[WASTE-list] Re: Zoom factor

[Marco Piovanelli]

On Fri, 20 Oct 2006 16:51:29 -0500,
Dan Korn (dkorn at printable.com) wrote:



>Does anyone have any ideas on how to implement a zoom factor in  

>WASTE, like the EM_SETZOOM message for a Rich Edit control?

>

>I found this old thread:

> http://www.merzwaren.com/waste/listarchive2000/58869.html

>

>But I didn't find it very helpful.

>

>I don't see anything built into WASTE to support a zoom factor.  I'm  

>using a customized version of the WText/CWasteEdit PowerPlant  

>wrappers.  So I'm thinking about two possibilities:

>

>1. Include the zoom factor in my calculations that convert points to  

>Fixed points for point size, leading, etc., when I call  

>WESetAttributes.  This will certainly work, but when the user changes  

>the zoom factor, I'll have to get all the text, apply the new sizes  

>to it, and then set it again, which will certainly be slow.

>

>2. Get the GrafPort into which WASTE is drawing and apply a  

>transformation and clipping to it.  This should be fast, but I'm not  

>sure if it will mess up WASTE's internals.  And it's more work.

>

>Any other ideas?


While so far WASTE 3.0 doesn't have built-in support for a zoom
factor, either, I think it would make both of the approaches you
suggest above easier to implement.  Let me try and explain why.

Option #1.

First of all, WASTE 3.0 uses a floating-point geometry internally,
and stores font sizes, vertical shifts, embedded object rectangles,
etc., in floating-point format, so scaling can be performed with
minimal rounding errors.  Secondly, WASTE 3.0 has a concept of
"display" styles as distinct from "stored" styles.

"Stored styles" represent collections of attributes associated
with certain text ranges.  The text model maintains a single,
shared copy of each unique collection in its style table.
"Display styles" are ATSUStyle objects generated dynamically,
on-demand, when a range of text needs to be laid out or rendered
into a Core Graphics context.  There's a well-defined bottleneck
that creates a display style from a stored style.  If you customize
this bottleneck (by overring a virtual function), you can easily
scale the font size on a per-layout basis, without affecting the
underlying model at all.

Option #2.

WASTE 3.0 draws everything using Core Graphics, and Core Graphics
makes it easy to apply arbitrary affine transforms to a drawing
context, something that would be hard or impossible to do in a
Quickdraw world.

Granted, both approaches would require the source code (at least
until built-in zooming is implemented), but likely less work than
grafting zooming support onto WASTE 2.1.



                                        -- marco

-- 
It's not the data universe only, it's human conversation.
They want to turn it into a one-way flow that they have entirely
monetized. I look at the collective human mind as a kind of
ecosystem. They want to clear cut it. They want to go into the
rainforest of human thought and mow the thing down.