
		# rubocop:enable Metrics/AbcSize
		@@ -594,7 +613,7 @@ module Banzai
		match.to_s
		end
		end
		end # end of def call
		end # end of class GridTableFilter
		end # end of module Filter
		end # end of module Banzai
		end
		end
		end
		end