[Rd] Re: Hershey Fonts, PNG, SVG, .. graphics formats

Paul Murrell paul@stat.auckland.ac.nz
Fri, 6 Oct 2000 13:58:42 +1300


> I wonder if we couldn't just use the PNG , SVG, support from the
>   GNU plotutils
> from which we already have the Hershey scalable vector fonts [->
> The new plotutils (July 2000) do support these (and will probably even
> better in the future, i.e., we can build on others people free software.
> Paul [who built "Hershey" into R], do you think this path is worth

What I wanted to do with the Hershey vector fonts was JUST grab all of the
vector font database information and all of the stuff that converted a
character string into a set of line-drawing commands (using the vector font
database).  This proved possible for the vector fonts, but that was because
I was able to find a fairly clear interface between the vector font code and
the device-handling code.  This meant I was able replace calls to the GNU
plotutils devices with calls to R graphics devices without (I think) too
much horribleness.

I was deliberately not trying to figure out how the GNU plotutils device
code worked, but the vague impression I obtained was that it was quite
complicated and differed quite a bit from R's concept of a device.

Now, without having looked at the PNG/SVG code, my guess would be that it
might be much more tightly bound to GNU's device-handling code - i.e.,
probably harder to only extract the PNG/SVG bits we want.

On a more positive note ... in general, the GNU plotutils code (at least the
bits I saw) impressed me.  It seemed well-documented, well-organised, and
VERY thorough.  Mad though it may sound, perhaps we might want to adopt ALL
of their device-handling code rather than trying to just get the PNG/SVG
bits (maybe one way to make our postscript woes disappear :)

I _suspect_ that a significant amount of time and effort would be involved
in trying to understand the PNG/SVG code (or all of the device-handling
code).  However, I also suspect that it would be worth the effort because
from what I've seen there's a lot of really good stuff in there.


r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch