On Sat, May 02, 2020 at 01:16:30AM -0700, beauleetien...@gmail.com wrote: > On 2020/05/02 07:38:58, hanwenn wrote: > > I don't completely understand, though: if we put the "utf-8-string" > directly > > into the SVG output, the SVG browser might make other font choices, > making the > > outline potentially incorrect. Is this a problem in practice?
"utf-8-string" never appeared directly in the svg output and that doesn't change. Typically (without changing options) you would end up with <text font-family="serif"></text>. > There isn't really an "outline" anymore; the bounding box is used. That > is based on the specified font, but if a font were substituted, there > will be slight differences. They wouldn't be too big though and so > wouldn't be a problem. Yes, and to expand on this: - lilypond was *already* falling back on the bounding box for pure utf-8-string stencils - but when a grob mixed utf-8-string stencils with something else (a note for example), it would give the grob a skyline *just* from the note, meaning the text portion of the grob would collide with other grobs - this patch changes the behaviour of stencils that mix utf-8-string and other kinds of stencil to be the same as pure utf-8-string (i.e. fall back to the bounding box). > > https://codereview.appspot.com/582010043/
