XPath finds bogusetag in case of unescaped double-quotes

[scrubber]

This snippet:

require 'rubygems'
require 'hpricot'
require 'open-uri'

doc = Hpricot(open('http://www.funmobile.com/catalog/ringtone/polyringtones/catId-00201340'))
records = doc/'/html/body/div/div/table/tr/td/table/tr/td[2]/a[1]'
records.map[14..16].each {|e| p e}

Shows that though the above XPath should find 20 records, it finds just 19. The missing one (the 15th in the records array) should be matched by the XPath, too (in Firefox it is, for example) but it is not - it is returned as a bogusetag.

Is this a serious problem? Just because a project is depending on this, and it would be super great to know what is the estimate for fixing it. Thanks a lot!

[lwu]

Looks like the root cause seems to be bad HTML on their end:

   title="Cell Phone Polyphonic Ringtones - Polyphonic Ringtones - Themes (Movies, TV) Polyphonic Ringtones - Going The Distance (From "Rocky") - Theme"

I haven't looked much at Hpricot's error handling code, but you could probably work around this ("Rocky"!) temporarily by escaping those " characters with " yourself, before sending the doc to Hpricot.