> > +<DT>%t#u
> > +<DD>Date of request, sub-second component with # decimals. 0# will
> > be zero padded. # defaults to 03 for 3 decimals zero padded.
>
> I prefer the generic field width method..
>
> %#tu
>
> Ths should also be considerably easier to parse..
Not really, since my current parser looks for prefixes to decide what
to do. It may very well be a very suboptimal parser, I didn't think
much about design (if I did at all).
>
> > +<DT>%rm
> > +<DD>Request method
> > +<DT>%ru
> > +<DD>Request URI, without the query string
> > +<DT>%rq
> > +<DD>Request URI, only the query string
> > +<DT>%rp
> > +<DD>Request protocol (i.e. HTTP/1.1)
> > <DT>%R
> > -<DD>Request line, including query string
> > +<DD>Request line, complete
>
>
> Missing a "Request URI, including query string". My suggestion is %rU.
Feel free to modify the specs, it can really be a work in progress.
> Should also clarify if %ru includes ? or not. But I guess it does in
> which case %rU can be seen as redundant.
I agree with you. Maybe an opcode that defines the whole query
can be added at a later time, but it's just an optimization.
-- /kinkieReceived on Thu Mar 21 2002 - 01:16:47 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:14:52 MST