blueprint-html2slim 1.1.0 → 1.3.0

data/lib/blueprint/html2slim/slim_extractor.rb

@@ -0,0 +1,286 @@
+require_relative 'slim_manipulator'
+
+module Blueprint
+  module Html2Slim
+    class SlimExtractor < SlimManipulator
+      def extract_file(file_path)
+        content = read_file(file_path)
+        structure = parse_slim_structure(content)
+
+        # Handle different extraction modes
+        sections_to_remove = []
+        sections_to_keep = []
+
+        extracted = if options[:outline]
+                      extract_outline(structure, options[:outline])
+                    elsif options[:selector]
+                      extract_by_selector(structure, options[:selector])
+                    else
+                      # Original keep/remove logic
+                      sections_to_remove = normalize_selectors(options[:remove] || [])
+                      sections_to_keep = normalize_selectors(options[:keep] || [])
+
+                      # Default removals if not keeping specific sections
+                      if sections_to_keep.empty? && sections_to_remove.empty?
+                        sections_to_remove = %w[doctype head nav header footer script]
+                      end
+
+                      # Extract content
+                      extract_content(structure, sections_to_keep, sections_to_remove)
+                    end
+
+        # Remove wrapper if requested (not for outline mode)
+        extracted = remove_outer_wrapper(extracted) if options[:remove_wrapper] && !options[:outline]
+
+        # Rebuild the Slim content
+        new_content = rebuild_extracted_content(extracted)
+
+        # Write to output file
+        output_path = options[:output] || file_path.sub(/\.slim$/, '_extracted.slim')
+        write_file(output_path, new_content)
+
+        # Build appropriate response based on extraction mode
+        if options[:outline]
+          {
+            success: true,
+            mode: 'outline',
+            depth: options[:outline]
+          }
+        elsif options[:selector]
+          {
+            success: true,
+            mode: 'selector',
+            selector: options[:selector]
+          }
+        else
+          {
+            success: true,
+            removed: sections_to_remove,
+            kept: sections_to_keep.empty? ? nil : sections_to_keep
+          }
+        end
+      rescue StandardError => e
+        { success: false, error: e.message }
+      end
+
+      private
+
+      def normalize_selectors(selectors)
+        return [] unless selectors
+
+        selectors.flat_map do |selector|
+          selector.split(',').map(&:strip).map(&:downcase)
+        end
+      end
+
+      def extract_content(structure, keep_selectors, remove_selectors)
+        result = []
+        skip_until_indent = nil
+        keep_until_indent = nil
+
+        structure.each_with_index do |item, _index|
+          # If we're in skip mode, check if we've exited the skipped section
+          if skip_until_indent
+            next if item[:indent_level] > skip_until_indent
+
+            skip_until_indent = nil
+            # Continue processing this line
+
+          end
+
+          # If we're in keep mode, track when we exit
+          keep_until_indent = nil if keep_until_indent && item[:indent_level] <= keep_until_indent
+
+          # Determine what to do with this line
+          if !keep_selectors.empty?
+            # We have keep selectors - only keep matching sections
+            if should_keep_line?(item, keep_selectors)
+              # This line matches a keep selector
+              keep_until_indent = item[:indent_level]
+              result << item
+            elsif keep_until_indent && item[:indent_level] > keep_until_indent
+              # We're inside a kept section
+              result << item
+            end
+            # Otherwise, skip this line
+          elsif should_remove_line?(item, remove_selectors)
+            # No keep selectors - use remove logic
+            skip_until_indent = item[:indent_level]
+            next
+          # Skip this line and all its children
+          else
+            result << item
+          end
+        end
+
+        result
+      end
+
+      def should_remove_line?(item, selectors)
+        return false if selectors.empty?
