Re: AFM font handling corrected?

[Rune Zedeler]

Han-Wen Nienhuys wrote:

> IIRC the 1000 in an AFM is relative to either the design size or the
> baselineskip of the font. Isn't an em 10pt in cmr10? 

I tried to (dirtily) hack afm.cc to multiply the sizes by 10 if the font 
is CMR10.
It does not give exactly the same results with and without the afm-file.
Unfortunately I made a test file which produces bad results no matter if 
the cmr afm-files are there or not. The result is worse with the 
afm-files, but it is also wrong without them :-(

\score {
    \notes <
        \context Staff \notes { c d e f g a b c' c d e f }
	\context Lyrics \lyrics { this is a loooooooooooooooooooooooooong test 
aaaa }
    >
}


the long word overlaps the next word.
Without the afm-files (i.e. on a tetex-system) it is approx "ng" that 
overlaps, where as with the afm-files, "oong" overlaps.
Can anybody else reproduce this bug?

The bug is also there in 1.6, so 1.8 is not in danger :-)

The font-metrics code really needs an intense look.

-Rune