Re: spacing problem

[Han-Wen Nienhuys]

On Tue, Aug 12, 2008 at 5:26 PM, Werner LEMBERG <address@hidden> wrote:

>> You're looking for whatever goes into the 2nd argument of the Spring
>> constructor.
>
> Hmm.  In file note-spacing.cc, function Note_Spacing::get_spacing,
> which is used in Spacing_Spanner::musical_column_spacing, I see this
> code:
>
>  Real distance = skys[LEFT].distance (skys[RIGHT]);
>  Real min_dist = max (0.0, distance);
>  ...
>  Spring ret (max (0.0, ideal), min_dist);
>  ...
>  return ret;
>
> It seems to me that `distance' can even become zero.  I can't say
> whether this can cause problems due to lack of knowledge of the source
> code.  On the other hand, musical_column_spacing contains this code in
> case the above one isn't called:
>
>  spring = Spring (max (base_note_space, options->increment_),
>                   options->increment_);
>
> This looks better, however it means that *all* notes, regardless of
> its duration, have a minimum width of `spacing-increment' (which is
> the Scheme parameter assotiated with `increment_'); the default value
> is the width of a note head.  Can this be correct?  Isn't this too

Yep.

> small in tight situations?  Note heads might (almost) touch...

In tight situations, the rods take over the spacing.  I think there is
a small threshold that makes sure they don't touch.

> IMHO, this should be configurable.

You could look into the Note_spacing class, which could decide to add
some more fixed width if the left side is a dotted note, or perhaps
scale up the fixed width for notes longer than 1/8th. (If you do
anything, please make sure it does the right thing for polyphonic
situations too.)

I am not sure if you should use durations rather than dottedness.
Should a dotted 64th need a lot of space to accomodate for the dot, or
should it be treated like a very short note?

-- 
Han-Wen Nienhuys - address@hidden - http://www.xs4all.nl/~hanwen