Ruby 1.9.3 リファレンスマニュアル > ライブラリ一覧 > rexml/parsers/sax2parserライブラリ

library rexml/parsers/sax2parser

要約

SAX2 と同等の API を持つストリーム式の XML パーサ。

コールバックをパーサオブジェクトに REXML::Parsers::SAX2Parser#listen で 設定してから REXML::Parsers::SAX2Parser#parse を呼び出すことで、 パーサからコールバックが呼び出されます。

コールバックには2種類あって、ブロックを使う方式と REXML::SAX2Listener を include したクラスのオブジェクトを使う方式があります。詳しくは REXML::Parsers::SAX2Parser#listen を参照してください。

REXML::Parsers::StreamParser のパーサよりは高機能です。

require 'rexml/parsers/sax2parser'
require 'rexml/sax2listener'

parser = REXML::Parsers::SAX2Parser.new(<<XML)
<root n="0">
  <a n="1">111</a>
  <b n="2">222</b>
  <a n="3">333</a>
</root>
XML

elements = []
parser.listen(:start_element){|uri, localname, qname, attrs|
  elements << [qname, attrs]
}
as = []
parser.listen(:start_element, ["a"]){|uri, localname, qname, attrs|
  as << [qname, attrs]
}
texts = []
parser.listen(:characters, ["a"]){|c| texts << c }
parser.parse
elements # => [["root", {"n"=>"0"}], ["a", {"n"=>"1"}], ["b", {"n"=>"2"}], ["a", {"n"=>"3"}]]
as # => [["a", {"n"=>"1"}], ["a", {"n"=>"3"}]]
texts # => ["111", "333"]

仕様確認サンプル

require 'rexml/parsers/sax2parser'
require 'rexml/sax2listener'

xml = <<EOS
<?xml version="1.0" encoding="UTF-8" ?>
<?xml-stylesheet type="text/css" href="style.css"?>
<!DOCTYPE root SYSTEM "foo" [
  <!ELEMENT root (a+)>
  <!ELEMENT a>
  <!ENTITY bar "barbarbarbar">
  <!ATTLIST a att CDATA #REQUIRED xyz CDATA "foobar">
  <!NOTATION foobar SYSTEM "http://example.org/foobar.dtd">
  <!ENTITY % HTMLsymbol PUBLIC
     "-//W3C//ENTITIES Symbols for XHTML//EN"
     "xhtml-symbol.ent">
  %HTMLsymbol;
]>
<root xmlns="http://example.org/default"
      xmlns:foo="http://example.org/foo"
      xmlns:bar="http://example.org/bar"><![CDATA[cdata is here]]>
  <a foo:att='1' bar:att='2' att='&lt;'>
  <bar:b />
  </a>
  &amp;&amp; <!-- comment here--> &bar;
</root>
EOS

class Listener
  #include REXML::SAX2Listener
  def method_missing(name, *args)
    p [name, *args]
  end
  def respond_to_missing?(name, include_private)
    name != :call
  end
end

parser = REXML::Parsers::SAX2Parser.new(xml)
parser.listen(Listener.new)
parser.parse
# >> [:start_document]
# >> [:xmldecl, "1.0", "UTF-8", nil]
# >> [:progress, 39]
# >> [:characters, "\n"]
# >> [:progress, 91]
# >> [:processing_instruction, "xml-stylesheet", " type=\"text/css\" href=\"style.css\""]
# >> [:progress, 91]
# >> [:characters, "\n"]
# >> [:progress, 144]
# >> [:doctype, "root", "SYSTEM", "foo", nil]
# >> [:progress, 144]
# >> [:elementdecl, "<!ELEMENT root (a+)"]
# >> [:progress, 144]
# >> [:elementdecl, "<!ELEMENT a"]
# >> [:progress, 159]
# >> [:entitydecl, "bar", "barbarbarbar"]
# >> [:progress, 190]
# >> [:attlistdecl, "a", {"att"=>nil, "xyz"=>"foobar"}, " \n  <!ATTLIST a att CDATA #REQUIRED xyz CDATA \"foobar\">"]
# >> [:progress, 245]
# >> [:notationdecl, "foobar", "SYSTEM", nil, "http://example.org/foobar.dtd"]
# >> [:progress, 683]
# >> [:entitydecl, "HTMLsymbol", "PUBLIC", "-//W3C//ENTITIES Symbols for XHTML//EN", "xhtml-symbol.ent", "%"]
# >> [:progress, 683]
# >> [:progress, 683]
# >> [:progress, 683]
# >> [:characters, "\n"]
# >> [:progress, 683]
# >> [:start_prefix_mapping, nil, "http://example.org/default"]
# >> [:start_prefix_mapping, "foo", "http://example.org/foo"]
# >> [:start_prefix_mapping, "bar", "http://example.org/bar"]
# >> [:start_element, "http://example.org/default", "root", "root", {"xmlns"=>"http://example.org/default", "xmlns:foo"=>"http://example.org/foo", "xmlns:bar"=>"http://example.org/bar"}]
# >> [:progress, 683]
# >> [:cdata, "cdata is here"]
# >> [:progress, 683]
# >> [:characters, "\n  "]
# >> [:progress, 683]
# >> [:start_element, "http://example.org/default", "a", "a", {"foo:att"=>"1", "bar:att"=>"2", "att"=>"&lt;"}]
# >> [:progress, 683]
# >> [:characters, "\n  "]
# >> [:progress, 683]
# >> [:start_element, "http://example.org/bar", "b", "bar:b", {}]
# >> [:progress, 683]
# >> [:end_element, "http://example.org/bar", "b", "bar:b"]
# >> [:progress, 683]
# >> [:characters, "\n  "]
# >> [:progress, 683]
# >> [:end_element, "http://example.org/default", "a", "a"]
# >> [:progress, 683]
# >> [:characters, "\n  &amp;&amp; "]
# >> [:progress, 683]
# >> [:comment, " comment here"]
# >> [:progress, 683]
# >> [:characters, " barbarbarbar\n"]
# >> [:progress, 683]
# >> [:end_element, "http://example.org/default", "root", "root"]
# >> [:end_prefix_mapping, nil]
# >> [:end_prefix_mapping, "foo"]
# >> [:end_prefix_mapping, "bar"]
# >> [:progress, 683]
# >> [:characters, "\n"]
# >> [:progress, 683]
# >> [:end_document]

クラス

REXML::Parsers::SAX2Parser

SAX2 と同等の API を持つストリーム式の XML パーサクラス。