+
+        line = item[:stripped]
+
+        selectors.any? do |selector|
+          case selector
+          when 'doctype'
+            item[:type] == :doctype
+          when 'script', 'style', 'link', 'meta'
+            line.start_with?(selector)
+          else
+            # Check for element match or class/id match
+            element_info = element_selector(line)
+            if element_info
+              element_info[:element] == selector ||
+                element_info[:selector].include?(".#{selector}") ||
+                element_info[:selector].include?("##{selector}")
+            else
+              false
+            end
+          end
+        end
+      end
+
+      def should_keep_line?(item, selectors)
+        return true if selectors.empty?
+
+        line = item[:stripped]
+
+        selectors.any? do |selector|
+          element_info = element_selector(line)
+          if element_info
+            element_info[:element] == selector ||
+              element_info[:selector].include?(".#{selector}") ||
+              element_info[:selector].include?("##{selector}")
+          else
+            false
+          end
+        end
+      end
+
+      def remove_outer_wrapper(structure)
+        return structure if structure.empty?
+
+        # Find the minimum indentation level
+        min_indent = structure.map { |item| item[:indent_level] }.min
+
+        # If there's only one element at the minimum level, remove it
+        root_elements = structure.select { |item| item[:indent_level] == min_indent }
+
+        if root_elements.size == 1 && root_elements.first[:type] == :element
+          # Remove the wrapper and decrease indentation of all children
+          structure = structure[1..-1].map do |item|
+            item[:indent_level] -= 1 if item[:indent_level] > min_indent
+            item
+          end
+        end
+
+        structure
+      end
+
+      def rebuild_extracted_content(structure)
+        return '' if structure.empty?
+
+        # Normalize indentation - find minimum and adjust
+        min_indent = structure.map { |item| item[:indent_level] }.min || 0
+
+        structure.map do |item|
+          adjusted_indent = item[:indent_level] - min_indent
+          indent_string(adjusted_indent) + item[:stripped]
+        end.join("\n")
+      end
+
+      def extract_outline(structure, max_depth)
+        result = []
+
+        structure.each do |item|
+          # Include items up to the specified depth
+          result << item if item[:indent_level] < max_depth
+        end
+
+        result
+      end
+
+      def extract_by_selector(structure, selector)
+        result = []
+        in_selected_section = false
+        selected_indent = nil
+
+        # Parse the CSS selector
+        selector_parts = parse_css_selector(selector)
+
+        structure.each do |item|
+          # Check if we're exiting a selected section
+          if in_selected_section && selected_indent && item[:indent_level] <= selected_indent
+            in_selected_section = false
+            selected_indent = nil
+          end
+
+          # Check if this item matches the selector
+          if !in_selected_section && matches_selector?(item, selector_parts)
+            in_selected_section = true
+            selected_indent = item[:indent_level]
+            result << item
+          elsif in_selected_section
+            result << item
+          end
+        end
+
+        result
+      end
+
+      def parse_css_selector(selector)
+        # Support basic CSS selectors: element, #id, .class, element.class, element#id
+        parts = {}
+
+        # Handle complex selectors like div.container#main
+        if selector =~ /^([a-z][a-z0-9]*)?([#.][\w\-#.]*)?$/i
+          parts[:element] = ::Regexp.last_match(1)
+          selector_part = ::Regexp.last_match(2)
+
+          if selector_part
+            # Extract ID
+            parts[:id] = ::Regexp.last_match(1) if selector_part =~ /#([\w\-]+)/
+
+            # Extract classes
+            classes = selector_part.scan(/\.([\w\-]+)/).flatten
+            parts[:classes] = classes unless classes.empty?
+          end
+        elsif selector.start_with?('#')
+          # Just an ID
+          parts[:id] = selector[1..-1]
+        elsif selector.start_with?('.')
+          # Just a class
+          parts[:classes] = [selector[1..-1]]
+        else
+          # Just an element
+          parts[:element] = selector
+        end
+
+        parts
+      end
+
+      def matches_selector?(item, selector_parts)
+        line = item[:stripped]
+        element_info = element_selector(line)
+
+        return false unless element_info
+
+        # Check element match
+        return false if selector_parts[:element] && !(element_info[:element] == selector_parts[:element])
+
+        # Check ID match
+        return false if selector_parts[:id] && !element_info[:selector].include?("##{selector_parts[:id]}")
+
+        # Check class matches
+        if selector_parts[:classes]
+          selector_parts[:classes].each do |cls|
+            return false unless element_info[:selector].include?(".#{cls}")
+          end
+        end
+
+        true
+      end
+    end
+  end
+end

data/lib/blueprint/html2slim/slim_fixer.rb

@@ -0,0 +1,145 @@
+require_relative 'slim_manipulator'
+
+module Blueprint
+  module Html2Slim
+    class SlimFixer < SlimManipulator
+      def fix_file(file_path)
+        content = read_file(file_path)
+        original_content = content.dup
+
+        fixes_applied = []
+
+        if options[:fix_slashes] != false
+          content, slash_fixes = fix_slash_prefix(content)
+          fixes_applied.concat(slash_fixes)
+        end
+
+        if options[:fix_multiline] != false
+          content, multiline_fixes = fix_multiline_text(content)
+          fixes_applied.concat(multiline_fixes)
+        end
+
+        if options[:dry_run]
+          if fixes_applied.any?
+            puts "\nChanges that would be made to #{file_path}:"
+            puts "  Fixes: #{fixes_applied.join(", ")}"
+            show_diff(original_content, content) if options[:verbose]
+          else
+            puts "No issues found in #{file_path}"
+          end
+        elsif content != original_content
+          write_file(file_path, content)
+        end
+
+        { success: true, fixes: fixes_applied }
+      rescue StandardError => e
+        { success: false, error: e.message }
+      end
+
+      private
+
+      def fix_slash_prefix(content)
+        fixes = []
+        lines = content.split("\n")
+
+        lines.map!.with_index do |line, index|
+          stripped = line.strip
+          indent = line[/\A */]
+
+          # Fix text that starts with / after an element
+          if stripped =~ %r{^([a-z#.][^\s/]*)\s+(/[^/].*)$}i
+            element_part = ::Regexp.last_match(1)
+            text_part = ::Regexp.last_match(2)
+
+            # Convert to pipe notation
+            new_lines = [
+              "#{indent}#{element_part}",
+              "#{indent}#{" " * @indent_size}| #{text_part}"
+            ]
+
+            fixes << "slash text at line #{index + 1}"
+            new_lines
+          # Fix standalone text starting with slash (not a comment)
+          elsif stripped.start_with?('/') && !stripped.start_with?('/!') && !stripped.match?(%r{^/\s})
+            fixes << "slash text at line #{index + 1}"
+            "#{indent}| #{stripped}"
+          else
+            line
+          end
+        end
+
+        lines.flatten!
+        new_content = lines.join("\n")
+
+        [new_content, fixes]
+      end
+
+      def fix_multiline_text(content)
+        fixes = []
+        lines = content.split("\n")
+        result_lines = []
+        i = 0
+
+        while i < lines.size
+          line = lines[i]
+          stripped = line.strip
+          indent = line[/\A */]
+
+          # Check if this is an element with multiline text content
+          if stripped =~ /^([a-z#.][^\s]*)\s+(.+)$/i && i + 1 < lines.size
+            element_part = ::Regexp.last_match(1)
+            first_text = ::Regexp.last_match(2)
+
+            # Look ahead to see if next lines are continuation text
+            next_line_indent = lines[i + 1][/\A */]
+
+            if next_line_indent.size > indent.size && !lines[i + 1].strip.empty?
