From strandh at labri.fr  Wed Mar  1 00:06:21 2006
From: strandh at labri.fr (Robert Strandh)
Date: Wed, 1 Mar 2006 01:06:21 +0100
Subject: [gsharp-devel] rotate-notehead on rests
In-Reply-To: <sqoe0rv7lp.fsf@cam.ac.uk>
References: <sqoe0rv7lp.fsf@cam.ac.uk>
Message-ID: <17412.58749.750742.654667@serveur5.labri.fr>

Christophe Rhodes writes:
 > 
 > People might also be interested to compare a screenshot I took this
 > morning, <http://www-jcsu.jesus.cam.ac.uk/~csr21/cris-gsharp.png>,
 > with the opening of the lilypond score I produced a few years ago,
 > <http://www-jcsu.jesus.cam.ac.uk/~csr21/music/scores/cris.ps>.
 > Entering the notes for this was a true breeze; the lyric entry mode
 > needs some more work (and so does lyrics display, actually), but
 > overall I think it's promising.

It is indeed, especially since the Lilypond version has numerous
rendering flaws, such as stems that are not properly attached to their
corresponding note heads, notes heads that are not positioned the same
way on different staff steps, ugly `bumps' in some ties (despite the
higher resolution), etc.  
-- 
Robert Strandh

---------------------------------------------------------------------
Greenspun's Tenth Rule of Programming: any sufficiently complicated C
or Fortran program contains an ad hoc informally-specified bug-ridden
slow implementation of half of Common Lisp.
---------------------------------------------------------------------


From strandh at labri.fr  Wed Mar  1 00:16:38 2006
From: strandh at labri.fr (Robert Strandh)
Date: Wed, 1 Mar 2006 01:16:38 +0100
Subject: [gsharp-devel] rotate-notehead on rests
In-Reply-To: <sqoe0rv7lp.fsf@cam.ac.uk>
References: <sqoe0rv7lp.fsf@cam.ac.uk>
Message-ID: <17412.59366.564540.803459@serveur5.labri.fr>

Christophe Rhodes writes:
 > 
 > At some point in the reorganization, C-r stopped working on rests.
 > Attached is a patch which gets it working again.  (My tree is
 > otherwise mutant from too many key signatures, so please apply :-)

Patch applied.  Thanks!
-- 
Robert Strandh

---------------------------------------------------------------------
Greenspun's Tenth Rule of Programming: any sufficiently complicated C
or Fortran program contains an ad hoc informally-specified bug-ridden
slow implementation of half of Common Lisp.
---------------------------------------------------------------------


From csr21 at cam.ac.uk  Wed Mar  1 10:16:31 2006
From: csr21 at cam.ac.uk (Christophe Rhodes)
Date: Wed, 01 Mar 2006 10:16:31 +0000
Subject: [gsharp-devel] 8va treble clef
In-Reply-To: <sqbqwrv25x.fsf@cam.ac.uk> (Christophe Rhodes's message of "Tue, 
	28 Feb 2006 17:11:38 +0000")
References: <sqbqwrv25x.fsf@cam.ac.uk>
Message-ID: <sq8xruh3ls.fsf@cam.ac.uk>

Christophe Rhodes <csr21 at cam.ac.uk> writes:

> Attached is a patch implementing a treble8 clef (which is a treble
> clef with implicit octaviation).  I'm not really proposing this for
> direct inclusion; I haven't designed a glyph for it, for a start.

Apart from the glyph problem, which can be finessed for now (because
in at least some versions of music typesetting an octave-treble clef
is set using the same glyph as the normal treble clef, and it is left
to context for the readers to work out that the octave needs to be
adjusted), I'm happy with this (attached) patch.  It introduces three
new functions: F-POSITION, B-POSITION and BOTTOM-LINE, of which the
first two aren't in fact quite general enough but simply replicate the
current functionality.  I would be wary of documenting these
functions, as I suspect that a final clef protocol will look slightly
different.

If these caveats are acceptable, please apply :-)

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: treble8-2.diff
URL: <https://mailman.common-lisp.net/pipermail/gsharp-devel/attachments/20060301/a7291d61/attachment.ksh>
-------------- next part --------------

Cheers,

Christophe

From csr21 at cam.ac.uk  Wed Mar  1 10:42:31 2006
From: csr21 at cam.ac.uk (Christophe Rhodes)
Date: Wed, 01 Mar 2006 10:42:31 +0000
Subject: [gsharp-devel] make tempo a segment attribute
Message-ID: <sq4q2ih2eg.fsf@cam.ac.uk>

Hi,

Les Cris de Paris, as I notate it, is taken at a tempo of
approximately 80 semibreves per minute -- about four times faster than
the default playback setting in gsharp.  So I implemented (attached)
setting the tempo for playback on a per-segment basis, which I think
fits in with the notion of segments as large-scale musical objects.

I haven't implemented a user interface beyond the simple command;
eventually, a tempo marking might be displayed as a metronome mark or
(for the second and subsequent segments) as a tempo relationship.

