lilypond-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Cleaned up style (issue 4951062)


From: lemzwerg
Subject: Re: Cleaned up style (issue 4951062)
Date: Fri, 09 Sep 2011 06:46:12 +0000


http://codereview.appspot.com/4951062/diff/1/mf/feta-kievan.mf
File mf/feta-kievan.mf (right):

http://codereview.appspot.com/4951062/diff/1/mf/feta-kievan.mf#newcode78
mf/feta-kievan.mf:78: fill z1{dir -6.9} .. z2 .. z3 & z3 .. z4 .. z5 &
z5 -- z6 & z6 .. z7 .. z8 & z8{left} .. z9 & z9 .. z10 ... {dir
-76.9}cycle;
fill z1{dir -6.9}
      .. z2
      .. z3
      & z3
      .. z4
      .. z5
      & z5
      -- z6
      & z6
      .. z7
      .. z8 & z8{left}
      .. z9 & z9
      .. z10
      ... {dir -76.9}cycle;

Well, I generally prefer vertical alignment since it makes the code much
easier to read.  However, it shouldn't be a religious thing.  For
example, given that the fill command is quite long and clearly separated
into subpaths, the following looks good also:

  fill z1{dir -6.9} .. z2 .. z3
       & z3 .. z4 .. z5
       & z5 -- z6
       & z6 .. z7 .. z8
       & z8{left} .. z9
       & z9 .. z10... {dir -76.9}cycle;

http://codereview.appspot.com/4951062/diff/6001/mf/feta-kievan.mf
File mf/feta-kievan.mf (right):

http://codereview.appspot.com/4951062/diff/6001/mf/feta-kievan.mf#newcode32
mf/feta-kievan.mf:32: x1 = 0.09 * staff_space;
I'm not really happy with the MF code: It looks like a direct
translation of PostScript points into MF points.  This definitely works
but is quite clumsy.  If you look at the design of other LilyPond
glyphs, you can see that all coordinates rely on meta-parameters which
control the appearance.  For your glyphs, we have to directly shift
coordinates in case of a change, with a great chance to inadvertently
break important relationships within the glyph shape.

I would really like to see a `LilyPond approach' for Kievan notes also,
treating the current approach as a temporary solution.

http://codereview.appspot.com/4951062/



reply via email to

[Prev in Thread] Current Thread [Next in Thread]