Amrita and XML

summary

amrita can be used with XML as well as HTML.

amrita has REXML-based XML parser: amrita's HTML parser was developed very ad hoc way. I have found no major problem with it yet , but I think it's good idea to have an option for XML: a strictly defined standard.
amrita can generate XHTML documents as well as HTML.: amrita produces xhtml output from xhtml template.
amrita can treat XML documents as a template.: amrita has a little code depend upon HTML's DTD(tag structure), and they can be detached easily. So any XML(not XHTML) document can be used as a template with amrita.
amrita can get model data from XML documents: REXML has very natural API for ruby. And it is easy to make model datas for amrita from XML documents read by REXML.
amx: yet anothor style-sheet for XML: see docs/Tour2

XHTML Document

code and output

code:

require "amrita/template"
include Amrita

tmpl_text = <<-END
<?xml version="1.0"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
    <title>xhtml sample</title>
</head>
<body>
  <h1 id=title>title</h1>
  <p id=body>body text</p>
  <hr />
</body>
</html>
   END

data = {
  :title => "SAMPLE1",
  :body => "members of this HASH will be inserted here and title"
}

tmpl = TemplateText.new(tmpl_text)
tmpl.prettyprint = true
# tmpl.xml = true # use REXML parser
tmpl.asxml = true
tmpl.expand(STDOUT, data)

output:

<?xml version="1.0"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
  <head>
    <title>xhtml sample</title>
  </head>
  <body>
    <h1>SAMPLE1</h1>
    <p>members of this HASH will be inserted here and title</p>
    <hr />
  </body>
</html>

Basicaly amrita produce xhtml output from xhtml template, html4.0 from html4.0 template.So programers don't need tobother about comformation for some paticular standards or browser or devices. Only designers (template writers) do.

The only thing programers should do is set asxml flag to Template object.If this was set a single tag like <hr> will be printed like <hr />.

XML template

code and output

code:

data = {
  :table1=>[
    { :name=>"Ruby In A Nutshell", :author=>"Yukihiro Matsumoto, David L. Reynolds", :isbn=>"0596002149" }
    { :name=>"Programming Ruby", :author=>"David Thomas, Andrew Hunt", :isbn=>"0201710897" },
    { :name=>"The Ruby Way", :author=>"Hal Fulton", :isbn=>"0672320835" },
  ]
}

xml_tmpl = TemplateText.new <<END
<booklist>
  <book id="table1">
    <title id="name" />
    <author id="author" />
    <isbn id="isbn" />
  </book>
</booklist>
END

xml_tmpl.xml = true # use REXML-based parser
puts "------------XML output ----------------------"
xml_tmpl.expand(STDOUT, data)

output:

<booklist>
  <book>
    <title>Ruby In A Nutshell</title>
    <author>Yukihiro Matsumoto, David L. Reynolds</author>
    <isbn>0596002149</isbn>
  </book><book>
    <title>Programming Ruby</title>
    <author>David Thomas, Andrew Hunt</author>
    ..........

description

xml_tmpl.xml = true # use REXML-based parser
puts "------------XML output ----------------------"
xml_tmpl.expand(STDOUT, data)

amrita loads templates on demand. If +xml+ flag is set when +expand+ is called, amrita uses REXML based parser.

You can use single model data for two templates. So single code with amrita produce both XML output and HTML output.For detail see sample/tour/xml1.rb .

Use XML document as a model data

XML document(data) + HTML template ==> HTML document

This may be some kind of style-sheet.

code and output

code:

require "amrita/template"
require "rexml/document"
include Amrita

doc = REXML::Document.new <<END
<booklist>
  <book isbn="0596002149">
    <title>Ruby In A Nutshell</title>
    <author>Yukihiro Matsumoto</author>
    <author>David L. Reynolds</author>
  </book>
  <book isbn="0201710897">
    <title>Programming Ruby</title>
    <author>David Thomas</author>
    <author>Andrew Hunt</author>
  </book>
  <book isbn="0672320835">
    <title>The Ruby Way</title>
    <author>Hal Fulton</author>
  </book>
</booklist>
END

table = doc.elements.to_a("booklist/book").collect do |book|
  { 
    :title=>book.elements['title'],
    :authors=>book.elements.to_a('author').collect do |a|
      { :name=>a }
    end,
    #:isbn=>book.attributes['isbn']
    :isbn=>e(:a, :href=>"http://www.amazon.com/exec/obidos/ASIN/#{book.attributes['isbn']}") {
      book.attributes['isbn']
    }
  }
end

data = { :table1=>table }

html_tmpl = TemplateText.new <<END
<table border="1">
  <tr><th>title</th><th>author</th><th>ISBN</th></tr>
  <tr id=table1>
    <td id="title">
    <td><span id="authors"><span id="name"></span><br></span>
    <td id="isbn">
  </tr>
</table>
END

html_tmpl.prettyprint = true
#html_tmpl.set_hint(HtmlCompiler::AnyData.new)
html_tmpl.expand(STDOUT, data)

output:

