On Sat, Oct 06, 2007 at 05:37:46PM +0200, Enrico Forestieri wrote: > On Fri, Oct 05, 2007 at 05:23:25PM +0200, Enrico Forestieri wrote: > > > For screen representation of integrals, LyX uses two different fonts > > leading to an inconsistent look. See the attached integrals.lyx. > > > > There's already in place the infrastructure for using the integrals > > in the esint font, and some time ago I built a ttf version of this > > font (I am attching it here). > > > > There's a problem with Qt4, though. As in the case of a soft-hyphen, > > affecting the Omega symbol, Qt4 doesn't print the glyph of a character > > when it thinks that it is white space. This impacts \dotsint (and > > \dotsintop) such that this symbol would not show up by simply > > dropping esint10.ttf in the LyX fonts directory, because its code > > point is 0x09 corresponding to a horizontal tab. Unfortunately, in > > this case we cannot use a workaround similar to that devised for Omega. > > > > I can think of two solutions: > > 1) using the attached patch (esint.diff), or > > 2) moving the dotsint glyph in esint10.ttf to another code point. > > > > Solution 1 is what we use now when the esint font is missing, > > constructing the missing glyph by means of other glyphs. > > Solution 2 would provide a real glyph for \dotsint, but the font > > would not match the TeX font with the same name. > > > > Given that now the math fonts are private to LyX, I would implement > > solution 2, as I don't see any drawback. > > > > Opinions? > > Lacking any opinion on this matter, I opted for solution 2, but > instead of moving the glyph I simply remapped it to an unused > position in order to maintain compatibility with the TeX font.
Jürgen, what about branch? -- Enrico
