On Monday 04 August 2003 18:45, Niklas Bergh wrote: > > On Monday 04 Aug 2003 17:09, Ian Clarke wrote: > > > I suspect the ratio between HTML and non-HTML data will be more-or-less > > > constant irrespective of the total amount of WWW information - the > > > point still remains that much time will be wasted trying to compress > > > uncompressable data unless the client looks at the MIME-types to > > > determine suitability for compression. > > > > I completely agree. This is precisely why I suggested detecting mime > > types and > > skipping compression on files that are already compressed. > > Could you explain what you mean by 'detecting' mimetypes? How do you > propose that to be done?
I am talking about implementing this in fproxy. When you insert files using fproxy, you also tell it what the mime type is. Additionally, mime type gets passed by the browser when a file is uploaded, so this can be used, too. Then, we can check the file's extension to try to determine the mime type. Finally, mime type can often be determined by reading the first few bytes of the file to try to recognize the header. As you can see, there are ways, and most of them quite easy. Gordan _______________________________________________ devl mailing list [EMAIL PROTECTED] http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl
