A link-checking checking tool called inspect.exe is already in existence
https://www.boost.org/doc/libs/1_70_0/tools/inspect/
and in use in an attempt to name'n'shame authors, but sadly not every library
author acts on it to update or remove links (a tiresome task).
Authors do not seem to use it when developing their documentation but building
the tool inspect.exe (jamfile provided so painless) and running it in the folder
containing their /html to create a local check.
(This tool also makes some other checks for other common mistakes (most common
is failure to prevent MAX and MIN macros being used when calling
numeric_limits<>::max() and min()).
But thanks for re-raising this issue.
Paul
Paul A. Bristow
Prizet Farmhouse
Kendal, Cumbria
LA8 8AB UK
-----Original Message-----
From: Boost On Behalf Of Tinko Bartels via Boost
Sent: 20 April 2019 03:08
To: boost@lists.boost.org
Cc: Tinko Bartels
Subject: Re: [boost] [doc][geometry][gil][signals] Broken links for Boost 1.70
Hi,
I also noticed these broken links in the boost documentation for geometry.
It made me wonder, whether there are more dead links in the boost documentation,
so I tried to check that with a JS script. I only managed to get parts of the
website running locally, and I didn't scan older versions than 1_70_0
recursively, so my results are definitely incomplete, but I did discover more
broken links:
The following documents list pages and below those pages is an unordered list of
hyperlinks on those pages that seem to be broken.
internal links that are probably broken:
http://46.101.197.7/internal.html
external links that are probably broken:
http://46.101.197.7/external.html
I'm not claiming that these lists are complete or without false positives but
all links on it, that I checked manually, seem to be dead. Maybe the lists still
help to see some broken links that might have gone unnoticed so far.
Since link rot is natural, I wonder whether it would make sense to implement
some check for this in the CI for the website so that breakages like this one
could be discovered automatically in the future. I'd volunteer to implement that
later this year but only if the general idea is considered sensible or viable by
the boost website maintainers.
Best regards,
Tinko Bartels
--
Sent from: http://boost.2283326.n4.nabble.com/Boost-Dev-f2600599.html
_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
