docubot 0.2.2 → 0.3

data/bin/docubot

@@ -99,7 +99,11 @@ if ARGS[:create]
 		end
 	end
 else
+	start = Time.now
 	bundle = DocuBot::Bundle.new( ARGS[:directory] )
+	lap = Time.now
+	puts "%.2fs to prepare the bundle..." % (lap-start)
 	bundle.write( ARGS[:writer], ARGS[:output] )
+	puts "%.2fs to write everything." % (Time.now-lap)
 end
 

data/lib/docubot.rb

@@ -20,12 +20,16 @@ module FileUtils
 end
 
 module DocuBot
-	VERSION = '0.2.2'
+	VERSION = '0.3'
 	DIR     = File.expand_path( File.dirname( __FILE__ ) )
 	
 	TEMPLATE_DIR = DIR / 'docubot/templates'
 	SHELL_DIR    = DIR / 'docubot/shells'
 	Dir.chdir( SHELL_DIR ){ SHELLS = Dir['*'] }
+	
+	def self.id_from_text( text )
+		text.strip.gsub(/[^\w.:-]+/,'-').gsub(/^-|-$/,'')
+	end
 end
 
 require 'docubot/snippet'

data/lib/docubot/bundle.rb

@@ -34,11 +34,16 @@ class DocuBot::Bundle
 					
 					# TODO: Move this bloat elsewhere.
 					if page.toc?
-						html = page.to_html
-						page.toc.scan /[a-z][\w.:-]*/i do |id|
-							# TODO: Maybe a lightweight HTML parser would be faster here? (Certainly more robust.)
-							if title = html[/\b(?:id|ID) *= *['"]#{id}['"][^>]*>([^<]+)/,1]
-								page << DocuBot::SubLink.new( page, title.strip, id )
+						ndoc = page.nokodoc
+						toc = page.toc
+						ids = if toc[','] # Comma-delimited toc interpreted as generated ids
+							toc.split(/,\s*/).map{ |title| DocuBot.id_from_text(title) }
+						else
+							toc.scan /[a-z][\w.:-]*/i
+						end
+ 						ids.each do |id|
+							if ele = ndoc.at_css("##{id}")
+								page << DocuBot::SubLink.new( page, ele.inner_text, id )
 							else
 								warn "Could not find requested toc anchor '##{id}' on #{page.html_path}"
 							end

data/lib/docubot/index.rb

@@ -29,14 +29,15 @@ class DocuBot::Index
 	def process_page( page )
 		page.keywords.split(/,\s*/).each{ |key| add( key, page ) } if page.keywords?
 
-		html = page.to_html
+		# FIXME: This is substantially slower (but way more correct) than regexp only.
 		unless page['no-index'] && page['no-index'].include?( 'headings' )
-			#TODO: Fix the regex to use a backreference to ensure the correct closing tag, once 1.8x support is not necessary
-			html.scan( %r{<h[1-6][^>]*>(.+?)</h[1-6]>}im ){ |captures| add( captures.first, page ) }
+			%w[h1 h2 h3 h4 h5 h6].each do |hn|
+				page.nokodoc.css(hn).each{ |head| add( head.inner_text, page ) }
+			end
 		end
 
 		unless page['no-index'] && page['no-index'].include?( 'definitions' )
-			html.scan( %r{<dt[^>]*>(.+?)</dt>}im ){ |captures| add captures.first, page }
+			page.nokodoc.css("dt").each{ |defn| add( defn.inner_text, page ) }
 		end
 	end
 	

data/lib/docubot/page.rb

@@ -1,7 +1,10 @@
 # encoding: UTF-8
 require 'yaml'
+require 'nokogiri'
+
 class DocuBot::Page
 	META_SEPARATOR = /^\+\+\+\s*$/ # Sort of like +++ATH0
+	AUTO_ID_ELEMENTS = %w[ h1 h2 h3 h4 h5 h6 legend caption dt ].join(',')
 
 	attr_reader :pages, :type, :folder, :file, :meta
 	attr_accessor :parent, :bundle
@@ -32,7 +35,9 @@ class DocuBot::Page
 			end
 			
 			# Raw markup, untransformed
-			@raw = parts.last
+			if @raw = parts.last && parts.last.strip
+				@raw = @raw
+			end
 		end
 	end
 	def []( key )
@@ -89,8 +94,9 @@ class DocuBot::Page
 	def to_html
 		return @cached_html if @cached_html
 
