# Relative urls - browser plugin? scraper? other?

**URL:** <https://www.ruby-forum.com/t/relative-urls-browser-plugin-scraper-other/155856>\
**Category:** Rails\
**Created:** [December 17, 2008, 5:07pm UTC](https://www.ruby-forum.com/t/relative-urls-browser-plugin-scraper-other/155856 "2008-12-17T17:07:43Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![jshow](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/977dab/32.png) [@jshow](https://www.ruby-forum.com/u/jshow)\
**Post date:** [December 17, 2008, 5:07pm UTC](https://www.ruby-forum.com/t/relative-urls-browser-plugin-scraper-other/155856/1 "2008-12-17T17:07:43Z")

</div>

I have a very strange situation that I’ve been tracking for some  
months now.

This may or may not be a rails issue. The symptoms don’t feel like it,  
so I guess I’m appealing to my rails colleagues to lend me their  
thoughts.

Throughout the day, [HomeStars.com](http://HomeStars.com) will get a number of invalid  
requests(\< 50) such as :

/companies/1199238-action-assembly/forums - is not valid

/companies/1199238-action-assembly - is valid  
/forums - is valid

/forums is linked to in the menu on each page (so is /sessions/new  
which also exhibits the same behaviour)

All I can surmise is that something is hitting the forums relative url  
thinking it’s relative, but it’s an absolute reference.

[Forum](https://www.ruby-forum.com/forums)

So this feels like either a bug in a scraper, or a page pre-loader  
(plugin that visits links on a page to pre-load them).

I see enough of them, with differing user agents, and vastly different  
IPs, and not enough requests to not feel like a scraper.

So I’m left with thinking it’s a browser plugin - pre-loading.

any thoughts or pointers are appreciated.

thank you.  
Jodi