On Tue, Jul 14, 2009 at 04:13:12PM +0900, Maximilian Albert wrote: > 2009/7/14 Patrick McCarty <pnor...@gmail.com>: > > > Unfortunately, this would be very difficult. Elements are dumped into > > the SVG file in the order they occur in the page stencil, and (almost) > > every one is independently positioned as well. > > Ah, okay. That's what I though. Out of interest: At the time when > these elements get written into the SVG file, do they know about their > mutual relationships? E.g., does a beam know which note heads it > belongs to (or vice versa)?
No. The closest thing the elements possess that relates to this is their *grob cause*. For example the NoteHead grob is often the grob cause for the "noteheads.s2" glyph. > > Some of the related stencils are dumped consecutively (e.g. the > > horizontal StaffSymbol lines), but I imagine this would require some > > postprocess optimization, which is an interesting feature request. > > > > I'll add this to the wiki. > > Following my question above, here is another random idea: Would it be > possible to (perhaps optionally) set the "id" attributes of the > elements to something meaningful from which at least the function of > this element could be deduced, like "staffline01" or "barline42" or > something similar? This could be a first step to enable some > postprocessing. BTW, I'm not saying that you should implement this. > I'm just interested whether it would be possible. That's possible, but it would be much easier to insert comments based on the "grob cause" mentioned above. It would look something like <!-- NoteHead --> <path transform="translate(...) scale(...)" d="..." /> Would that be reasonable? > P.S.: What's all this about text element being rasterized or converted > to paths? I can edit them as regular text elements in Inkscape without > problems. Now that I think about it, I'm not really sure what Mark was referring to. Possibly the low graphics quality at 100% zoom in Firefox. Thanks, Patrick _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
