blueprint-html2slim 1.3.0 → 1.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: d27f4f166468ae3be90116e4ff94d8c9901eb72a033ef69b14822d6ed8ce8273
-  data.tar.gz: c7916bc01602176018608dc723e49816670d6e6688d890aedde2ec09b4901a4a
+  metadata.gz: 8659e98075970808c3f6b98f3bf5630df930b0164f4010cc57d2393a6646e403
+  data.tar.gz: cbae380fbe76e804e24d79ba393480864f9cdbd2858d4395d55f1527dacba5aa
 SHA512:
-  metadata.gz: d019b9a611db510f8bd3ed39187c651c40dbd1dab414e7012cbb97d8e4cf06222034d45a1d06db44bf7709a3b822a12021af205fccf5b3482c325b9ca3377f89
-  data.tar.gz: 21da628071a957513c4faa67aedb8fe63ee76274efc3c9f08f6c0aeeae353b89017e7785b7fb144a8378718aaf297b9cebf1ba61ad9a623e79de411aad045dc1
+  metadata.gz: 85ee1ae5d5bc5ed1a67478c9255681fe3925141c575c29ae439abbfb77409b74002e881a12cbec444f2133659feaca90487012d44e8e73acbe1ac47ed7886458
+  data.tar.gz: 173c15886f0865772500f67cb04d06cfa425921fb0dc2fca77a0975003c6f8835ab013a3b78593dcf86824bba9899e22c5ddbb95b7b47168e2030bd248039379

data/CHANGELOG.md

@@ -2,6 +2,27 @@
 
 All notable changes to this project will be documented in this file.
 
+## [1.3.1] - 2025-01-16
+
+### Fixed
+- **SlimExtractor CSS Selector Support**: 
+  - Fixed child combinator selectors (`body > section`) not working
+  - Now correctly extracts multiple matching elements instead of just the first
+  - Added support for parent-child relationship validation in selectors
+- **Improved Content Extraction**:
+  - Enhanced default removal list to include `html`, `body`, `script` elements
+  - Added automatic cleanup of orphaned comments when sections are removed
+  - Fixed extraction logic to handle multiple sections properly
+- **Enhanced CSS Selector Parser**:
+  - Added `parse_simple_selector()` for individual selector components
+  - Implemented `matches_child_selector()` for parent-child verification
+  - Added `find_parent_item()` for structure hierarchy navigation
+
+### Changed
+- Default remove list now includes: `doctype html head nav header footer script body`
+- CSS selector extraction now handles multiple matching elements in single operation
+- Orphaned comments are automatically cleaned up during extraction
+
 ## [1.3.0] - 2025-01-16
 
 ### Added

data/lib/blueprint/html2slim/slim_extractor.rb

@@ -22,7 +22,7 @@ module Blueprint
 
                       # 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]
+                        sections_to_remove = %w[doctype html head nav header footer script body]
                       end
 
                       # Extract content
@@ -32,6 +32,9 @@ module Blueprint
         # Remove wrapper if requested (not for outline mode)
         extracted = remove_outer_wrapper(extracted) if options[:remove_wrapper] && !options[:outline]
 
+        # Clean up orphaned comments
+        extracted = clean_orphaned_comments(extracted)
+
         # Rebuild the Slim content
         new_content = rebuild_extracted_content(extracted)
 
@@ -203,26 +206,48 @@ module Blueprint
 
       def extract_by_selector(structure, selector)
         result = []
-        in_selected_section = false
-        selected_indent = nil
-
+        @current_structure = structure # Store for parent lookup
+        
         # 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
+        # For child selectors like "body > section", find all matching sections
+        if selector_parts[:combinator] == :child
+          structure.each do |item|
+            if matches_selector?(item, selector_parts)
+              # Add this item and all its children
+              result << item
+              # Add children until we hit the same or lower indent level
+              item_index = structure.index(item)
+              next unless item_index
+              
+              (item_index + 1...structure.size).each do |i|
+                child_item = structure[i]
+                break if child_item[:indent_level] <= item[:indent_level]
+                result << child_item
+              end
+            end
           end
+        else
+          # Original single-section logic for simple selectors
+          in_selected_section = false
+          selected_indent = nil
+
+          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
+            # 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
         end
 
@@ -230,7 +255,27 @@ module Blueprint
       end
 
       def parse_css_selector(selector)
-        # Support basic CSS selectors: element, #id, .class, element.class, element#id
+        # Support CSS selectors: element, #id, .class, element.class, element#id
+        # Also support child combinator: parent > child
+        parts = {}
+
+        # Handle child combinator (e.g., "body > section")
+        if selector.include?(' > ')
+          parent_child = selector.split(' > ').map(&:strip)
+          if parent_child.size == 2
+            parts[:parent] = parse_simple_selector(parent_child[0])
+            parts[:child] = parse_simple_selector(parent_child[1])
+            parts[:combinator] = :child
+            return parts
+          end
+        end
+
+        # Handle simple selectors
+        parts.merge!(parse_simple_selector(selector))
+        parts
+      end
+
+      def parse_simple_selector(selector)
         parts = {}
 
         # Handle complex selectors like div.container#main
@@ -261,6 +306,48 @@ module Blueprint
       end
 
       def matches_selector?(item, selector_parts)
+        # Handle child combinator selectors
+        if selector_parts[:combinator] == :child
+          return matches_child_selector?(item, selector_parts)
+        end
+
+        # Handle simple selectors
+        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
+
+      def matches_child_selector?(item, selector_parts)
+        # For child selector, we need to check if this item matches the child
+        # and verify its parent matches the parent selector
+        
+        # First check if this item matches the child selector
+        return false unless matches_simple_selector?(item, selector_parts[:child])
+
+        # Then find its parent and check if it matches the parent selector
+        parent_item = find_parent_item(item)
+        return false unless parent_item
+
+        matches_simple_selector?(parent_item, selector_parts[:parent])
+      end
+
+      def matches_simple_selector?(item, selector_parts)
         line = item[:stripped]
         element_info = element_selector(line)
 
@@ -281,6 +368,62 @@ module Blueprint
 
         true
       end
+
+      def find_parent_item(target_item)
+        # Find the parent of the target item by looking for the previous item
+        # with lower indentation level
+        target_indent = target_item[:indent_level]
+        target_line_num = target_item[:line_number]
+        
+        # Search backwards from target item to find parent
+        return nil unless @current_structure
+        
+        @current_structure.reverse.each do |item|
+          next if item[:line_number] >= target_line_num
+          
+          if item[:indent_level] < target_indent
+            return item
+          end
+        end
+        
+        nil
+      end
+
+      def clean_orphaned_comments(structure)
+        result = []
+        
+        structure.each_with_index do |item, index|
+          # If this is a comment, check if the next non-comment item exists
+          if item[:type] == :html_comment
+            # Look ahead to see if there's meaningful content after this comment
+            has_following_content = false
+            
+            (index + 1...structure.size).each do |next_index|
+              next_item = structure[next_index]
+              
+              # If we find content at the same or lower indent level, keep the comment
+              if next_item[:indent_level] <= item[:indent_level] && 
+                 next_item[:type] != :html_comment
+                has_following_content = true
+                break
+              end
+              
+              # If we find indented content, keep the comment
+              if next_item[:indent_level] > item[:indent_level]
+                has_following_content = true
+                break
+              end
+            end
+            
+            # Only keep the comment if there's following content
+            result << item if has_following_content
+          else
+            result << item
+          end
+        end
+        
+        result
+      end
     end
   end
 end

data/lib/blueprint/html2slim/version.rb

@@ -1,5 +1,5 @@
 module Blueprint
   module Html2Slim
-    VERSION = '1.3.0'.freeze
+    VERSION = '1.3.1'.freeze
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: blueprint-html2slim
 version: !ruby/object:Gem::Version
-  version: 1.3.0
+  version: 1.3.1
 platform: ruby
 authors:
 - Vladimir Elchinov