Stefan Haller wrote:
> 
> M-Tx handles melismas automatically based on the beams and slurs; this
> is great, but not flexible enough (unless I'm missing something) in
> cases where you have to override this default handling (and I often
> have).
> 
> Consider the following example:
> 
> ================================
> Style: Singer
> Meter: C
> 
> %% w2i
> [ c8 d1 e1 ] [ d8 e1 f1 ] e2 |
> L: A-men
> ================================
> 
> Now of course I want the first syllable of `Amen' to be a melisma of 6
> notes, and the second syllable should be on the final half note.
> 
> The first thing I tried to achieve this is putting a slur around the
> first six notes (I don't really want this though -- I often need
> melismas *without* slurs).  This doesn't work at all; M-Tx complains
> about:
> 
>    Melisma starts with slur but ends with beam: ERROR on line 5
> 
> no matter in what order I put the ( ) and [ ].  This looks like a bug to
> me.
> 

Yes, the error message is a bug.  :-/  It goes onto my list.  
The correct PMX code is generated, though.

>
> Putting in \beginmel\ and \endmel\ as inline TeX doesn't work either,
> because they interfere with the \endmel commands that M-Tx puts in
> itself.
>
>
> What I end up doing now is this:
>
> ================================
> Style: Singer
> Meter: C
>
> %% w2i
> [ c8 d1 e1 ] [ \nolyr\ d8 e1 f1 ] e2 |
> L: A-men
> ================================
>
> This works quite well in this simple example, but it is a pain when
> there are very long melismas.  (Defining \let\nl=\nolyr\ helps some, but
> not much.)

You have several possibilities.

--1----------
Style: Singer
Meter: C

%% w2i
( c8 d1 e1 d8 e1 f1 ) e2 |
L: A-men
-------------
This produces all six notes slurred, and grouped as two beamed sets.

--2----------
Style: Singer
Meter: C

%% w2i
m2200 [ c8 d1 e1 d8 e1 f1 ] e2 |
L: A-men
-------------
This produces all six notes unslurred and beamed as one set.
You may want to restore common meter afterwards with m4400.

--3----------
Style: Singer
Meter: C
Options: i

%% w2i
( [ c8 d1 e1 d8 e1 f1 ] ) e2 |
L: A-men
-------------
This produces all six notes slurred, and beamed as one set.  
I don't recommend using the i option as a habit, but because
of the spurious error message it is required here.

--4----------
Style: Singer
Meter: C

%% w2i
[[ \beginmel\ c8 d1 e1 ] [[ d8 e1 \endmel\ f1 ]  e2 |
L: A-men
-------------
This produces the six notes unslurred, beamed as two groups of three,
and belonging to just one melisma.  In general using the [[ (( {{
constructions turns off the M-Tx automatic melisma generator and leaves
everything up to you.  If you have a lot of this I recommend
-------------
[[ MR1 \beginmel\ M c8 d1 e1 ] [[ d8 e1 MR2 \endmel\ M f1 ]  e2 |
-------------
after which 'MP1' equals '\beginmel\' and 'MP2' equals '\endmel\'.

These four cases cover all possibilities I can think of.  The last one
does the same as your solution with \nolyr, but is probably less
of "a pain when there are very long melismas."

> I suggest adding a new pair of commands to M-Tx to begin and end a
> melisma; unfortunately I can't think of any characters that could be
> used for this, but maybe someone else can.  If { } were not already
> taken, I would suggest to use them like this:
> 
> { [ c8 d1 e1 ] [ d8 e1 f1 ] } e2 |
> 
> with proper nesting of any beams or slurs that might occur inside such a
> group.

I'm a bit reluctant to invent a new notation for something that is not too
cumbersome.  Instead, I propose adding the above solutions to the 'Annotated
Examples' section of the manual.

Dirk Laurie

Reply via email to