使用saxerator gem（ruby）时，put_attributes_in_hash！ 方法获取标签的内部文本而不是属性

Question

I'm attempting to use the saxerator gem to parse an xml file, and I want to get the attributes for some xml tags. I am using the put_attributes_in_hash! method to accomplish this, and it's working well, except for tags that have inner text. The method ignores attributes for these tags, and instead grabs the inner text and the inner text alone. I want to be sure I can get the attributes for ANY tag, whether it contains text or not.

Here's my xml

<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660734.jpg" 
    ModificationDate="2014-10-03T09:50:14" Primary="true">Welcome home!
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660736.jpg" 
    ModificationDate="2014-10-03T09:50:38">Bathroom 1
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660737.jpg" 
    ModificationDate="2014-10-03T09:50:43">Bedroom 1
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660738.jpg" 
    ModificationDate="2014-10-03T09:50:48">Bedroom 2
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660739.jpg" 
    ModificationDate="2014-10-03T09:50:54">Bedroom 3
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660740.jpg" 
    ModificationDate="2014-10-03T09:50:58">Backyard
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660741.jpg" 
    ModificationDate="2014-10-03T09:51:05">Dining room
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660742.jpg" 
    ModificationDate="2014-10-03T09:51:10">Living room
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660743.jpg" 
    ModificationDate="2014-10-03T09:51:19">Bathroom 2
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/6660744.jpg" 
    ModificationDate="2014-10-03T09:51:24">Kitchen
</PropertyPhoto>
<PropertyPhoto 
    ImageUrl="http://example.com/L600/9987810.png" 
    ModificationDate="2015-03-09T15:06:17">Rental Images-07
</PropertyPhoto>

and my output looks like this...

"PropertyPhoto"=>["Welcome home!", "Bathroom 1", "Bedroom 1", "Bedroom 2", "Bedroom 3", "Backyard", "Dining room", "Living room", "Bathroom 2", "Kitchen", "Rental Images-07"]

and here's an example of it working for tags with no inner text

xml:

<Property PropertyID="1234567" LocalPropertyID="">

output:

"PropertyID"=>"1234567",
  "LocalPropertyID"=>"",

(in a hash)

and I'm invoking put_attributes_in_hash in the proper, functional way:

feed = Saxerator.parser(Net::HTTP.get(@xml_file)) do |config|
   config.put_attributes_in_hash!
end

Please help!

Answer 1

Saxerator author here. I know this is way late, apologies. The documentation for this could use some improvement, as you're not the first person to wonder what happens to attributes on elements that have inner text .

The objects that appear to be strings in your array (eg "Welcome home!" ) are actually StringElement objects, and you can access the attributes on them through the #attributes method.

As of Saxerator 0.9.5 this fact is hidden because the library inherited from the String object, which inherits String#inspect and gives no clue that there's anything else on that object. I plan to release a new version soon that makes this more-clear by relying on Delegate classes and exposing its attributes through inspect .