From csr21 at cantab.net Sun Aug 19 20:42:02 2007 From: csr21 at cantab.net (Christophe Rhodes) Date: Sun, 19 Aug 2007 21:42:02 +0100 Subject: [gsharp-devel] my current development directions and issues Message-ID: <87y7g72r51.fsf@cantab.net> Hi, I thought it might interest people who read this list but don't perpetually idle on IRC to know some of what I'm doing at the moment. I'm looking, based partly on a late-night conversation in a bar in Berlin, at implementing support for Mensural (black) notation, with a view to allowing editing and typesetting of chant (as in the Liber Usualis and similar collections). I'm also simultaneously attempting to learn how to use the git distributed version control system, which is why you can see a mirror of my development tree (and even play along, should you so wish) at . The neumes and ligatures which are currently implemented can be seen at ; at present, there's no support for stems, accidentals; I've drawn glyphs for the F-clef and the diamond punctum, but there's no UI for them yet. (There are other slightly broken bits of the UI and display code, and it's as yet completely uncoupled to the MIDI output, and and...). I hope the lurking mediaeval music experts will pitch in! Bugs that I've found or remembered in the course of this development include: * Issues with non-6 staff sizes: ** beams are drawn wrong at size 8 or 12 (and possibly others) using the CLX backend: an example of what I get is at ; there are two different problems shown there: the vertical misalignment of some of the beam components, and the apparent miscalculation of the horizontal extent ** miscalculation of the position for the first note on a staff; with large sizes (towards the top of the range), the first note is almost completely obscured by a treble clef. ** the space allowed for multiple staves in a system is inadequate with large staff sizes. (Larger staff sizes are potentially important in the context of mensural notation because the neume glyphs are small; in fact, for printing it might be worth making the staff lines thinner so that the neumes are more clearly distinguishable.) * Data structure badness ** M-x Insert Segment After RET followed by M-x Delete Segment gives some kind of data structure corruption. This might be simple to fix. * ESA? issues ** C-h b lists a non-helpful set of commands; I /think/ that the commands it lists are those in the `gsharp' command table, rather than those which are in the currently applicable command table. Cheers, Christophe From campbell at mumble.net Wed Aug 29 04:44:13 2007 From: campbell at mumble.net (Taylor R Campbell) Date: Wed, 29 Aug 2007 04:44:13 +0000 Subject: [gsharp-devel] bugs in rendering tight-knit chords with ties Message-ID: <20070829044413.91A85983B2@pluto.mumble.net> I took these two screenshots this evening: Gsharp appears to be getting confused by chords with ties so close together. I'm also informed that the second screenshot is due to proximity to the end of the line (which is where those clusters were). From strandh at labri.fr Wed Aug 29 15:49:08 2007 From: strandh at labri.fr (Robert Strandh) Date: Wed, 29 Aug 2007 17:49:08 +0200 Subject: [gsharp-devel] bugs in rendering tight-knit chords with ties In-Reply-To: <20070829044413.91A85983B2@pluto.mumble.net> References: <20070829044413.91A85983B2@pluto.mumble.net> Message-ID: <18133.38260.102995.605442@serveur5.labri.fr> Taylor R Campbell writes: > > I commited what I believe to be a fix for this problem. > I haven't started working on this one yet. -- 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. ---------------------------------------------------------------------