octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: fltk printing [PDF problems]


From: Michael D Godfrey
Subject: Re: fltk printing [PDF problems]
Date: Sun, 08 Aug 2010 12:27:10 -0700
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.7) Gecko/20100720 Fedora/3.1.1-1.fc13 Thunderbird/3.1.1

On 08/08/2010 12:16 PM, Ben Abbott wrote:
On Aug 1, 2010, at 3:48 PM, Shai Ayal wrote:

> On Tue, Jul 27, 2010 at 9:21 PM, bpabbott <address@hidden> wrote:
>> Regarding gl2ps, I've made some assumptions.
>> (1) That the GL2PS_PS/EPS/PDF/TEX/SVG/PGF formats will each be supported.
> As usual, its not so simple. The when outputting in PDF format, GL2PS
> does not respect text alignment, so all labels are placed wrong. If
> someone has a little PDF knowledge, this could probably be fixed in
> GL2PS.
> 
>> (2) That the output will respect the figure's papersize, paperorientation,
>> and paperposition property.
> This is also not supported by GL2PS, but I will try to implement it,
> at least for ps and maybe for eps
> 
> Shai
Shai / others,

I've played with direct PDF output (no ghostscript). The direct approach produces a superior image for surface plots.

If you try a plot of "surf (peaks)" with the current sources the surface plot has some small gaps visible (for me anyway) along the mesh. However, a larger problem is the time needed to open and display the resulting pdf file (several seconds).
This could be partly due to your older ghostscript.  I noticed a substantial reduction in
PDF size as of some recent ghostscript update.  But, I think that ghostscript is still not
doing a very efficient job.  For example, pdftex (goes directly from TeX to pdf) produces
a much smaller pdf then going through ghostscript.

In any case the direct to pdf route should be better if the text alignment can be fixed.
As an experiment I modified __fltk_print__.m to allow pdf to be created directly (patch is attached). The result is a file that loads instantly and has no visible defects in the surface plot.

However, as you (Shai) noted above the alignment of the text objects is wrong.

If direct PDF output by GL2PS place the text objects correctly, it think it possible to use PDF as the basis for all output (rather than EPS as is done now).

Another solution would be to determine why surface plots exported as EPS render so slowly, and fix that.

Unfortunately, I have little value as a c/c++ programmer.

Ben


Michael



reply via email to

[Prev in Thread] Current Thread [Next in Thread]