[Fwd: [Fwd: FT2 design question]]

[Stefan Seefeld]

Here is the response from a friend working on berlin whome I forwarded
your mail. I hope this will clear up at least what we are aiming at a bit:


Graydon Hoare wrote:

> sequence of normalized unicode characters
> 
> "wysiwyg is difficult"
> 
> breaks into sequence of minimally nonconflicting chunks
> 
> w:y:s:i:w:y:g: :i:s: :d:i:ffi:c:u:l:t
> 
> each computes its width from drawing kit's font metrics in order to
> facilitate flowing of the text in display-device space. the flowing takes
> place _only_ when editing. once the line is edited, it does not get
> re-flowed (i.e. chunks are not transferred between lines) when printed.
> 
> w :y:s:i :w :y:g:   :i :s:   :d:i :ffi:c:u:l :t
> .7:1:1:.5:.7:1:1:1.8:.5:1:1.8:1:.5:3  :1:1:.3:.8
> 
> as a part of the fresco layout system, each chunk was given stretchyness
> in layout terms, by applying the unicode test for a "spacing" character.
> 
> w:y:s:i:w:y:g: :i:s: :d:i:ffi:c:u:l:t
> R:R:R:R:R:R:R:F:R:R:F:R:R:R  :R:R:R:R
> (R = rigid, F = flexible)
> 
> when user goes to print, print kit does _not_ reflow the text, but when it
> measures the chunks in the printer's font it gets different values:
> w:y:s:i:w:y:g: :i:s: :d:i:ffi:c:u:l:t
> 1:1:1:1:1:1:1:1:1:1:1:1:1:1  :1:1:1:1
> 
> oh no! it's monospace! no big deal. when the layout algorithm runs, it
> will stretch the individual space chunks to compensate. It does this
> because the chunks are in an aligned HBox. It works like any other. It'll
> look a _little_ weird, but postscript (as a print backend) lets you do
> this, and you wind up with the exact same set of words occupying the exact
> same set of lines on the exact right size layout box -- just more space
> shows up here and there. Look at a magazine with justified text someday --
> they all do this.
> 
> likewise, in the case that the display space said there was more width
> than the printer (which is highly unlikely, if you look at how hints work,
> but anyway) you can just compress the spaces between words. in a
> pathological case, you'll get words starting to overlap, but you'd need
> such terribly mismatched fonts for that to be a problem that this is
> probably the least of your problems.
> 
> likewise, the inter-line spacing is set up in screen space, and translated
> (with re-layout) to printer space. since you are actually _asking_ the
> font to measure the text in each space, you get exact information about
> what sort of "hidden machinery" the font will apply to your chunks. i.e.,
> it can apply any machinery it likes, so long as the size it reports is
> indeed the size of the thing it returns.

Regards,        Stefan
_______________________________________________________              
              
Stefan Seefeld
Departement de Physique
Universite de Montreal
email: address@hidden

_______________________________________________________

      ...ich hab' noch einen Koffer in Berlin...