On 8 August 2015 at 00:01, masti <mastigm@gmail.com> wrote:
(dp0
Vhttp://www.european-athletics.org/european-athletics-awards-night/baldini-says-goodbye-at-the-giro-al-sas.html
p1
(...)

This is indeed pickle format 1, representing a python dict that looks like:

{u'http://www.european-athletics.org/european-athletics-awards-night/baldini-says-goodbye-at-the-giro-al-sas.html':
  [(u'Stefano Baldini', 1404150662.121833, '404 Not Found'),
   (u'2010 w lekkoatletyce', 1404569856.34401, '404 Not Found')]

}

which is the same as the format I get when I run weblinkchecker now. 

But it tries to report deadlinks the were previously reported. So this makes me think the script does not recognize properly old records.

Based on the code, I don't think there is a check for that. The code only checks if the talk page already contains the URL, and this is the same for both compat and core.

Merlijn