REXML -- Parsing doctypes (and some some bugs)

ruby-talk
1

(1) How does one parse a doctype in REXML?

Hack solution --

···

#+++++++
require 'rexml/document'
data =<<EOF
<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE Resource SYSTEM "foo.dtd">
<Resource>
     <bar>baz</bar>
</Resource>
EOF
p REXML::Document.new(data).doctype.instance_variable_get("@long_name")[1..-2] # => "foo.dtd"
#+++++++

There must be a simple solution I am missing, what is it?

(2) In rexml/dtd/elementdecl, PATTERN_RE is defined twice, issuing a warning (see http://www.ruby-doc.org/stdlib/libdoc/rexml/rdoc/classes/REXML/DTD/ElementDecl.html)

(3)
#+++++++
require 'rexml/dtd/dtd'
data =<<EOF
<!ELEMENT foo (bar,baz)>
<!ELEMENT bar (#PCDATA)>
<!ELEMENT baz (#PCDATA)>
EOF

p REXML::DTD::Parser.parse(data)
# => /opt/local/lib/ruby/1.8/rexml/dtd/elementdecl.rb:8: warning: already initialized constant PATTERN_RE
# => /opt/local/lib/ruby/1.8/rexml/dtd/dtd.rb:25:in `parse_helper': undefined method `PATTERN_RE' for REXML::DTD::ElementDecl:Class (NoMethodError)
# => from /opt/local/lib/ruby/1.8/rexml/dtd/dtd.rb:14:in `parse'
#+++++++

-- Daniel