Pete Jordan (perlmonger) wrote in lj_dev,
Pete Jordan

  • Mood:
  • Music:

CSS sanitising

I store my stylesheet on my own server - this makes it far less hassle to tweak than having it embedded in the style (and, in theory at least, means it can be cached in browsers).

Since the CSS sanitiser was interposed into the process, I’ve been getting occasional fetch errors on the thing. I’m guessing this is down to LJ (for whatever reason) failing to grab the thing successfully from my server, though there aren’t any errors in my logs.

It looks like LJ fetches (and presumably processes) the remote stylesheet on every request, I can’t think of any other reason for the following sequence: - - [04/Jul/2006:14:16:09 +0100] “GET /css/lj.css HTTP/1.1” 206 6066 “-” “libwww-perl/5.803” - - [04/Jul/2006:14:23:45 +0100] “GET /css/lj.css HTTP/1.1” 206 6066 “-” “libwww-perl/5.803” - - [04/Jul/2006:14:37:12 +0100] “GET /css/lj.css HTTP/1.1” 206 6066 “-” “libwww-perl/5.803” - - [04/Jul/2006:14:42:28 +0100] “GET /css/lj.css HTTP/1.1” 206 6066 “-” “libwww-perl/5.803” - - [04/Jul/2006:14:48:31 +0100] “GET /css/lj.css HTTP/1.1” 206 6066 “-” “libwww-perl/5.803”

I don’t understand why the sanitised version isn’t cached at LJ’s end. Even though traffic on my journal isn’t likely to cause any network or processing issues, that might not be true elsewhere. In any case, if LJ has a problem with fetching, it should at least serve the last processed version not a 404.

I wonder about the 206 status too - the file is exactly 6066 bytes long, so what partial content is LJ asking for?

  • Post a new comment


    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded