Re: robots.txt, etc?
Will Guaraldi <[email protected]>
| Newsgroups | gmane.comp.web.pyblosxom.user |
|---|---|
| Message-ID | <[email protected]> |
A while back I was tossing around adding the ability for PyBlosxom to have a "resources" directory where files like the ones you mention could all hang out. I didn't implement it for a couple of reasons: 1) I figured it's probably a slow way for files to get sourced since the request has to go through PyBlosxom and then get served by PyBlosxom. 2) I also wanted this system to make flavour packs easier to modularize. So if a flavour pack required a css file (and they almost all do), then the css file would stay in the same directory as the flavour pack and still get served up by PyBlosxom. I didn't get further than that. I still like the idea, though, and maybe it's something that should get handled in a plugin so that we can more easily handle other behaviors. /will On Thu, 25 Jan 2007, Ryan Barrett wrote: > > hi all. i'm curious how people on this list handle robots.txt, favicon.ico, > and other well-known URLs on pyblosxom sites. > > pyblosxom is a cgi program, not a web server, so technically, we can duck the > issue by claiming that GET /robots.txt should be handled directly by the web > server. however, many pyblosxom-based sites use pyblosxom to serve *all* > requests, including /*, so this isn't really fair. :P > > on my site, i use the filekicker plugin, which serves matching files in the > filesystem when they're requested directly. it only lets pyblosxom render > stories and file lists when they don't match a file on the filesystem. > > with filekicker, i just drop a robots.txt file in the root of my datadir, and > filekicker serves it happily. however, .txt is the default extension for > entries, so when pyblosxom sees a request for /robots, it tries to parse > robots.txt as an entry, promptly dies, and the browser receives a 500. > > should pyblosxom do anything about special, well-known URLs like /robots.txt, > /favicon.ico, and others? i'm inclined to say that if a matching file exists > in the datadir, we should serve it, and we should 404 on requests for /robots. > > thoughts? > > -Ryan ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV