Reading large XML file with Nokogiri

Question

I'm having trouble reading a (somewhat) large XML file with Nokogiri, but can't figure out where things are going wrong. The file content looks like the following (just a single node included for readability):


End | Globally Sourced Menswear2016-04-02T01:25:30+00:00391WRY3924TYN/AComme des Garcons x Artek Standard Eau De Toilettehttp://www.endclothing.com/eu/comme-des-garcons-x-artek-standard-eau-de-toilette-wry3924ty.html89.00 EURin stockCDG ParfumPerfume & FragrancenewFounded by 4 young idealists in 1935, Finnish design company Artek produce modern furniture to promote the modern culture of habitation. Here they collaborate with Comme des Garçons Parfum to produce a fragrance dubbed 'Standard', ironic for a scent that is anything but.
Notes Include:

Thyme
Black Pepper
Patchouli
Cedar Wood
Citrus

Due to recent changes in regulations, we are unable to ship aftershaves and perfumes to certain destinations outside of the EU. For full details, please email info@endclothing.co.uk or call +44 191 231 3983.
]]>http://media.endclothing.com/media/catalog/product/1/8/18-03-2016_commedesgarcons_xartekstandardeaudetoilette_100ml_sh_1.jpghttp://media.endclothing.com/media/catalog/product/1/8/18-03-2016_commedesgarcons_xartekstandardeaudetoilette_100ml_sh_2.jpgFRDPD Priority Service9.00 EURDEDPD Priority Service9.00 EURDKDPD Priority Service9.00 EURNLDPD Priority Service9.00 EURITDPD Priority Service9.00 EURSEDPD Priority Service9.00 EURBEDPD Priority Service9.00 EURATDPD Priority Service15.00 EURIEParcel Force Priority Service15.00 EURESDPD Priority Service15.00 EURLVDPD Priority Service19.00 EURHRDPD Priority Service35.00 EURCYFEDEX Priority Service45.00 EURHUDPD Priority Service15.00 EURPTDPD Priority Service19.00 EUREEDPD Priority Service25.00 EURLUDPD Priority Service9.00 EURSKDPD Priority Service15.00 EURBGDPD Priority Service25.00 EURGRFEDEX Priority Service25.00 EURPLDPD Priority Service15.00 EURLTDPD Priority Service19.00 EURSIDPD Priority Service15.00 EURFIParcel Force Priority Service19.00 EURCZDPD Priority Service15.00 EURLIFEDEX Priority Service35.00 EURMCDPD Priority Service15.00 EURCHParcel Force Priority Service15.00 EUR

I've tried the following code to read the stream, and while the separate parts seem to work just fine (data outputs a string which appears to be valid XML to me), Nokogiri appears to be unable to read the string, and just crashes or returns nothing for my xpath queries.

url = "http://www.endclothing.com/media/end_feeds/eu.xml.gz"
stream = open(url, 'Accept-Encoding' => 'gzip')
data = Zlib::GzipReader.new(stream).read
page = Nokogiri::XML(data)

page.xpath("//entry")
=> []

har07 · Accepted Answer

The XML has default namespace declared at the root element level :

xmlns="http://www.w3.org/2005/Atom"

In XML, descendant elements without prefix inherits default namespace from ancestor implicitly. That said, entry element that you tried to get is in the root element's default namespace.

On the other side, in XPath, element without prefix always considered in empty namespace. To reference element in XML's default namespace using XPath, we need to map a prefix to the default namespace URI and use that prefix in our XPath, for example :

page.xpath("//d:entry", 'd' => 'http://www.w3.org/2005/Atom')

Reading large XML file with Nokogiri

Answers (1)

Related Questions