-		contents = if @raw
+		contents = if @raw && !@raw.empty?
 			# Directories with no index.* file will not have any @raw
+			# TODO: Swap the order of these once we're sure that all converters will pass raw HTML through.
 			html = DocuBot::convert_to_html( self, @raw, @type )
 			DocuBot::process_snippets( self, html )
 		end
@@ -104,14 +110,29 @@ class DocuBot::Page
 		haml = master_templates / "#{template}.haml" unless File.exists?( haml )
 		haml = master_templates / "page.haml"        unless File.exists?( haml )
 		haml = Haml::Engine.new( IO.read( haml ), DocuBot::Writer::HAML_OPTIONS )
-		contents = haml.render( Object.new, :contents=>contents, :page=>self, :global=>@bundle.toc, :root=>root )
+		html = haml.render( Object.new, :contents=>contents, :page=>self, :global=>@bundle.toc, :root=>root )
+
+		# Add IDs to elements, only if a toc entry might reference one.
+		if @raw && @meta['toc'] && @meta['toc'][',']
+			nokodoc( html ).css( AUTO_ID_ELEMENTS ).each do |node|
+				next if node.has_attribute?('id')
+				node['id'] = DocuBot.id_from_text(node.inner_text)
+			end
+			html = @nokodoc.at_css('body').children.to_html
+		end		
 
-		@cached_html = contents
+		@cached_html = html
 	end
 	def to_html!
 		@cached_html=nil
 		to_html
 	end
+	def nokodoc( html=nil )
+		@nokodoc ||= Nokogiri::HTML(html || to_html)
+	end
+	def nokodoc!
+		@nokodoc = Nokogiri::HTML(to_html!)
+	end
 end
 
 class DocuBot::SubLink

data/lib/docubot/templates/index.haml

@@ -1,14 +1,21 @@
--# TODO: Breakdown by letter.
-%ul#index
-	- global.index.entries.sort_by{ |k,p| k.downcase }.each do |keyword,pages|
-		- if pages.length == 1
-			%li
-				%a{ :href => pages.first.html_path }= keyword
-		- else
-			%li
-				= keyword
-				%ul
-					- pages.each do |page|
-						%li
-							%a{ :href => page.html_path }= page.title
+!!! Strict
+%html
+	%head
+		%meta(http-equiv='Content-Type' content='text/html; charset=utf-8')
+		%title= global.title + " Index"
+		%link{:rel=>'stylesheet', :type=>'text/css', :href=>"#{root}common.css", :media=>'all'}
+	%body
+		-# TODO: Breakdown by letter.
+		%ul#index
+			- global.index.entries.sort_by{ |k,p| k.downcase }.each do |keyword,pages|
+				- if pages.length == 1
+					%li
+						%a{ :href => pages.first.html_path }= keyword
+				- else
+					%li
+						= keyword
+						%ul
+							- pages.each do |page|
+								%li
+									%a{ :href => page.html_path }= page.title
 

data/lib/docubot/templates/toc.haml

@@ -1,7 +1,9 @@
 !!! Strict
 %html
 	%head
+		%meta(http-equiv='Content-Type' content='text/html; charset=utf-8')
 		%title= global.title
+		%link{:rel=>'stylesheet', :type=>'text/css', :href=>"#{root}common.css", :media=>'all'}
 		%link{:rel=>'stylesheet', :type=>'text/css', :href=>"#{root}toc.css", :media=>'all'}
 		%script{:type=>'text/javascript', :src=>"#{root}toc.js"}
 	%body

data/lib/docubot/writers/chm.rb

@@ -8,11 +8,14 @@ class DocuBot::CHMWriter < DocuBot::HTMLWriter
 	
 	def write( destination=nil )
 		super( nil )
+		lap = Time.now
 		@chm_path = destination || "#{@bundle.source}.chm"
 		@toc = @bundle.toc
 		write_hhc
 		write_hhk
 		write_hhp
+		puts "...%.2fs to write the CHM support files" % ((lap=Time.now)-lap)
+		
 		# This will fail if a handle is open to it on Windows
 		begin
 			FileUtils.rm( @chm_path ) if File.exists?( @chm_path )
@@ -22,12 +25,13 @@ class DocuBot::CHMWriter < DocuBot::HTMLWriter
 				process.Terminate if process.CommandLine.include? @chm_path.gsub('/','\\')
 			end
 		end
-		# TODO: output timing and progress results
 		`hhc.exe "#{FileUtils.win_path @hhp}"`.gsub( /[\r\n]+/, "\n" )
+		puts "...%.2fs to create the CHM" % ((lap=Time.now)-lap)
 		
