A parser for TomDoc based on TomDoc 1.0.0-rc1 (02adef9b5a)
The TomDoc specification can be found at:
The latest version of the TomDoc specification can be found at:
github.com/mojombo/tomdoc/blob/master/tomdoc.md
To choose TomDoc as your only default format see
Saved Options at
RDoc::Options for instructions on setting up a
.rdoc_options file to store your project default.
There are a few differences between this parser and the specification. A best-effort was made to follow the specification as closely as possible but some choices to deviate were made.
A future version of RDoc will warn when a MUST or MUST NOT is violated and may warn when a SHOULD or SHOULD NOT is violated. RDoc will always try to emit documentation even if given invalid TomDoc.
Here are some implementation choices this parser currently makes:
This parser allows rdoc-style inline markup but you should not depended on it.
This parser allows a space between the comment and the method body.
This parser does not require the default value to be described for an optional argument.
This parser does not examine the order of sections. An Examples section may precede the Arguments section.
This class is documented in TomDoc format. Since this is a subclass of the RDoc markup parser there isn't much to see here, unfortunately.
Creates a new TomDoc parser. See also RDoc::Markup.parse
# File rdoc/tom_doc.rb, line 129
def initialize
super
@section = nil
@seen_returns = false
end
Parses TomDoc from text
A String containing TomDoc-format text.
RDoc::TomDoc.parse <<-TOMDOC This method does some things Returns nothing. TOMDOC # => #<RDoc::Markup::Document:0xXXX @parts=[...], @file=nil>
Returns an RDoc::Markup::Document representing the TomDoc format.
# File rdoc/tom_doc.rb, line 83
def self.parse text
parser = new
parser.tokenize text
doc = RDoc::Markup::Document.new
parser.parse doc
doc
end
Extracts the Signature section's method signature
An RDoc::Comment that will be parsed and have the signature extracted
Returns a String containing the signature and nil if not
# File rdoc/tom_doc.rb, line 99
def self.signature comment
return unless comment.tomdoc?
document = comment.parse
signature = nil
found_heading = false
found_signature = false
document.parts.delete_if do |part|
next false if found_signature
found_heading ||=
RDoc::Markup::Heading === part && part.text == 'Signature'
next false unless found_heading
next true if RDoc::Markup::BlankLine === part
if RDoc::Markup::Verbatim === part then
signature = part
found_signature = true
end
end
signature and signature.text
end
Builds a paragraph from the token stream
Unused
Returns an RDoc::Markup::Paragraph.
# File rdoc/tom_doc.rb, line 172
def build_paragraph margin
p :paragraph_start => margin if @debug
paragraph = RDoc::Markup::Paragraph.new
until @tokens.empty? do
type, data, = get
case type
when :TEXT then
@section = 'Returns' if data =~ /\AReturns/
paragraph << data
when :NEWLINE then
if :TEXT == peek_token[0] then
paragraph << ' '
else
break
end
else
unget
break
end
end
p :paragraph_end => margin if @debug
paragraph
end
Builds a verbatim from the token stream. A verbatim in the Examples section will be marked as in Ruby format.
The indentation from the margin for lines that belong to this verbatim section.
Returns an RDoc::Markup::Verbatim
# File rdoc/tom_doc.rb, line 158
def build_verbatim margin
verbatim = super
verbatim.format = :ruby if @section == 'Examples'
verbatim
end
Turns text into an Array of tokens
A String containing TomDoc-format text.
Returns self.
# File rdoc/tom_doc.rb, line 223
def tokenize text
text.sub!(/\A(Public|Internal|Deprecated):\s+/, '')
setup_scanner text
until @s.eos? do
pos = @s.pos
# leading spaces will be reflected by the column of the next token
# the only thing we loose are trailing spaces at the end of the file
next if @s.scan(/ +/)
@tokens << case
when @s.scan(/\r?\n/) then
token = [:NEWLINE, @s.matched, *token_pos(pos)]
@line_pos = char_pos @s.pos
@line += 1
token
when @s.scan(/(Examples|Signature)$/) then
@tokens << [:HEADER, 3, *token_pos(pos)]
[:TEXT, @s[1], *token_pos(pos)]
when @s.scan(/([:\w][\w\[\]]*)[ ]+- /) then
[:NOTE, @s[1], *token_pos(pos)]
else
@s.scan(/.*/)
[:TEXT, @s.matched.sub(/\r$/, ''), *token_pos(pos)]
end
end
self
end