On Monday 02 April 2007 21:47, Patrick R. Michaud wrote: > Here's how I currently see the above: > > link=Cat1 # link=+Cat1 > link=+Cat1,Cat2 # link=+Cat1,+Cat2 > link=-Cat1,Cat2 # link=-Cat1,+Cat2 > link=-Cat1,Cat2,+Cat3 # link=-Cat1,+Cat2,+Cat3 > > In other words, any values without + or - are part of a required > "or grouping" -- i.e., at least one of the specified values must be > present. (If there's only one, it must be present.)
So "link=-Cat1,Cat2,+Cat3,Cat4" means "Cat1 forbidden, Cat3 required, (Cat2 and/or Cat4) required". Yes, this is more user-friendly. Okay, this is really good, didn't think of that. I like this! > > > > For me what is absolutely essential is the speed of a pagelist, > > > > so I prefer to only use .pageindex and not open every file for > > > > inspection unless impossible (so: no PageTextVar search for now). > > ...and this is why the expanded link= parameter is taking a bit of time > to design and implement, because I think to be done correctly we need to > preserve the speeds provided by pageindex, as well as provide an > unambigous interpretation for link=cat1,cat2 (while also providing for > the other interpretation). I completely agree. And in any case, whatever means link=cat1,cat2 it will be better than what it does now. Thanks a lot! Petko _______________________________________________ pmwiki-users mailing list [email protected] http://www.pmichaud.com/mailman/listinfo/pmwiki-users