 		# Clean out the intermediary files
 		FileUtils.rm( [ @hhc, @hhp, @hhk ] )
 		FileUtils.rm_r( @html_path )
+		puts "...%.2fs to clean up temporary files" % ((lap=Time.now)-lap)
 		
 		# Spin a new thread so it doesn't hold up the Ruby process, but sleep long enough for it to get going.
 		Thread.new{ `hh.exe "#{FileUtils.win_path @chm_path}"` }

data/lib/docubot/writers/html.rb

@@ -3,6 +3,8 @@ class DocuBot::HTMLWriter < DocuBot::Writer
 	
 	# Specify nil for destination to place "<source>_html" next to the source.
 	def write( destination=nil )
+		start = Time.now
+		
 		source = @bundle.source
 		@html_path = destination || File.dirname(source)/"#{File.basename source}_html"
 		FileUtils.rm_rf(@html_path) if File.exists?(@html_path)
@@ -54,6 +56,7 @@ class DocuBot::HTMLWriter < DocuBot::Writer
 			File.open( 'glossary-terms.js', 'w' ){ |f| f << @bundle.glossary.to_js }
 		end
 		
+		puts "...%.2fs to write the HTML" % (Time.now - start)
 	end
 end
 

data/test/site1_html/A Slight Change of Heart/1 Ze First Page in Ze Section.html

@@ -0,0 +1,30 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html>
+<head>
+<meta content='text/html; charset=utf-8' http-equiv='Content-Type'>
+<link href='../common.css' media='all' rel='stylesheet' type='text/css'>
+<title>Ze First Page in Ze Section</title>
+</head>
+<body>
+<div id='content'>
+<div id='pagetop'>
+<div id='breadcrumb'>
+Table of Contents
+<span class='sep'>&gt;</span>
+<a href='../A Slight Change of Heart/index.html'>A Slight Change of Heart</a>
+<span class='sep'>&gt;</span>
+Ze First Page in Ze Section
+</div>
+<h1 id='title'>Ze First Page in Ze Section</h1>
+</div>
+<div id='pagebody'>
+<div id='mainbody'><p>The title of this should be "Ze First Page in Ze Section"</p>
+</div>
+<div id='pagefooter'>
+Copyright ©2010.
+All Rights Reserved.
+</div>
+</div>
+</div>
+</body>
+</html>

data/test/site1_html/A Slight Change of Heart/2 Another Page in the Section.html

@@ -0,0 +1,30 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html>
+<head>
+<meta content='text/html; charset=utf-8' http-equiv='Content-Type'>
+<link href='../common.css' media='all' rel='stylesheet' type='text/css'>
+<title>Another Page in the Section</title>
+</head>
+<body>
+<div id='content'>
+<div id='pagetop'>
+<div id='breadcrumb'>
+Table of Contents
+<span class='sep'>&gt;</span>
+<a href='../A Slight Change of Heart/index.html'>A Slight Change of Heart</a>
+<span class='sep'>&gt;</span>
+Another Page in the Section
+</div>
+<h1 id='title'>Another Page in the Section</h1>
+</div>
+<div id='pagebody'>
+<div id='mainbody'><p>The title of this page should be "Another Page in the Section".</p>
+</div>
+<div id='pagefooter'>
+Copyright ©2010.
+All Rights Reserved.
+</div>
+</div>
+</div>
+</body>
+</html>

data/test/site1_html/A Slight Change of Heart/3_more_crap.html

@@ -0,0 +1,42 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html>
+<head>
+<meta content='text/html; charset=utf-8' http-equiv='Content-Type'>
+<link href='../common.css' media='all' rel='stylesheet' type='text/css'>
+<title>Additional Helpful Information</title>
+</head>
+<body>
+<div id='content'>
+<div id='pagetop'>
+<div id='breadcrumb'>
+Table of Contents
+<span class='sep'>&gt;</span>
+<a href='../A Slight Change of Heart/index.html'>A Slight Change of Heart</a>
+<span class='sep'>&gt;</span>
+Additional Helpful Information
+</div>
+<h1 id='title'>Additional Helpful Information</h1>
+<div id='author'>Gavin Kistner</div>
+</div>
+<div id='pagebody'>
+<div id='mainbody'><h2 id='frist-post'>Title Verification</h2>
+<p>The title of this page should be "Additional Helpful Information", not "More Crap".</p>
+<h2 id='moar'>TOC Verification</h2>
+<p>The TOC for this page should have sub-entries for the referenced headings.</p>
+
+<ul>
+<li>The links should be named for the headings.</li>
+<li>The links should link to the headings.</li>
+</ul>
+<h2 id='dumb'>Exclusivity</h2>
+<p>The TOC should not, however, have this last heading.</p>
+
+</div>
+<div id='pagefooter'>
+Copyright ©2010.
+All Rights Reserved.
+</div>
+</div>
+</div>
+</body>
+</html>

data/test/site1_html/A Slight Change of Heart/3_more_crap.html#frist-post

@@ -0,0 +1,29 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html>
+<head>
+<meta content='text/html; charset=utf-8' http-equiv='Content-Type'>
+<link href='../common.css' media='all' rel='stylesheet' type='text/css'>
+<title>Title Verification</title>
+</head>
+<body>
+<div id='content'>
+<div id='pagetop'>
+<div id='breadcrumb'>
+Table of Contents
+<span class='sep'>&gt;</span>
+<a href='../A Slight Change of Heart/index.html'>A Slight Change of Heart</a>
+<span class='sep'>&gt;</span>
+Title Verification
+</div>
+<h1 id='title'>Title Verification</h1>
+</div>
+<div id='pagebody'>
+<div id='mainbody'></div>
+<div id='pagefooter'>
+Copyright ©2010.
+All Rights Reserved.
+</div>
+</div>
+</div>
+</body>
+</html>

data/test/site1_html/A Slight Change of Heart/3_more_crap.html#moar

@@ -0,0 +1,29 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html>
+<head>
+<meta content='text/html; charset=utf-8' http-equiv='Content-Type'>
+<link href='../common.css' media='all' rel='stylesheet' type='text/css'>
+<title>TOC Verification</title>
+</head>
+<body>
+<div id='content'>
+<div id='pagetop'>
+<div id='breadcrumb'>
+Table of Contents
+<span class='sep'>&gt;</span>
+<a href='../A Slight Change of Heart/index.html'>A Slight Change of Heart</a>
+<span class='sep'>&gt;</span>
+TOC Verification
+</div>
+<h1 id='title'>TOC Verification</h1>
+</div>
+<div id='pagebody'>
+<div id='mainbody'></div>
+<div id='pagefooter'>
+Copyright ©2010.
+All Rights Reserved.
+</div>
+</div>
+</div>
+</body>
+</html>

data/test/site1_html/A Slight Change of Heart/index.html

@@ -0,0 +1,28 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html>
+<head>
+<meta content='text/html; charset=utf-8' http-equiv='Content-Type'>
+<link href='../common.css' media='all' rel='stylesheet' type='text/css'>
+<title>A Slight Change of Heart</title>
+</head>
+<body>
+<div id='content'>
+<div id='pagetop'>
+<div id='breadcrumb'>
+Table of Contents
+<span class='sep'>&gt;</span>
+A Slight Change of Heart
+</div>
+<h1 id='title'>A Slight Change of Heart</h1>
+</div>
+<div id='pagebody'>
+<div id='mainbody'>
+</div>
+<div id='pagefooter'>
+Copyright ©2010.
+All Rights Reserved.
+</div>
+</div>
+</div>
+</body>
+</html>

data/test/site1_html/_index.html

@@ -0,0 +1,17 @@
+<ul id='index'>
+<li>
+<a href='headers.html'>Delivery Options</a>
+</li>
+<li>
+<a href='A Slight Change of Heart/3_more_crap.html'>Exclusivity</a>
+</li>
+<li>
+<a href='headers.html'>I like Chicken, I like Liver</a>
+</li>
+<li>
+<a href='A Slight Change of Heart/3_more_crap.html'>Title Verification</a>
+</li>
+<li>
+<a href='A Slight Change of Heart/3_more_crap.html'>TOC Verification</a>
+</li>
+</ul>