patio11 · December 31, 2025 22:54
diff --git a/markdown_export_plugin.rb b/markdown_export_plugin.rb
 # Jekyll plugin to generate .md versions of posts for LLM consumption
 #
 # For any post not opting out via `no_markdown_export: true` in frontmatter,
 # generates a .md file at the same URL path with .md extension.
 #
 # Only includes allowlisted frontmatter fields to avoid exposing sensitive
 # internal fields like noindex, seo settings, etc.
 #
 # (Claude Code wrote most of this, FWIW.) I release this work into the public domain. -- patio11

 module Jekyll
  class MarkdownExportGenerator < Generator
    safe true
    priority :low

    ALLOWED_FRONTMATTER = %w[
      title
      date
      author
      permalink
      categories
      description
      blurb
    ].freeze

    # These fields are only included if explicitly set in source frontmatter
    EXPLICIT_ONLY_FRONTMATTER = %w[
      excerpt
    ].freeze

    def generate(site)
      site.posts.docs.each do |post|
        next if post.data['no_markdown_export']

        # Determine output path from post URL
        post_url = post.url.to_s.chomp('/')
        dir = File.dirname(post_url).sub(/^\//, '')  # Remove leading slash
        filename = File.basename(post_url, '.*') + '.md'

        # Build the markdown content
        content = build_export_content(post)

        # Write to a temp location and add as static file
        export_path = File.join(site.source, '.markdown_exports', dir)
        FileUtils.mkdir_p(export_path)

        file_path = File.join(export_path, filename)
        File.write(file_path, content)

        site.static_files << MarkdownExportFile.new(site, export_path, '', filename, dir)
      end
    end

    def build_export_content(post)
      # Build filtered frontmatter
      filtered = {}
      ALLOWED_FRONTMATTER.each do |key|
        if post.data.key?(key) && post.data[key] && post.data[key].to_s.strip != ''
          filtered[key] = post.data[key]
        end
      end

      # Include explicit-only fields only if they appear in source frontmatter
      source_frontmatter = extract_source_frontmatter(post.path)
      EXPLICIT_ONLY_FRONTMATTER.each do |key|
        if source_frontmatter.include?(key)
          filtered[key] = post.data[key] if post.data[key]
        end
      end

      # Always use canonical author name
      # Commented this out for the gist to save copy/pasters some pain
      # filtered['author'] = 'Patrick McKenzie (patio11)'

      # Get raw markdown content from source file
      raw_content = extract_raw_content(post.path)

      # Build output
      output = "---\n"
      filtered.each do |key, value|
        output += yaml_line(key, value)
      end
      output += "---\n\n"
      output += raw_content
      output
    end

    def extract_raw_content(path)
      return '' unless File.exist?(path)

      content = File.read(path)
      # Remove YAML frontmatter (everything between --- and ---)
      if content =~ /\A---\s*\n(.*?\n?)^---\s*\n/m
        content = $'  # Everything after the frontmatter
      end
      content
    end

    def extract_source_frontmatter(path)
      return [] unless File.exist?(path)

      content = File.read(path)
      if content =~ /\A---\s*\n(.*?\n?)^---\s*\n/m
        frontmatter_text = $1
        # Extract keys from YAML (simple regex approach)
        frontmatter_text.scan(/^(\w+):/).flatten
      else
        []
      end
    end

    def yaml_line(key, value)
      case value
      when Array
        if value.empty?
          "#{key}: []\n"
        else
          "#{key}:\n" + value.map { |v| "  - #{yaml_escape(v)}" }.join("\n") + "\n"
        end
      when Time, DateTime
        "#{key}: #{value.iso8601}\n"
      when Date
        "#{key}: #{value.to_s}\n"
      else
        "#{key}: #{yaml_escape(value.to_s)}\n"
      end
    end

    def yaml_escape(str)
      # Quote strings that contain special characters
      if str =~ /[:\#\[\]\{\}\,\&\*\?\|\-\<\>\=\!\%\@\`]/ || str =~ /\A[\s]/ || str =~ /[\s]\z/ || str.include?("\n")
        "\"#{str.gsub('"', '\\"').gsub("\n", "\\n")}\""
      else
        str
      end
    end
  end