<table border="1">
  <tr>
  <th>title</th>
  <th>author</th>
  <th>ISBN</th>
  </tr>
  <tr>
  <td>Ruby In A Nutshell</td>
  <td>Yukihiro Matsumoto<br>David L. Reynolds
  <br>
  </td>
  <td><a href="http://www.amazon.com/exec/obidos/ASIN/0596002149">0596002149</a></td>
  </tr>
  <tr>
  <td>Programming Ruby</td>
  <td>David Thomas<br>Andrew Hunt
  <br>
  </td>
  <td><a href="http://www.amazon.com/exec/obidos/ASIN/0201710897">0201710897</a></td>
  </tr>
  <tr>
  <td>The Ruby Way</td>
  <td>Hal Fulton<br>
  </td>
  <td><a href="http://www.amazon.com/exec/obidos/ASIN/0672320835">0672320835</a></td>
  </tr>
</table>

description

table = doc.elements.to_a("booklist/book").collect do |book|
  {
    :xxx=>.....
  }
end

This code generate an Array of Hash on <book> element. <book> element is a REXML::Element data. So you can get any node or attribute you need by REXML's API.

...
  { 
    :title=>book.elements['title'],
...

book.elements['title'] is the first <title> element of <book> element.

...
  :authors=>book.elements.to_a('author').collect do |a|
    { :name=>a }
  end,
...

In this sample, a book has one title but can have many authors. So authors shuold be treated as an Array with +to_a+, generate Array of Hash by Ruby's standard method +collect+.

...
  :isbn=>e(:a, :href=>"http://www.amazon.com/exec/obidos/ASIN/#{book.attributes['isbn']}") {
    book.attributes['isbn']
  }
...

I want to insert a direct link to amazon. e(...) { ... } generate a <a>tag like

<a href="http://www.amazon.com/exec/obidos/ASIN/0596002149">0596002149</a>

and insert it into template.

This idea was extended for document processing in sample/tour/xml3.rb and reached to amx. See docs/Tour2 for detaile.

convert a Ruby object to a XML entry

When you map a ruby object to XML entry, some members are mapped to attribute and others are to sub elements.

This sample shows how to do it by generate ls -l information as an XML.

code and output

code:

# amrita can use an existing class for model data.
# To show this ability, this sample uses Ruby's system class
# File::Stat.

require "amrita/template"
include Amrita

class File
  class Stat
    include Amrita::ExpandByMember

    def entry(name)
      a(:name=>name, :type=>ftype) { self }
    end

    def size_or_nil
      size if ftype == "file"
    end

    def mode_str
      ret = "-rwxrwxrwx"
      /(.*)(.)(.)(.)(.)(.)(.)(.)(.)(.)$/ =~ sprintf("%b",mode) 
      $~[2..10].each_with_index do |b,n|
        ret[n+1] = "-" if b != "1" 
      end
      ret[0] = "d" if $~[1] == "100000"
      ret 
    end

    def unix_inf
      a(:dev=>dev, :uid=>uid, :gid=>gid) { self }
    end

  end
end

tmpl = TemplateText.new <<END
  <file id="filelist">
    <size id="size_or_nil" />
    <mode id="mode_str" />
    <times>
      <ctime id="mtime" />
      <mtime id="mtime" />
      <atime id="atime" />
    </times>
    <unix_inf id="unix_inf">
      <inode id="ino" />
    </unix_inf>
  </file>
END

dir = ARGV.shift || '*'

filelist = Dir[dir].collect do |f|
  File::stat(f).entry(f)
end

data = { :filelist=>filelist }
tmpl.xml = true
tmpl.expand(STDOUT, data)

output:

<file name="CVS" type="directory">
  <mode>drwxr-xr-x</mode>
  <times>
    <ctime>Tue Sep 03 11:07:10 JST 2002</ctime>
    <mtime>Tue Sep 03 11:07:10 JST 2002</mtime>
    <atime>Thu Sep 05 07:30:39 JST 2002</atime>
  </times>
  <unix_inf uid="1000" gid="1000" dev="770">
    <inode>652250</inode>
  </unix_inf>
</file>
<file name="precompile.rb" type="file">
  <size>2596</size>
  <mode>-rw-r--r--</mode>
  <times>
    <ctime>Mon Aug 26 09:12:11 JST 2002</ctime>
    <mtime>Mon Aug 26 09:12:11 JST 2002</mtime>
    <atime>Thu Sep 05 09:26:48 JST 2002</atime>
  </times>
  <unix_inf uid="1000" gid="1000" dev="770">
    <inode>310411</inode>
  </unix_inf>
</file>
<file name="amstest.ams" type="file">
.....

description

def entry(name)
  a(:name=>name, :type=>ftype) { self }
end

This method generate an AttrArray that put some value of self to XML attribute and make sub-elements with itself.

def size_or_nil
  size if ftype == "file"
end

If the file is not a normal file, this method returns nil and The <size> element will be deleted.

Contents

Documentation

Community

Amrita and XML

summary

XHTML Document

code and output

description

XML template

code and output

description

Use XML document as a model data

code and output

description

convert a Ruby object to a XML entry

code and output

description