Merge custom langs and comment chars from jdp/master

bin/rocco

@@ -1,11 +1,14 @@
 #!/usr/bin/env ruby
-#/ Usage: rocco [-o <dir>] <file>...
+#/ Usage: rocco [-l <lang>] [-c <chars>] [-o <dir>] <file>...
 #/ Generate literate-programming-style documentation for Ruby source <file>s.
 #/
 #/ Options:
-#/   -o, --output=<dir>   Directory where generated HTML files are written
+#/   -l, --language=<lang>  The Pygments lexer to use to highlight code
+#/   -c, --comment-chars=<chars>
+#/                          The string to recognize as a comment marker
+#/   -o, --output=<dir>     Directory where generated HTML files are written
 #/
-#/       --help           Show this help message
+#/       --help             Show this help message
 
 require 'optparse'
 
@@ -28,9 +31,12 @@ end
 # Parse command line options, aborting if anything goes wrong.
 output_dir = '.'
 sources = []
+options = {}
 ARGV.options { |o|
   o.program_name = File.basename($0)
   o.on("-o", "--output=DIR") { |dir| output_dir = dir }
+	o.on("-l", "--language=LANG") { |lang| options[:language] = lang }
+	o.on("-c", "--comment-chars=CHARS") { |chars| options[:comment_chars] = Regexp.escape(chars) }
   o.on_tail("-h", "--help") { usage($stdout, 0) }
   o.parse!
 } or abort_with_note
@@ -76,8 +82,8 @@ Dir.mkdir output_dir if !File.directory?(output_dir)
 
 # Run each file through Rocco and write output.
 sources.each do |filename|
-  rocco = Rocco.new(filename, sources)
+  rocco = Rocco.new(filename, sources, options)
-  dest = File.join(output_dir, File.basename(filename, '.rb') + '.html')
+  dest = filename.split('.')[0..-2].join('.') + '.html'
   puts "rocco: #{filename} -> #{dest}"
   File.open(dest, 'wb') { |fd| fd.write(rocco.to_html) }
 end

lib/rocco.rb

@@ -57,13 +57,18 @@ end
 #### Public Interface
 
 # `Rocco.new` takes a source `filename`, an optional list of source filenames
-# for other documentation sources, and an optional `block`. When `block` is
+# for other documentation sources, an `options` hash, and an optional `block`.
-# given, it must read the contents of the file using whatever means necessary
+# The `options` hash respects two members: `:language`, which specifies which
-# and return it as a string. With no `block`, the file is read to retrieve data.
+# Pygments lexer to use; and `:comment_chars`, which specifies the comment
+# characters of the target language. The options default to `'ruby'` and `'#'`,
+# respectively.
+# When `block` is given, it must read the contents of the file using whatever
+# means necessary and return it as a string. With no `block`, the file is read
+# to retrieve data.
 class Rocco
   VERSION = '0.3'
 
-  def initialize(filename, sources=[], &block)
+  def initialize(filename, sources=[], options={}, &block)
     @file = filename
     @data =
       if block_given?
@@ -71,7 +76,10 @@ class Rocco
       else
         File.read(filename)
       end
+    defaults = { :language => 'ruby', :comment_chars => '#' }
+    @options = defaults.merge(options)
     @sources = sources
+    @comment_pattern = Regexp.new("^\\s*#{@options[:comment_chars]}")
     @sections = highlight(split(parse(@data)))
   end
 
