Re: Weird errors - again

[Mats Bengtsson]

> GNU LilyPond 1.3.103.
> Now processing: `star-spangled-banner.ly'
> Parsing...
> Interpreting music...
> warning: no one to print a repeat brace
> 
> star-spangled-banner.ly:106:7: warning: Nothing to connect extender to on the 
> left.  Ignoring extender request.:
>   Oh __
>         \repeat "fold" 2 { }
> [8][16]
> star-spangled-banner.ly:126:68: warning: Nothing to connect extender to on 
> the left.  Ignoring extender request.:
>       Oh say, does that star -- span -- gled ban -- ner yet wave, __
>                                                                      
> 
> star-spangled-banner.ly:127:22: warning: Nothing to connect extender to on 
> the left.  Ignoring extender request.:
>       O'er the land __
>                        of the free and the home of the brave.

We all get these warnings.

> [24][27]
> Preprocessing elements... 
> Calculating column positions... kpathsea: Running mktextfm  feta-nummer8
> mktextfm: Running mf \mode:=ljfour; mag:=1; nonstopmode; input feta-nummer8
> This is METAFONT, Version 2.7182 (Web2C 7.3.1)
> kpathsea: Running mktexmf  feta-nummer8
> 
> ! I can't find file `feta-nummer8'.
> <*> ...ur; mag:=1; nonstopmode; input feta-nummer8

Your teTeX installation does not find the Lilypond fonts.
Is the environment MFINPUTS set? If so, does it include a
path to the Lilypond mf/ directory? (I think the RPM:s
install shell initialization files /etc/profile.d/lilypond-*
that set this variable, see also the INSTALL.txt file). 
However, setting MFINPUTS this way may cause the problem
that feta*.tfm and feta*.*pk files are generated in every
directory where you run ly2dvi and an alternative route is
described in the FAQ on pk and tfm files, see 
http://appel.lilypond.org/wiki/index.php3?LilyPondFaqs
If you already used the latter solution and the link is
still pointing at the correct directory, it may be enough
to run 'mktexlsr' as root to solve the problem.

(This font setup is a bit messy but the problem is that
every teTeX installation looks a bit different so it is
virtually impossible to make a generic solution that 
fits everybody.)

      /Mats