From: "Branko Strok" <[EMAIL PROTECTED]> > I believe I discovered another discrepancy in Plucker Find :( (I'm using > viewer 1.7 sent to me by Alex with the latest search fixes) > > Problem: > Some keywords are not discovered by using "sub pages" search (started at > least one depth up from the page that has the keyword). Aternatively same > keywords are dicovered by "in all pages" search and every other search > option. > Interestingly, once the keyword is viewed (we are talking about long > continous pages with terms in HTML unordered lists) in the browser a repeted > "sub" search is succesful. > > Suggestions?
My main suggestion is to post this to bugs.plkr.org so I can get to it eventually. The subpage search has not been tested by me very much. I fixed it so it wouldn't have a memory leak, but that's about all. Oh, here's a thought--in fact, I think this may be it. Subpage search has never been overhauled to deal with pages that consist of more than one 30K fragment. So if the page that the main page links to is one of these long pages, only data in the fragment actually linked to will be found. Obviously this needs to be fixed. I've got some ideas. I should, first of all, start the search at the first fragment of the current page. Second, after a fragment is searched, I should put at the head of the search queue the reference to the next fragment. That should do the job. But I am very busy with other things, so could you just post a bug report, and hopefully include a demo pdb for my eventual ease of testing, and I'll look into it in due time, hopefully by early January. It's a low priority for me, I'm afraid. As a quick workaround, enable the fragment link markers. Alex _______________________________________________ plucker-dev mailing list [EMAIL PROTECTED] http://lists.rubberchicken.org/mailman/listinfo/plucker-dev
