[tex-live] ovp2ovf with VPL input

[Lars Hellström]

Karl Berry skrev:
> Hi Lars,
> 
>     So, is this program just plain broken or what?
> 
> I don't know.  Perhaps the documentation is wrong when it says it works
> with the original VPL files, and it only works with the Omega VPL's.
> I've never looked into the actual differences between original VPL and OVP.

At least for the v1.x series of ovp2ovf (written in Web), it's just an 
extension with new properties of the basic PL format. It's possible 
that they did some radical language modification for v2.0 (written in C 
and Yacc), but it seems unlikely.

>     Moreover, the manpage refers to omega at ens.fr and 
>     http://www.gutenberg.eu.org/omega/, which are probably not helpful. 
> 
> There is no web site any more that I know of, but there is a mailing
> list omega at tug.org, which Yannis et al. are on.  Could you write there
> and ask for advice about the situation?

Done; so far neither a bounce nor any appearance in the omega archive 
(wow, that list hasn't seen a lot of traffic in its five year existence).

>  And/or look up the ovp2ovf
> sources and see if anything jumps out?

Given that the first error reported concerns parsing (of the very first 
line!!), and that the parser is in Yacc, I feel out of my depth.

FWIW, I did manage to compile ovp2ovf v1.11 late last night and got it 
to do the VPL->VF+TFM conversion, so using ovp2ovf for the "Big base 
font" project seems viable. Chronologically it is possible that the 
manpage (dated 2000) describes the v1.x series (last changed 2000), but 
I had to make a CHange to enable generation of VF+TFM output, so that 
hypothesis isn't entirely consistent either.

> Sorry I can't do more right now.

Well, right now I'm mainly worried by the prospect that I might come up 
with a great solution in a month or two, which pretty much nobody would 
be able to use for about a year because a key utility is buggy in 
TeXLive 2009. :-)

Lars