On 2004-12-19 13:20+1000 Andrew Roach wrote:
> I think the old topic "Re: [Plplot-devel] Freetype "bug" in Debian packages"
> is starting to get a bit stale, so I have changed it.
> At 10:07 AM 18/12/2004 -0800, Alan W. Irwin wrote:
>> So Andrew (Roach), let the other developers know if you need any help
>> getting the ball rolling on the programming of the centralized
>> infrastructure for core translation of Hershey indices to unicode indices
>> following the design that Rafael proposed (just sticking with #g and
>> access methods for now). Or if you don't have time over Christmas to do
>> programming, you might want to make your Hershey to unicode data file
>> available for download so some other developer can take a first crack at
>> necessary programming.
> I should, from this week, *finally* have some time to do some work on this. I
> have a "conceptional model" as to how I am going to go about it, and the raw
> table for conversion is, I think as advanced as I can make it with my eyes
> (although, as I have said before, many glyphs just look like squiggles to me
> !). I made the table by using a unicode browser program and a printed out
> copy of the Hershef font tables, and looked for matches. There are 949
> Hershey fonts by my count, and 760 of them I have been able to match up to a
> unicode glyph. I know that might sound like a lot are missing, but I think
> most of the important ones have been matched, and it is mainly the more
> abstract ones left over. The process is not infallible, so I am sure that
> many of those blanks could be filled up with another set of eyes, and I
> probably also have some "wrong" (one has to make some arbitary assumptions
> here and there). I made the table using OpenOffice's spreadsheet program, and
> if anyone wants to have a go at adding the missing glphs they are welcome to
> have the table, just let me know what format works best for you. Eventually I
> am going to have a plain ASCII three field table (hershey, unicode, hershey
> font face number), but while it is still being edited it is a little easier
> to maintain as a spreadsheet. Alternately, if anyone wants a list of the
> missing glpys only, I'd could send that across, or even post it here to the
> developers list.
My advice is not to worry about refining these index transformation data for
now. Instead, I suggest you simply plunge ahead with the minimal part of
programming that gets just #(nnnn) access to work with your current data as
we have already discussed, check in the results, and then the rest of us can
join in and help you.
For example, as soon as that #(nnnn) programming is done, I can also
reprogramme plsym to call plmtex with the appropriate #(nnnn) code. This
simple change will make all of the example 7 results be responsive to font
changes, not just the outer number display as now. Thus, simply running
example 7 alternatively with Hershey font and FreeSans font for -dev png
should give a good comparison between the two fonts for everything available
from Hershey and quickly turn up any remaining index transformation data
difficulties. Furthermore, since your index transformation data will be part
of a PLplot source code and therefore under cvs control, that gives
automatic and extremely useful coordination of updates of those data for our
developers who do sit down to compare Hershey and non-Hershey results for
In sum, everybody now deeply interested in the PLplot font revolution can
easily help out in a variety of ways as soon as #(nnnn) access minimally
works for true-type fonts.
Alan W. Irwin
Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).
Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project