Bear with me a moment, i think i might have *just* found and fixed a small but
important buglet with complex chp properties, we initted the chp from
the chp->istd but didnt init the istd from the pap istd in complex mode,
or we did originally and i fecked it up in my copy.
Also, i have had to modify slightly the initial values of para_fcfirst, coz
if the first para is in the complex appendum section rather than in the
simple section then it will fall through the search for the beginning
of the para, just setting the init to the fc of the first char appears to
do the bizz. Purify is a wonderful tool btw :-), i can't comment on the
veracity of the extracted properties, but give the new ver in cvs a try.
>> 4. Speed. The importer is currently passing along each character and
>> properties as it's read. Buffering runs with the same properties and adding
>> them all at once should speed things up drastically. (The piece table can
>> handle this by coalescing adjacent characters with the same properties, but
>> it should be *much* more efficient if done in the importer.)
>
>Yeah, I was going to do that once I had improved property support. I might
>get to it tomorrow. This week sometime, for sure.
Personally i'd be happier with getting everything stable and secure before
optimization. But still it shouldn't make much difference i suppose.
C.
>
>Justin
Real Life: Caolan McNamara * Doing: MSc in HCI
Work: Caolan.McNamara@ul.ie * Phone: +353-86-8790257
URL: http://www.csn.ul.ie/~caolan * Sig: an oblique strategy
Infintesimal gradations