Re: tablature.ly

[Marc Hohl]

Neil Puttock schrieb:
> 2009/4/25 Marc Hohl <address@hidden>:
>   
> > Hello tablature users*,
> >     
>
> Like Carl, I'm not a tablature user, so I can only comment on matters of coding.
>
> Some suggestions and thoughts follow below:
>
>   
> > % some publications use the triangled note head
> > % for palm mute, so here we go:
> > palmMuteOn = { \set shapeNoteStyles = #'#(do do do do do do do ) }
> >     
>
> You can use make-vector here, since every entry is the same symbol:
>
> palmMuteOn = { \set shapeNoteStyles = #(make-vector 7 'do)
>
>   
> > palmMuteOff = { \unset shapeNoteStyles }
> > % for single notes (or groups of notes within { ...} :
> > palmMute =  #(define-music-function (parser location note) (ly:music?)
> >      #{
> >         \set shapeNoteStyles = #'#(do do do do do do do )
> >         $note
> >         \unset shapeNoteStyles
> >      #})
> >     
>
> You've just defined palmMuteOn/Off.  Why not use them here?
>
> palmMute =  #(define-music-function (parser location note) (ly:music?)
>       #{
>          \palmMuteOn
>          $note
>          \palmMuteOff
>       #})
>
>   
> > % for single notes or groups of notes within {...}:
> > deadNotes = #(define-music-function (parser location notes) (ly:music?)
> >   #{
> >      \override NoteHead #'style = #'cross
> >      \set tablatureFormat = #x-tab-format
> >      $notes
> >      \unset tablatureFormat
> >      \revert NoteHead #'style
> >   #})
> >     
>
> As above, using deadNotesOn/Off.
>
>   
> > #(define-markup-command (customTabClef layout props tuning) (pair?)
> >    (let* ((num-strings (length tuning))
> >           ;; the number of strings has to be in 4...7
> >           (num-strings (cond ((< num-strings 4) 4)
> >                              ((> num-strings 7) 7)
> >                              (else num-strings)))
> >     
>
> This looks a bit strange, since you've just defined num-strings.  You
> can use min and max to keep it within range:
>
> (let* ((num-strings (min (max (length tuning) 4) 7))
>   
Thank you for your suggestions! As I am no expert in scheme, some things can
surely be done more elegant. I corrected my file as you suggested above.

>   
> >           (raise-value (- (* num-strings 0.4) 0.9))
> >     
>
> You can junk this, since #:vcenter in the markup below will do it automatically.
>
>   
Ok.
> >           (font-size (- (* num-strings 1.5) 7))
> >           (base-skip (cond ((= 4 num-strings) 1.55)
> >                            ((= 5 num-strings) 1.84)
> >                            ((= 6 num-strings)  2.00)
> >                            ((= 7 num-strings) 2.08)))
> >     
>
> Can you rework these so they're not hard-coded?
>   
That's a bigger problem: first, I used a definition as shown in the LSR 
and played a bit with the
values for base-skip, font-size and raise. Here is my first attempt:

tabClefIV = \markup {
 \raise #0.7 {
   \override #'(font-family . sans)
   \bold\fontsize #-1.0
   \override #'(baseline-skip . 1.44)
   \column { "T" "A" "B" }
 }
}

I found out that the base-line for 4 ... 7 strings follows a quadratic 
equation:
base-skip = ( 0.2 * num-strings + 0.4 )**2
but inserting this into the definition of customTabClef didn't work, and 
neither
Carl nor I could nail down the problem, so finally, I hard-coded the values.

If you can help me with that, it would be great.
> Imagine a user doesn't like the default staff-space setting for
> TabStaff (1.5).  If they change it, none of these empirical values
> will work properly.
>
>   
> >           (new-props (cons (list
> >                              '(font-family . sans)
> >                              (cons 'baseline-skip base-skip))
> >                             props)))
> >       (interpret-markup layout new-props
> >         (markup #:raise raise-value #:bold #:fontsize font-size
> >                #:column ("T" "A" "B")))))
> >     
>
> Do you want the TAB column left-aligned rather than centred (like the
> LSR snippet)?
>   

Ok, #:column changed to #:center-column, I simply have overlooked this.
>   
> > % Wrappers for the different clefs
> > % the argument is the string-tuning, which is automatically set.
> > sansSerifTabClef = #(define-music-function (parser location tuning) (pair?)
> >   #{
> >      \override TabStaff.Clef #'stencil = #ly:text-interface::print
> >     
>
> Use grob-interpret-markup instead of ly:text-interface::print:
>
> \override TabStaff.Clef #'stencil = $(lambda (grob)
> (grob-interpret-markup grob (make-customTabClef-markup tuning)))
>
> The docs are slightly lagging behind here, but
> ly:text-interface::print should only be used with objects which
> support text-interface.
>
>   
> >      \override TabStaff.Clef #'text = \markup \customTabClef $tuning
> >     
>
> Can be removed if using grob-interpret-markup.
>
>   
> > calligraphicTabClef = #(define-music-function (parser location tuning)
> > (pair?)
> >   #{
> >      \revert TabStaff.Clef #'stencil
> >      \set TabStaff.stringTunings = $tuning
> >   #})
> >     
>
>   
Ok, changed.
> On a general note, I'd prefer to keep the string tunings separate from
> setting the clef.  To set the traditional tab clef, we have the
> command \clef tab, so it would be nice to be able to set the modern
> style using e.g. \clef moderntab without having to use a new function.
>   
Yes, that would be better, but I didn't manage to get the information about
the number of strings internally, so I used the tuning as an argument.
Then, to avoid setting it twice, I wrote the wrappers.
On the other hand, I would have to set the tuning before setting \clef 
moderntab,
but this could be easily mentioned in the docs.

As said above, any help for the base-skip problem and finding a way to 
get the number of strings
without explicitly using the tuning as an argument are appreciated.

Thank you!

Marc
> Regards,
> Neil
>
>