@@ -98,13 +106,16 @@ class Rocco
 
   # Parse the raw file data into a list of two-tuples. Each tuple has the
   # form `[docs, code]` where both elements are arrays containing the
-  # raw lines parsed from the input file.
+  # raw lines parsed from the input file. The first line is ignored if it
+  # is a shebang line.
   def parse(data)
     sections = []
     docs, code = [], []
-    data.split("\n").each do |line|
+    lines = data.split("\n")
+    lines.shift if lines[0] =~ /^\#\!/
+    lines.each do |line|
       case line
-      when /^\s*#(?:\s+|$)/
+      when @comment_pattern
         if code.any?
           sections << [docs, code]
           docs, code = [], []
@@ -124,8 +135,11 @@ class Rocco
   def split(sections)
     docs_blocks, code_blocks = [], []
     sections.each do |docs,code|
-      docs_blocks << docs.map { |line| line.sub(/^\s*#\s?/, '') }.join("\n")
+      docs_blocks << docs.map { |line| line.sub(@comment_pattern, '') }.join("\n")
-      code_blocks << code.join("\n")
+      code_blocks << code.map do |line|
+        tabs = line.match(/^(\t+)/)
+        tabs ? line.sub(/^\t+/, '  ' * tabs.captures[0].length) : line
+      end.join("\n")
     end
     [docs_blocks, code_blocks]
   end
@@ -147,7 +161,7 @@ class Rocco
     # Pygments. We `popen` a read/write pygmentize process in the parent and
     # then fork off a child process to write the input.
     code_html = nil
-    open("|pygmentize -l ruby -f html", 'r+') do |fd|
+    open("|pygmentize -l #{@options[:language]} -f html", 'r+') do |fd|
       pid =
         fork {
           fd.close_read

lib/rocco/layout.rb

@@ -31,7 +31,7 @@ class Rocco::Layout < Mustache
       {
         :path => source,
         :basename => File.basename(source),
-        :url => File.basename(source, '.rb') + '.html'
+        :url => File.basename(source).split('.')[0..-2].join('.') + '.html'
       }
     end
   end

lib/rocco/tasks.rb

@@ -33,6 +33,15 @@
 #
 #     Rocco::make 'html/', ['lib/thing.rb', 'lib/thing/*.rb']
 #
+# Finally, it is also possible to specify which Pygments language you would
+# like to use to highlight the code, as well as the comment characters for the
+# language in the `options` hash:
+#
+#    Rocco::make 'html/', 'lib/thing/**/*.rb', {
+#      :language => 'io',
+#      :comment_chars => '#'
+#    }
+#
 
 # Might be nice to defer this until we actually need to build docs but this
 # will have to do for now.
@@ -42,17 +51,18 @@ require 'rocco'
 # of sugar over `Rocco::Task.new`. If you want your Rake task to be named
 # something other than `:rocco`, you can use `Rocco::Task` directly.
 class Rocco
-  def self.make(dest='docs/', source_files='lib/**/*.rb')
+  def self.make(dest='docs/', source_files='lib/**/*.rb', options={})
-    Task.new(:rocco, dest, source_files)
+    Task.new(:rocco, dest, source_files, options)
   end
 
   # `Rocco::Task.new` takes a task name, the destination directory docs
   # should be built under, and a source file pattern or file list.
   class Task
-    def initialize(task_name, dest='docs/', sources='lib/**/*.rb')
+    def initialize(task_name, dest='docs/', sources='lib/**/*.rb', options={})
       @name = task_name
       @dest = dest[-1] == ?/ ? dest : "#{dest}/"
       @sources = FileList[sources]
+      @options = options
 
       # Make sure there's a `directory` task defined for our destination.
       define_directory_task @dest
@@ -60,7 +70,7 @@ class Rocco
       # Run over the source file list, constructing destination filenames
       # and defining file tasks.
       @sources.each do |source_file|
-        dest_file = File.basename(source_file, '.rb') + '.html'
+        dest_file = File.basename(source_file).split('.')[0..-2].join('.') + '.html'
         define_file_task source_file, "#{@dest}#{dest_file}"
 
         # If `rake/clean` was required, add the generated files to the list.
@@ -92,7 +102,7 @@ class Rocco
       prerequisites = [@dest, source_file] + rocco_source_files
       file dest_file => prerequisites do |f|
         verbose { puts "rocco: #{source_file} -> #{dest_file}" }
-        rocco = Rocco.new(source_file, @sources.to_a)
+        rocco = Rocco.new(source_file, @sources.to_a, @options)
         File.open(dest_file, 'wb') { |fd| fd.write(rocco.to_html) }
       end
       task @name => dest_file