Re: Issue 1294 in lilypond: Version 2.13.35 does not properly represent lyric tie

[lilypond]

Comment #5 on issue 1294 by PhilEHolmes: Version 2.13.35 does not properly  
represent lyric tie
http://code.google.com/p/lilypond/issues/detail?id=1294

Don't think it's just windoze.

1) Could you confirm that the NR looks OK for you - see the screen-shot  
from the latest NR page 215 where the snippet is not properly displayed.

2) Original poster was running MAC. Full quote says:

"everything was fine with LyricTies, until I came up with a score (say:
lyrics in this score) when trying the current Lily devel.
What happened? In 2.12.3: all lyric ties have been placed beautifully
where they have to be – starting right at the bottom right corner of the
left letter and ending right at the bottom left corner of the following
letter. On Mac OS X Tiger, Lily 2.12.3 (and before) resp. Pango/FontConfig
has chosen "Helvetica" to borrow this special char from. Great – fits
excellent!

NOW on the same machine – same OS, same hardware (sorry, can't change that
much, it's a MacBook), no re-install ever since Lily 11.x lived here (may
I call this "stable"?) – now Lily 2.13.34 upwards finds "MS-Gothic" to be
tied to lyrically!!! It's not for the font, but for the way Mr U+203F is
invoked: suddenly the glyph is much more narrow, it starts beneath the
vertical center of the letter and ends halfway the wordspace, leaving the
same gap before the following letter, where it was intended to stop.

So WHY do Lily 2.12 and 2.13 on the SAME MACHINE choose different (system)
fonts for the lyric tie? I greped all Lily-containers for "Helvetica" or
"MS-Gothic" to discover a new preference for the new font, but without
success. Is it Pango to make the choice on its own leaving Lily back with
baddest spacing? Please, compare the clips and see, how ugly it's now on
Mac OS X :(( I've found some code to modify the lyric tie spacing, but
that's just a solution for my local machine, not suitable to give away ...

Setup: LP 2.12.3 and 2.13.11–35 on Mac OS X (10.4.11 Intel)"

And anyway - if it is a Windows problem, that's still a bug/regression?