cmless 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: bdb15f8e0031cc6134590b4476267f2de933d79b
+  data.tar.gz: c9648c323e096ff9a2bf767507ebf18a1ee0e5af
+SHA512:
+  metadata.gz: b59096de7971a1f6695581fbf8ad61c856f5df2fa0da03b4ad372f02982d03daa446d019f89ed593b26806801eb57f33166ea1c6039ac38d90bf7c9474ef862e
+  data.tar.gz: 9df87894d707f101b566a58d781086e7173bb50a16541a1016c1d61e1b7f84e544de8d70e80b624e86eb29f602ad7e30eb1425b3b3652db8c253fb17d9a397a8

data/lib/cmless.rb

@@ -0,0 +1,145 @@
+# require_relative 'cmless/cmless'
+# If this becomes multiple files, then require a subdirectory.
+# but if it's just one file, it's fine here.
+
+require 'redcarpet'
+require 'singleton'
+require 'nokogiri'
+
+class Cmless
+  
+  attr_reader :path
+  attr_reader :title
+  
+  private
+  
+  # You should use find_by_path rather than creating your own objects.
+  def initialize(file_path)
+    @path = self.class.path_from_file_path(file_path)
+    Nokogiri::HTML(Markdowner.instance.render(File.read(file_path))).tap do |doc|
+      @title = doc.xpath('//h1').first.remove.text
+
+      html_methods = self.class.instance_methods.
+        select { |method| method.to_s.match(/\_html$/) }
+      
+      if html_methods.include?(:head_html)
+        self.instance_variable_set('@head_html', Cmless.extract_head_html(doc))
+        html_methods.delete(:head_html)
+      end
+      
+      if html_methods.include?(:body_html)
+        self.instance_variable_set('@body_html', Cmless.extract_body_html(doc))
+        html_methods.delete(:body_html)
+      end
+      
+      html_methods.each do |method|
+        h2_name = method.to_s.gsub(/\_html$/, '').gsub('_',' ').capitalize
+        variable_name = "@#{method.to_s}"
+        self.instance_variable_set(variable_name, Cmless.extract_html(doc, h2_name))
+      end
+      
+      doc.text.strip.tap do |extra|
+        escaped = extra.gsub("\n",'\\n').gsub("\t",'\\t')
+        fail("#{file_path} has extra unused text: '#{escaped}'") unless extra == ''
+      end
+    end
+  end
+  
+  public
+  
+  # Instance methods:
+  
+  def ancestors
+    @ancestors ||= begin
+      split = path.split('/')
+      (1..split.size-1).to_a.map do |i|
+        self.class.objects_by_path[split[0,i].join('/')]
+      end
+    end
+  end
+  
+  def children
+    @children ||= begin
+      self.class.objects_by_path.select do |other_path, other_object|
+        other_path.match(/^#{path}\/[^\/]+$/) # TODO: escape
+      end.map do |other_path, other_object|
+        other_object
+      end
+    end
+  end
+  
+  
+  # Class methods:
+  
+  def self.objects_by_path
+    @objects_by_path ||=
+      begin
+        unless File.directory?(self.root_path)
+          raise StandardError.new("#{self.root_path} is not a directory")
+        end
+        Hash[
+          Dir[Pathname(self.root_path) + '**/*.md'].sort.map do |path|
+            object = self.new(path)
+            [object.path, object]
+          end
+        ]
+      end
+  end
+  
+  def self.find_by_path(path)
+    self.objects_by_path[path] || raise(IndexError.new("'#{path}' is not a valid path under '#{self.root_path}'; Expected one of #{self.objects_by_path.keys}"))
+  end
+
+  def self.path_from_file_path(file_path)
+    file_path.to_s.gsub(self.root_path.to_s+'/', '').gsub(/\.md$/, '')
+  end
+  
+  def self.extract_html(doc, title)
+    following_siblings = []
+    doc.xpath("//h2[text()='#{title}']").first.tap do |header|
+      raise IndexError.new("Can't find header '#{title}'") unless header
+      while header.next_element && !header.next_element.name.match(/h2/) do
+        following_siblings.push(header.next_element.remove)
+      end
+      header.remove
+    end
+    following_siblings.map { |el| el.to_s }.join
+  end
+  
+  def self.extract_head_html(doc)
+    siblings = []
+    body = doc.xpath('//body').first
+    while body.children.first && !body.children.first.name.match(/h2/)
+      siblings.push(body.children.first.remove)
+    end
+    siblings.map { |el| el.to_s }.join.strip
+  end
+  
+  def self.extract_body_html(doc)
+    siblings = []
+    body = doc.xpath('//body').first
+    while body.children.first
+      siblings.push(body.children.first.remove)
+    end
+    siblings.map { |el| el.to_s }.join.strip
+  end
+  
+  
+  # Utility class: (This could move.)
+  
+  class Markdowner
+    include Singleton
+  
+    def initialize()
+      @markdown = Redcarpet::Markdown.new(
+                    Redcarpet::Render::XHTML.new(with_toc_data: true), 
+                    autolink: true)
+    end
+  
+    def render(md_text)
+      return unless md_text
+      @markdown.render(md_text)
+    end
+  end
+  
+end

metadata

@@ -0,0 +1,46 @@
+--- !ruby/object:Gem::Specification
+name: cmless
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- Chuck McCallum
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2015-07-22 00:00:00.000000000 Z
+dependencies: []
+description: 'CMS alternative: Maintain content in markdown / Extract HTML and data
+  for display'
+email: chuck_mccallum@wgbh.org
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/cmless.rb
+homepage: https://github.com/WGBH/cmless
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.2.3
+signing_key: 
+specification_version: 4
+summary: CMS, but less
+test_files: []
+has_rdoc: