Andreas L. Delmelle wrote:
Anyway, the spec doesn't seem to provide for a way to indicate that the text
in this case needs to be centered across a line (--or am I missing something
again?)
Indeed. There is some "common knowledge" that excess space in justified
text should be distributed evenly over resizable spaces.
How 'easy' would it be to change the default expansion for leaders in the
case:
...
Why _exactly_ does it use a default width here?
Because its easy. And: conformant :-)
Is it possible to test at
that point whether in the current line an fo:leader has already been added?
Well, I could have tried to distribute excess space evenly. However, how
should the space be distributed in case a leader would exceed its maximun
length? Should it be maxed out and then desitribution of the residue space
continued with the rest of the eladers? Certainly an idea, but it requires
more extended housekeeping than I was willing to implement. You'll have
to keep an eye on leader alingment at the same time, which complicates
the whole stuff (that's why leader alignment will go out of the window
if there's stull excess space after all leaders have been maxed out and is
distributed over resizable spaces). You may even run into convergence
problems.
And if so, is it possible to adjust both leaders to expand accordingly?
It depends. In the case presented, certainly. But a general algorithm
able to deal with arbitrary combinations of leader-length MinOpMax and
alingments is by no means easy.
J.Pietschmann
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
