Extended maintenance of Ruby versions 1.8.7 and 1.9.2 ended on July 31, 2014. Read more

In Files

  • rdoc/template.rb


Class/Module Index [+]



Cheap-n-cheerful HTML page template system. You create a template containing:

  • variable names between percent signs (%fred%)

  • blocks of repeating stuff:

      ... stuff

You feed the code a hash. For simple variables, the values are resolved directly from the hash. For blocks, the hash entry corresponding to key will be an array of hashes. The block will be generated once for each entry. Blocks can be nested arbitrarily deeply.

The template may also contain

  ... stuff

stuff will only be included in the output if the corresponding key is set in the value hash.

Usage: Given a set of templates T1, T2, etc

   values = { "name" => "Dave", state => "TX" }

   t = TemplatePage.new(T1, T2, T3)
   File.open(name, "w") {|f| t.write_html_on(f, values)}
   res = ''
   t.write_html_on(res, values)

Public Class Methods

new(*templates) click to toggle source

templates is an array of strings containing the templates. We start at the first, and substitute in subsequent ones where the string !INCLUDE! occurs. For example, we could have the overall page template containing


and substitute subpages in to it by passing [master, sub_page]. This gives us a cheap way of framing pages

               # File rdoc/template.rb, line 132
def initialize(*templates)
  result = "!INCLUDE!"
  templates.each do |content|
    result.sub!(/!INCLUDE!/, content)
  @lines = LineReader.new(result.split($/))

Public Instance Methods

expand_line(line) click to toggle source

Given an individual line, we look for %xxx% constructs and HREF:ref:name: constructs, substituting for each.

               # File rdoc/template.rb, line 201
def expand_line(line)
  # Generate a cross reference if a reference is given,
  # otherwise just fill in the name part

  line.gsub!(/HREF:(\w+?):(\w+?):/) {
    ref = @context.lookup($1)
    name = @context.find_scalar($2)

    if ref and !ref.kind_of?(Array)
      "<a href=\"#{ref}\">#{name}</a>"

  # Substitute in values for %xxx% constructs.  This is made complex
  # because the replacement string may contain characters that are
  # meaningful to the regexp (like \1)

  line = line.gsub(/%([a-zA-Z]\w*)%/) {
    val = @context.find_scalar($1) 
    val.tr('\\', "\000")

rescue Exception => e
  $stderr.puts "Error in template: #{e}"
  $stderr.puts "Original line: #{line}"
substitute_into(lines, values) click to toggle source

Substitute a set of key/value pairs into the given template. Keys with scalar values have them substituted directly into the page. Those with array values invoke substitute_array (below), which examples a block of the template once for each row in the array.

This routine also copes with the IF:key directive, removing chunks of the template if the corresponding key does not appear in the hash, and the START: directive, which loops its contents for each value in an array

               # File rdoc/template.rb, line 161
def substitute_into(lines, values)
  skip_to = nil
  result = []

  while line = lines.read

    case line

    when /^IF:(\w+)/
      lines.read_up_to(/^ENDIF:#$1/) unless @context.lookup($1)

  when /^IFNOT:(\w+)/
      lines.read_up_to(/^ENDIF:#$1/) if @context.lookup($1)

    when /^ENDIF:/

    when /^START:(\w+)/
      tag = $1
      body = lines.read_up_to(/^END:#{tag}/)
      inner_values = @context.lookup(tag)
      raise "unknown tag: #{tag}" unless inner_values
      raise "not array: #{tag}"   unless inner_values.kind_of?(Array)
      inner_values.each do |vals|
        result << substitute_into(body.dup, vals)
      result << expand_line(line.dup)


write_html_on(op, value_hash) click to toggle source

Render the templates into HTML, storing the result on op using the method <<. The value_hash contains key/value pairs used to drive the substitution (as described above)

               # File rdoc/template.rb, line 144
def write_html_on(op, value_hash)
  @context = Context.new
  op << substitute_into(@lines, value_hash).tr("\000", '\\')