On Fri, Apr 09, 1999 at 10:16:46AM +0100, Tom Newton wrote: > FWIW, Maxwell (that I currently maintain) finds the necessary metrics by > decoding the pfa file. It seems to be "good enough" - Maxwell's printed > output is pretty good (although we don't do kerning), and it might be worth > considering as a fallback option if an afm file cannot be found. > > The code exists in Maxwell and would be relatively easy to extract. Maybe > you consider it to be unecessary bloat, though? Actually, I consider that a good feature to have, but I never really got around to looking at how to decode them from the font files. I also have a bug assigned to me about diacritic characters and metrics, where Adobe's ParseAFM code just returns 0 for character widths for anything with a charcode above 127 for our fonts. What little information the Adobe Type 1 specification documents tell me is along the lines of "if the width is specified to be 0, then the character is not in the font." Does Maxwell deal with high-side Latin 1 character metrics gracefully? Is it just a matter of getting better fonts? :) -- Shaw Terwilliger