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: tempo.diff
URL: <https://mailman.common-lisp.net/pipermail/gsharp-devel/attachments/20060301/f30f2841/attachment.ksh>
-------------- next part --------------

Cheers,

Christophe

From azervopoulos at gmail.com  Wed Mar 22 17:34:19 2006
From: azervopoulos at gmail.com (Thanassis Zervopoulos)
Date: Wed, 22 Mar 2006 19:34:19 +0200
Subject: [gsharp-devel] Can I import *.midi or *.mid or a sequence from CM
	in gsharp
Message-ID: <f8c1048a0603220934w2c0f94b4y1f9febedd14582ad@mail.gmail.com>

Hi,
  I've tried to save some notes and I saw that the format wasn't midi or mid
and I have some music recorded with CM (portmidi-record),I'm not sure if I
can pass a seq or even a midi,mid file directly to gsharp

Thanx in advance
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/gsharp-devel/attachments/20060322/2040bfd1/attachment.html>

From strandh at labri.fr  Thu Mar 23 01:28:26 2006
From: strandh at labri.fr (Robert Strandh)
Date: Thu, 23 Mar 2006 02:28:26 +0100
Subject: [gsharp-devel] Can I import *.midi or *.mid or a sequence from CM
	in gsharp
In-Reply-To: <f8c1048a0603220934w2c0f94b4y1f9febedd14582ad@mail.gmail.com>
References: <f8c1048a0603220934w2c0f94b4y1f9febedd14582ad@mail.gmail.com>
Message-ID: <17441.63930.564407.224105@serveur5.labri.fr>

Hello, 

Thanassis Zervopoulos writes:
 >   I've tried to save some notes and I saw that the format wasn't midi or mid
 > and I have some music recorded with CM (portmidi-record),I'm not sure if I
 > can pass a seq or even a midi,mid file directly to gsharp

No you can't.  The simple reason is that MIDI is both more general and
less general than classic music notation.

It is more general in that it encodes performance aspects that are
typically not encoded in the written score (such as short pauses
between notes or, on the contrary, short overlaps between two adjacent
notes).  It is less general in that it cannot distinguish between a
sharp note and a flat version of one note higher and many other
aspects of traditional music notation. 

-- 
Robert Strandh

---------------------------------------------------------------------
Greenspun's Tenth Rule of Programming: any sufficiently complicated C
or Fortran program contains an ad hoc informally-specified bug-ridden
slow implementation of half of Common Lisp.
---------------------------------------------------------------------


From csr21 at cam.ac.uk  Thu Mar 23 09:22:27 2006
From: csr21 at cam.ac.uk (Christophe Rhodes)
Date: Thu, 23 Mar 2006 09:22:27 +0000
Subject: [gsharp-devel] Can I import *.midi or *.mid or a sequence from CM
	in gsharp
In-Reply-To: <f8c1048a0603220934w2c0f94b4y1f9febedd14582ad@mail.gmail.com>
	(Thanassis Zervopoulos's message of "Wed, 22 Mar 2006 19:34:19 +0200")
References: <f8c1048a0603220934w2c0f94b4y1f9febedd14582ad@mail.gmail.com>
Message-ID: <sqlkv1tt0s.fsf@cam.ac.uk>

"Thanassis Zervopoulos" <azervopoulos at gmail.com> writes:

> ? I've tried to save some notes and I saw that the format wasn't
> midi or mid and I have some music recorded with CM
> (portmidi-record),I'm not sure if I can pass a seq or even a
> midi,mid file directly to gsharp

Not without doing some work.  (And automatic conversion is unlikely
ever to be fully satisfactory, for reasons detailed in Robert's
reply).

However, writing (for want of a better term) import filters into
gsharp (and export, though we have export to MIDI already) is clearly
important for all sorts of reasons, even if the results of the filter
need to be touched afterwards.  As it happens, I was discussing just
two days ago about what would need to be done to write an
"interpreter" for OPND files, which are a sequence of (Onset,
PitchName, Duration) triples.  This is one step closer than a MIDI
importer, as the "spelling" of the notes is included in the OPND
format, whereas it isn't in MIDI: a C-sharp and a D-flat have the same
MIDI pitch representation, whereas on a score they are noticeably
different.

Just to get you started, though, here is a toy example: consider a
music notation encoding which can represent one voice of crotchets
(quarter notes) the white notes between a4 and g5: the opening of
Tallis' second lamentation might be represented in this format as the
string "aabcdeea".  (I hope that everyone realises that this format is
not a particularly good one :-).  Then code to
convert/import/interpret this representation within gsharp is:

(in-package :gsharp)

(define-gsharp-command (interpret-string :name t) ((s 'string))
  ;; no error checking
  (dotimes (i (length s))
    (let ((pitch (+ 23 (digit-char-p (char s i) 36))))
      (insert-note pitch (if (upper-case-p (char s i))
                             *current-cluster*
                             (insert-cluster))))))

and indeed this code includes some extra bonus easter egg features!

Generalizing this to OPND files is relatively straightforward;
generating OPND representations from MIDI files involves a certain
amount more research, but there are some algorithms which perform
quite well, so getting pitch names is possible.  I believe that
performing acceptable voicing from arbitrary MIDI files is quite hard.

Cheers,

Christophe