-----BEGIN PGP SIGNED MESSAGE-----
It looks like we can use the Symbol font for Greek letters, and
we shouldn't need to substitute this for the fourth font. I can probably
implement this (unless someone else already knows how), but may need to
learn a little more postscript first. Yech...
I don't really know about the bounding box issues, but will give it
On Mon, 6 Dec 2004, Alan W. Irwin wrote:
> On 2004-12-06 09:41-0400 Thomas J. Duck wrote:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> > Hi Alan, Rafael,
> > I've got it! Please change the reference point code in proc_str to:
> > if (args->base == 2) /* not supported by plplot */
> > offset = ENLARGE * ft_ht / 2.; /* half font height */
> > else if (args->base == 1)
> > offset = 0.;
> > else
> > offset = -ENLARGE * ft_ht / 2.;
> > args->y += offset*cos(theta*PI/180.);
> > args->x -= offset*sin(theta*PI/180.);
> > Note that the offset calculation (line 2 and 6) and application
> > (lines 8 and 9) are both changed. This fixes up the problems with x03c.
> Thanks for finding the solution Tom, and committing it, Rafael.
> Clearly, example 3 it is not a perfect example for this work since the
> labels are not perfectly centred. The 0 label is closest and the other
> labels are further away from centre depending on quadrant and length of the
> number. Nevertheless, with Tom's fix both the Hershey and postscript fonts
> show similar label spacing patterns, and that is the best you can expect for
> different sized fonts. Also, I extensively tested example 3 for a good
> sample of non-integral values of -ori from 1 to 4, and the label spacing
> pattern continues to be consistent. Thus, I agree with Tom that the
> spacing issues have been solved for example 3. Nice work!
> However, I just noticed that does leave one minor boundary box issue still
> outstanding with example 3, and also one showstopper issue (lack of Greek
> letters) for scientific publication.
> * Boundary boxes incorrect for postscript fonts.
> (Note they are incorrect for Hershey fonts for non-integral -ori so I am not
> referring to that known issue.) To show the problem for integral -ori,
> prepare test0.ps and test1.ps with
> software@...> ./x03c -dev psc -o test0.ps -drvopt text=0 -ori 0
> software@...> ./x03c -dev psc -o test1.ps -drvopt text=1 -ori 0
> and display results with gv. test0.ps has a valid boundary box, but
> test1.ps does not. Here is what identify reveals:
> software@...> identify test?.ps
> test0.ps PS 444x375 444x375+0+0 DirectClass 488kb 0.000u 0:01
> test1.ps PS 457x390 457x390+0+0 DirectClass 522kb 0.000u 0:01
> So some adjustment of the boundary box is going on when postscript
> fonts are being used, but that adjustment is not correct according to gv
> results. You can also see similar results for -ori 1, etc.
> Tom, if you can figure out a quick fix for this minor issue, that would be
> great, but if not, we can live with it. We are quite pleased with everything
> you have done so far!
> * Inaccessible Greek letters (which is a showstopper issue for scientific
> I just noticed for example 3 that the Greek letters (e.g., #gh) just seem to
> revert to the Roman equivalent ("h" in this case) for postscript fonts. I
> am virtually positive that Greek letters are available in postscript fonts,
> but perhaps we should pick something different than ZapfChancery for the
> fourth font? That is just a guess because I am almost totally ignorant of
> what really happens when you specify Greek letters via the #g escape
> sequence in text. If that choice of fourth font is irrelevant to Greek
> letters, then perhaps we just need to change our access methods to make
> Greek letters accessible for postscript fonts?
> Andrew (Roach) and Rafael are much more familiar with PLplot font topics
> than I am. Do you guys have any ideas for making the greek postscript
> symbols available to PLplot?
> Alan W. Irwin
> email: irwin@...
> phone: 250-727-2902
> 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
> Linux-powered Science
Thomas J. Duck <tomduck@...>
Department of Physics and Atmospheric Science, Dalhousie University,
Halifax, Nova Scotia, Canada, B3H 3J5.
Tel: (902)494-1456 | Fax: (902)494-5191 | Lab: (902)494-3813
Public key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x17D965DB
Tier I CRC Chair in Atmospheric Science: http://www.atm.dal.ca/jobs/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)
-----END PGP SIGNATURE-----