Re: Spacer Rests Global Confusion

[Br. Samuel Springuel]

On 4/26/18 12:02 PM, R wrote:
> This shows the default duration as a sixteenth note which should be
> NOT even considered since it's within an uncalled variable. So why is
> this spacer value magically applied to music when it resides it a
> completely uncalled variable? This makes no sense to me. Even this
> snippet makes things worse.

On 4/26/18 1:34 PM, David Kastrup wrote:
> You are laboring under the wrong impression that there is a "default 
> duration" of a quarter note.  There isn't.  The_initial_  default 
> duration is that of a quarter note.  Whenever a duration is entered
> as a note or chord or s or syllable length, it becomes the new
> default duration.

If it helps, I tend to avoid thinking of it as "default" altogether.
Rather the parser understands the concept of "last duration encountered"
and that is initialized to be 4.  "Default" to me always implies "fall
back" position which could come into play under various circumstances
which I may (or may not) fully understand.  "Initialization" only
applies to the very beginning of the file.  After that I either have to
trace the parser's actions line by line (and maybe even character by 
character) or all bets are off.


On 4/26/18 1:04 PM, R wrote:
> I know but I never called it with a backslash was my point. It's 
> uncalled. I was trying to figure out what contents of an uncalled 
> variable leak out into impacting later code in the input file regardless 
> of it never being called ever. Thanks.

The way I think about this is to distinguish between the parser and the 
"interpreter" (which is basically everything else).  The parser reads 
the input from start to finish in order in order to create a "code" 
representation of it (a set of instructions).  It has to go in order and 
cannot jump around, so "last encountered..." necessarily refers to what 
comes first "physically" in the input file (including includes).  Once 
my input is translated into the "code" representation, the "interpreter" 
can follow the instructions therein to render the score.  Because the 
complete instruction set is available at this point, the "interpreter" 
is not bound by linearity and can jump around based on variable 
references and the like.

Of course, I have no idea if this is how it actually works.  It's just 
the working model I have in my head that thus far has proven productive 
when trying to grasp why LilyPond works the way it does.

--
✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝
Br. Samuel, OSB
St. Anselm’s Abbey
Washington, DC
(R. Padraic Springuel)

PAX ☧ ΧΡΙΣΤΟΣ