+              # This looks like multiline text that should use pipe notation
+              text_lines = [first_text]
+              j = i + 1
+
+              while j < lines.size
+                next_indent = lines[j][/\A */]
+                next_stripped = lines[j].strip
+
+                # Stop if we hit a line with same or less indentation
+                break if next_indent.size <= indent.size
+                # Stop if we hit Slim syntax
+                break if next_stripped =~ %r{^[=\-|/!#.]} || next_stripped =~ /^[a-z]+[#.\[]/i
+
+                text_lines << next_stripped
+                j += 1
+              end
+
+              if text_lines.size > 1
+                # Convert to proper multiline with pipes
+                result_lines << "#{indent}#{element_part}"
+                text_lines.each do |text|
+                  result_lines << "#{indent}#{" " * @indent_size}| #{text}"
+                end
+
+                fixes << "multiline text at line #{i + 1}"
+                i = j
+                next
+              end
+            end
+          end
+
+          result_lines << line
+          i += 1
+        end
+
+        new_content = result_lines.join("\n")
+        [new_content, fixes]
+      end
+
+      def show_diff(original, modified)
+        puts "\n--- Original ---"
+        puts original
+        puts "\n+++ Modified +++"
+        puts modified
+        puts
+      end
+    end
+  end
+end

data/lib/blueprint/html2slim/slim_manipulator.rb

@@ -0,0 +1,117 @@
+module Blueprint
+  module Html2Slim
+    class SlimManipulator
+      attr_reader :options
+
+      def initialize(options = {})
+        @options = options
+        @indent_size = options[:indent_size] || 2
+      end
+
+      protected
+
+      def read_file(file_path)
+        File.read(file_path, encoding: 'UTF-8')
+      end
+
+      def write_file(file_path, content)
+        return if options[:dry_run]
+
+        # Create backup if requested
+        if options[:backup]
+          backup_path = "#{file_path}.bak"
+          File.write(backup_path, read_file(file_path), encoding: 'UTF-8')
+        end
+
+        # Ensure content ends with newline
+        content += "\n" unless content.end_with?("\n")
+        File.write(file_path, content, encoding: 'UTF-8')
+      end
+
+      def parse_slim_structure(content)
+        lines = content.split("\n")
+        structure = []
+
+        lines.each_with_index do |line, index|
+          indent_level = line[/\A */].size / @indent_size
+          stripped = line.strip
+
+          next if stripped.empty?
+
+          structure << {
+            line: line,
+            stripped: stripped,
+            indent_level: indent_level,
+            line_number: index + 1,
+            type: detect_line_type(stripped)
+          }
+        end
+
+        structure
+      end
+
+      def detect_line_type(line)
+        case line
+        when /^doctype/i
+          :doctype
+        when %r{^/!}
+          :html_comment
+        when %r{^/\s}
+          :slim_comment
+        when /^-/
+          :ruby_code
+        when /^=/
+          :ruby_output
+        when /^ruby:/
+          :ruby_block
+        when /^\|/
+          :text_pipe
+        when /^[#.]/
+          :div_shorthand
+        when /^[a-z][a-z0-9]*/i
+          :element
+        else
+          :text
+        end
+      end
+
+      def indent_string(level)
+        ' ' * (@indent_size * level)
+      end
+
+      def rebuild_slim(structure)
+        structure.map do |item|
+          item[:modified_line] || item[:line]
+        end.join("\n")
+      end
+
+      def element_selector(line)
+        # Extract element, id, and classes from a Slim line
+        match = line.match(/^([a-z][a-z0-9]*)?([#.][\w\-#.]*)?/i)
+        return nil unless match
+
+        {
+          element: match[1] || 'div',
+          selector: match[2] || '',
+          full: match[0]
+        }
+      end
+
+      def has_slash_prefix_text?(line)
+        # Check if line has text that starts with forward slash
+        # This is a common issue that needs fixing
+        stripped = line.strip
+
+        # Check for inline text after element
+        if stripped =~ /^[a-z#.]/i
+          # Extract the text part after element definition
+          text_part = stripped.sub(/^[a-z][a-z0-9]*([#.][\w\-#.]*)?(\[.*?\])?/i, '').strip
+          return text_part.start_with?('/')
+        end
+
+        # Check for standalone text that starts with slash
+        stripped.start_with?('/') && !stripped.start_with?('/!') && !stripped.match?(%r{^/\s})
+      end
+    end
+  end
+end