Glenn McCorkle wrote:
>>Presently Arachne seems to select colo(u)rs at random from a selection
>>with the likelyhood of 2 items the same colour. It looks odd.
>>actually 2 colours the same 80% of the time.
>
>
> That shouldn't be too hard to do.
>
> I'll let you know when it's done. ;-)
I like my current colour schene (the blue-green one), so please keep it.
(Do whatever with the 2nd colour scheme, or any any other schemes of
your choice ;-)
> if(sizeof(struct HTTPrecord)==len)
> _____________________
>
> --- the changes I made ---
> //!!glennmcc: begin Apr 10, 2002
> //'postflag' is now 'dynamic' and moved to before 'lastseen'
> char dynamic; // document is dynamic
> // char postflag; // it is result of operation post
> long lastseen; // last seen time
> //!!glennmcc: end
>
> //!!glennmcc: begin Apr 10, 2002
> //must subtract 2 from 'len' to get the compatibility test to work
> if(sizeof(struct HTTPrecord)==len-2)
...and preferably sizeof(int), for future compatibility with other
platforms than DOS ;-)
> And just in case you hadn't guessed.....
>
> I LOVE doing this stuff. ;-)
Which is reason why open source development is so succesful for Linux
apps.... (well at least relatively....)
I like coding too. I hate packing installation packages and testing them...
> And why do we need to subtract 2 for the test to work ????
> (that does not match the length as written by core.exe)
sizeof(int). It is the boring binary file format
{int XX=sizeof(struct YY), struct YY}
> 'HTTPrecord / cacheitem' contains 11 "fields".
>
> Core.exe writes them into cache.idx 1,2,3,4,5,6,7,8,9,10,11
> Wwwman.exe now reads them from cache.idx 1,2,3,4,5,6,7,8,9,11,10
>
> Wwwman.exe now looks for a length of 2 bytes less than core.exe
> used so that it can to test for 'version compatibility'.
>
> Logic says that it can't work.... but it DOES work (perfectly).
>
> Figure that one out. <g g g g>
???
xCh.