  # Custom StaticFile that outputs to a different path than the source
  class MarkdownExportFile < StaticFile
    def initialize(site, base, dir, name, dest_dir)
      super(site, base, dir, name)
      @dest_dir = dest_dir
    end

    def destination(dest)
      File.join(dest, @dest_dir, @name)
    end
  end
 end
	# Jekyll plugin to generate .md versions of posts for LLM consumption
	#
	# For any post not opting out via `no_markdown_export: true` in frontmatter,
	# generates a .md file at the same URL path with .md extension.
	#
	# Only includes allowlisted frontmatter fields to avoid exposing sensitive
	# internal fields like noindex, seo settings, etc.
	#
	# (Claude Code wrote most of this, FWIW.) I release this work into the public domain. -- patio11

	module Jekyll
	class MarkdownExportGenerator < Generator
	safe true
	priority :low

	ALLOWED_FRONTMATTER = %w[
	title
	date
	author
	permalink
	categories
	description
	blurb
	].freeze

	# These fields are only included if explicitly set in source frontmatter
	EXPLICIT_ONLY_FRONTMATTER = %w[
	excerpt
	].freeze

	def generate(site)
	site.posts.docs.each do \|post\|
	next if post.data['no_markdown_export']

	# Determine output path from post URL
	post_url = post.url.to_s.chomp('/')
	dir = File.dirname(post_url).sub(/^\//, '') # Remove leading slash
	filename = File.basename(post_url, '.*') + '.md'

	# Build the markdown content
	content = build_export_content(post)

	# Write to a temp location and add as static file
	export_path = File.join(site.source, '.markdown_exports', dir)
	FileUtils.mkdir_p(export_path)

	file_path = File.join(export_path, filename)
	File.write(file_path, content)

	site.static_files << MarkdownExportFile.new(site, export_path, '', filename, dir)
	end
	end

	def build_export_content(post)
	# Build filtered frontmatter
	filtered = {}
	ALLOWED_FRONTMATTER.each do \|key\|
	if post.data.key?(key) && post.data[key] && post.data[key].to_s.strip != ''
	filtered[key] = post.data[key]
	end
	end

	# Include explicit-only fields only if they appear in source frontmatter
	source_frontmatter = extract_source_frontmatter(post.path)
	EXPLICIT_ONLY_FRONTMATTER.each do \|key\|
	if source_frontmatter.include?(key)
	filtered[key] = post.data[key] if post.data[key]
	end
	end

	# Always use canonical author name
	# Commented this out for the gist to save copy/pasters some pain
	# filtered['author'] = 'Patrick McKenzie (patio11)'

	# Get raw markdown content from source file
	raw_content = extract_raw_content(post.path)

	# Build output
	output = "---\n"
	filtered.each do \|key, value\|
	output += yaml_line(key, value)
	end
	output += "---\n\n"
	output += raw_content
	output
	end

	def extract_raw_content(path)
	return '' unless File.exist?(path)

	content = File.read(path)
	# Remove YAML frontmatter (everything between --- and ---)
	if content =~ /\A---\s\n(.?\n?)^---\s*\n/m
	content = $' # Everything after the frontmatter
	end
	content
	end

	def extract_source_frontmatter(path)
	return [] unless File.exist?(path)

	content = File.read(path)
	if content =~ /\A---\s\n(.?\n?)^---\s*\n/m
	frontmatter_text = $1
	# Extract keys from YAML (simple regex approach)
	frontmatter_text.scan(/^(\w+):/).flatten
	else
	[]
	end
	end

	def yaml_line(key, value)
	case value
	when Array
	if value.empty?
	"#{key}: []\n"
	else
	"#{key}:\n" + value.map { \|v\| " - #{yaml_escape(v)}" }.join("\n") + "\n"
	end
	when Time, DateTime
	"#{key}: #{value.iso8601}\n"
	when Date
	"#{key}: #{value.to_s}\n"
	else
	"#{key}: #{yaml_escape(value.to_s)}\n"
	end
	end

	def yaml_escape(str)
	# Quote strings that contain special characters
	if str =~ /[:\#\[\]\{\}\,\&\*\?\\|\-\<\>\=\!\%\@\`]/ \|\| str =~ /\A[\s]/ \|\| str =~ /[\s]\z/ \|\| str.include?("\n")
	"\"#{str.gsub('"', '\\"').gsub("\n", "\\n")}\""
	else
	str
	end
	end
	end

	# Custom StaticFile that outputs to a different path than the source
	class MarkdownExportFile < StaticFile
	def initialize(site, base, dir, name, dest_dir)
	super(site, base, dir, name)
	@dest_dir = dest_dir
	end

	def destination(dest)
	File.join(dest, @dest_dir, @name)
	end
	end
	end
No results found