Re: catching broken links on website and documentation

[Jan Nieuwenhuizen]

Trevor Daniels writes:

Hi Trevor,

> I am puzzled by some of the entries.  For example, in the output for today 
> (5 May) there is the entry:
> 
> 
>    30 /doc/v2.11/Documentation/user/lilypond/Piano-templates
>         
> http://lilypond.org/doc/v2.11/Documentation/user/lilypond/Piano-sections

Yes, that's odd.  Here's one of the log entries it was created from

193.253.100.25 - - [04/May/2008:10:14:23 +0200] "GET 
/doc/v2.11/Documentation/user/lilypond/Piano-templates HTTP/1.0" 404 316 
"http://lilypond.org/doc/v2.11/Documentation/user/lilypond/Piano-sections"; 
"Mozilla/5.0 (X11; U; Linux i686; en; rv:1.8.1.12) Gecko/20080208 
(Debian-1.8.1.12-4) Epiphany/2.22" 193.253.100.25.313961209882632232

> a few lines from the top.  I'm puzzled because the templates were moved from 
> the Notation Reference (the old user manual) to the Learning Manual some 
> months ago, and
> 
>    /doc/v2.11/Documentation/user/lilypond/Piano-templates
> 
> no longer appears on the lilypond.org website.  Do you perhaps have some old 
> pages left on the machine on which this cron job runs?

Indeed, this is weird.  The script runs on lilypond.org where it reads
the apache log.  We (where is plan9 after 20 freakin' years?) can't go
back to may 4 to check if the file was still there.  

Either we have cruft on lilypond.org, or epiphany has a caching feature
where it will display pages from its cache that have been removed
on the server but to the user look like live pages.  Such pages
possibly have broken links to other removed pages.

I'm not sure if there's much we can do about the latter case, other than
ignore them in the report.

Greetings,
Jan.

-- 
Jan Nieuwenhuizen <address@hidden> | GNU LilyPond - The music
typesetter
http://www.xs4all.nl/~jantien       | http://www.lilypond.org