have opted for the continuation bit, and have committed the viewer code,
PyPlucker code and DBFormat.html that uses this.
Everything seems to work, except:
- Font changes make the place of bookmarks and history wrong sometimes when
one had been viewing a fragment-boundary location. Basically, going to the
bookmark/history can end up going to the top of the second fragment rather
than the exact displayed point (so one no longer sees the bottom of the
previous fragment). This is going to be at most one screen off, and only
happens when the font has changed. Keeping track of the location across
font changes was hard to do, and I guess a bit more work is needed. But
it's not a big deal, I suspect. I'll fix it when I find an elegant way to
do it. Anyway, it's not a surprise that font changes should throw things
off (try a font size change in IE!)
- There is a report of my fragment-sorting PyPlucker failing on a site under
python 2.2. Don't know if it's a problem with my code specifically or with
running PyPlucker under python 2.2 or both, but the old code worked under
python 2.2, so probably it has something to do with my code. I am working
on this. This is not in the continuation-marker code, but in the
ordered-assignment of ids code that I had committed before this report.
Minor issues:
- There are some unnecessary screen redraws. (This is just an
aesthetic/speed issue.)
- Autoscroll up should be slower than before when one is far down a
fragment. (But who uses it?)
- Search in page should search a whole series of fragments, but still only
searches the current fragment. I should fix this.
- Going to a particular place in the scrollbar or using the percentage go to
can be off because on fragments that haven't been pre-rendered, I estimate
the pixel height using the character size. This is going to be off if there
is graphics. I think in practice this is going to work fine as long as the
fragments are more or less uniform. For a text-only page that I was testing
on, I was always within 1%. This is not really fixable if one wants speedy
response. Anyway, the estimates improve the more has been rendered, so one
only really has errors in practice when one moves to unread areas.
- 42 million pixel vertical maximum. I can raise this to 2^31 (a factor 50
higher), but it's fairly low priority.
Alex
--
Dr. Alexander R. Pruss
Department of Philosophy
Georgetown University
Washington, DC 20057-1133 U.S.A.
e-mail: [EMAIL PROTECTED]
online papers and home page: www.georgetown.edu/faculty/ap85
----------------------------------------------------------------------------
-
"Philosophiam discimus non ut tantum sciamus, sed ut boni efficiamur."
- Paul of Worczyn (1424)
_______________________________________________
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev
