[Bug-gnupod] Podcast character data truncated and ' not escaped in XML generation

[H. Langos]

Hi Adrian,

Podcast titles that contain escaped xml characters (like &) often get
truncated at that character ... The reason is that the parser doesn't
have to return all the character data at once.

See: http://perl-xml.sourceforge.net/faq/#char_events

So a podcast containing <title>The new Q&amp;A Show</title> can result in
"The new Q"

I guess the same error occurs when you read the GNUtunesDB.xml to create
the itunes.db

Also I noticed that in XMLhelper::xescaped you seem to have forgotten to
escape the ' character (single quote or apostrophe).

The later bug probably rarely caused problems because the first bug
already truncated cdata at that point.

If you need a podcast to test it on, I'd recommend this one:

http://www.thenakedscientists.com/naked_scientists_podcast.xml

It contains XML entities in the titles and lots of apostrophs in the
descriptions. (thanx David for pointing me there)

Cheers
-henrik