Some bugs in 1.5.4x

[Mats Bengtsson]

Here's a list of bugs that I've noticed in the latest 
revisions, including 1.5.46:

- 1. For some scores, I get tons of warnings and errors. Some examples:

programming error: Adding reverse spring! (Continuing; cross thumbs)
programming error: Insane spring found. Setting to unit spring. (Continuing; 
cross thumbs)
programming error: Skipped something ?! (Continuing; cross thumbs)
programming error: No last visible stem (Continuing; cross thumbs)
programming error: Beam dir not set. (Continuing; cross thumbs)

  I think all of these can be seen when processing the regression 
  test, so I don't have to send any specific examples.

- 2. Sometimes, beams collide with note heads. Similar problems
  also appear for tremolo flags. Examples:
\score{
  \notes\relative c'{
  \times 2/3{[d16 fis' d,]} \times 2/3{[cis g'' cis,,]}
  a'16 cis a, g''' % Used to give a nice beam directed upwards.
  <{\stemUp a,,4:32} \context Voice=lower{\stemDown d,:32}>
  }
}
  In these situations, the stem length should be increased 
  automatically. By the way, I didn't compose these examples
  myself, they are all taken from music I typeset.

- 3. Automatic beaming doesn't work well togeter with grace notes:

fis8 e16 d \grace g8 fis e16 d 

  Strangely enough it seems to work sometimes (when the grace notes
  start a new bar?). The following example beams the grace notes
  instead of the main notes:

  c16 b a g \grace g8 fis16 e \grace fis8 e16 d

- 4. I find that the distance between the grace notes and the 
  main notes often is too large to look well.

- 5. When you have a key signature change at the end of the 
  last bar of a repeat, the new key signature overwrites the
  repeat dots (at least, I've seen this when it occured at a
  line break).

Sorry, I know I shouldn't mention too many problems in the 
same email. 

  /Mats