<hbot> Just appeared in GrokLaw - http://www.groklaw.net : <hbot> SCO's 12/19 Letter to Lehman Compared with Generic 12/19 "Dear Unix Licensee" Letter entities aren't getting unescaped.
The raw XML for that item is: <item> <title>SCO's 12/19 Letter to Lehman Compared with Generic 12/19 &quot;Dear Unix Licensee&quot; Letter</title> <link>http://www.groklaw.net/article.php?story=20040217233342200</link> <description>Here is the December 19 letter to Lehman Brothers, compared with the generic December 19th "Dear UNIX Licensee" letter, thanks to the creativity of Groklaw member Hykin. The blue is new material in the Lehman letter, and the red text is edited out of the UNIX Licensee letter and does not appear in the Lehman letter. It's helpful in several ways, one of them being that it shows what the missing words likely are in the Lehman Brothers letter. It also shows some files have been dropped from the list. So... there wasn't just one form letter sent to everyone, after all.</description> </item> Just pulled that to make sure it wasn't getting double-quoted in their output. It's not.
Er, scuse me, I'm not reading that very well am I. Looks like a bug in GeekLog.
Yeah we're using a real XML parser now, based on expat, so it seems likely that this will be doing it right. ;-)